The 4^th International Rice Blast Conference

ABSTRACTS

Changsha, China

October 9-14, 2007

Oral Presentations

Current Status and Future Prospects of Researchon Blast Disease in Rice (Oryza sativa)

Gurdev S. Khush and K. K. Jena

University of California, Davis CA 95616-7008, USA

Email:gurdev@khush.org

Rice is the most important food security crop and staple of half the world population. Major increases in rice production occurred during last four decades of last century as a result of wide scale adoption of green revolution technology. Demand for rice continues to increase as a result of population increase and improvement in living standards particularly in Africa and Latin America. However, rate of increase of rice production has slowed down. It is estimated we will have to produce 30% more rice in 2030. For this purpose we need rice varieties with high yield potential and greater yield stability. Breakdown of blast resistance is the major cause of yield instability in several rice growing areas. Efforts are underway to develop rice varieties with durable blast resistance.

More than 40 major genes as well as QTL for resistance to blast have been identified. Monogenic resistance is less stable but varieties with pyramided monogenes or QTL are durably resistant. Rice research should focus on identifying more durably resistant genes, tagging of these genes with molecular markers and pyramiding these genes or QTL through molecular marker aided selection. Candidate gene identification through rice functional genomics has great potential for developing more durably resistant varieties.

Progress in Breeding of Super Hybrid Rice

Long-Ping Yuan

China National Hybrid Rice R&D Center, Changsha, Hunan, 410128 China

In order to meet food requirement in the 21^st century, a super rice breeding program was set up by China Ministry of Agriculture in 1996. The yield targets for single season crop of rice hybrids are as follows.

Phase Ⅰ (1996~2000): 10.5 t/ha.

Phase Ⅱ (2001~2005): 12 t/ha.

Several pioneer super hybrid rice varieties had been developed by 2000, which met the yield target of the Phase Ⅰ. More than 20 demonstration locations with 6.7 or 67ha each where the average yield was over 10.5 t/ha in 2000. The yield of these pioneer super rice hybrids in large-scale commercial production (1.2-2.0 million ha) is 8.5 t/ha in recent years.

Good progress had been made on developing Phase Ⅱ super rice hybrids. A promising two-line indica/japonica combination, P88s/0293, yielded more than 12 t/ha at five demonstration locations with around 7 ha each in 2003 and at twelve locations in 2004. This second generation super hybrid rice was released for commercial production in 2006. Based on the above achievements, the phase Ⅲ super hybrid rice breeding program is proposed, in which the yield target is 13.5 t/ha and can be fulfilled by 2010. The technological approaches for breeding super hybrid rice mainly are: ①Morphological improvement, ② Using inter-subspecific heterosis and ③ Utilization of biotechnology. The details of these technical issues are discussed in the full paper.

An “omics” Interrogation of Pathogenesis

by the Rice Blast Fungus

Ralph Dean

Center for Integrated Fungal Research, Dept. Plant Pathology, North Carolina State University, Raleigh, NC 27606, USA

Magnaporthe grisea is the causal agent of rice blast, the most devastating disease of rice worldwide and is a seminal model to elucidate the basis of pathogen–host interactions. Following the completion of the genome sequence of both the fungus and its host, rice, research is focused on functional and comparative approaches to uncover the molecular and evolutionary foundation of fungal pathogenesis. Whole genome microarray analysis of appressorium initiation, development and maturation induced by hydrophobic surfaces and by cAMP revealed a core set of 357 genes that were differentially expressed when compared to conidial germination under non-inductive conditions. In addition to genes involved in lipid, carbohydrate and secondary metabolism, numerous genes involved in protein turn over and amino acid catabolism were significantly induced. The critical requirement for protein catabolism, including endo-proteases and key enzymes involved in shuttling carbon back into the Kreb cycle, was demonstrated by gene knockout. To examine the evolution of pathogenesis, a semi-automated high throughput computational platform to facilitate large-scale comparative analyses was developed. Comparison of gene sets from 11 fully sequenced fungal pathogens and related non-pathogens revealed evidence for duplication of genes associated with several functional categories including signaling and hydrolytic activities. A database, FEGA (Fungal Evolutionary Genomic Analyses), was created and provides information about gene copy number in a genome, gene family constituency among genomes, and functional descriptions of gene families, including Gene Ontology (GO)-based functional annotation. Indices of selection (Ka/Ks ratios) and divergence (Ks, synonymous nucleotide substitution) are also included. Other efforts are currently focused on examination of novel non-coding transcripts, transcriptional networks and protein-protein interactions to define the circuitry regulating rice-rice blast interactions.

Genome-wide Identification of Genes Controlling Hyphal Growth of Magnaporthe Oryzae

You-Liang Peng

The State Key Laboratory for Agrobiotechnology and Department of Plant Pathology, China Agricultural University, Beijing 100094

Magnaporth oryzae infects host plants by a process involving conidiation, appressorium formation, penetration and invasive growth of infection hyphae. Understanding of molecular mechanisms regulating the infection process will contribute to design of novel approaches for the disease control. We suppose that many of genes required for in vitro colony growth are also necessary for invasive growth of infection hyphae in M. oryzae, and thus started genome-wide identification of genes that control colony growth. In order to identify genes that control colony growth of M. oryzae, an insertion mutagenesis library was generated with a field isolate P131. The library contained about 68000 independent transformants. Through screening the library, over 600 mutants were identified, showing slow colony growth. Genetic co-segregation analysis was carried out on about 400 of the mutants, which showed that the phenotype change in about 40% of the mutants was co-segregated with the selection marker of hygromycin resistance, suggesting that not all the mutations were caused by the insertion. TAIL PCR was used to isolate the sequences flanking the insertion sites. So far, the sequence in about 60 co-segregation mutants was determined, and through gene complementation and targeted gene disruption, more than ten genes were proven to control the colony growth. A detailed report will be given on the genetic network controlling colony growth of M. oryzae.

Systems Biology Initiatives for Magnaporthe oryzae

Yong-Hwan Lee

Department of Agricultural Biotechnology and Center for Fungal Genetic Resources, Seoul National University, Seoul 151-921, Korea

Magnaporthe grisea is a causal agent of rice blast and considered as a model pathogen for studying plant-microbe interactions. This is due to not only economic significance of rice blast disease worldwide but genetic and molecular tractability of this fungal pathogen. Currently whole genome sequence of strain 70-15 is available in the public database. To decipher fungal pathogenicity factors at genome-wide level in this fungus, we initiated a large-scale insertional mutagenesis using Agrobacterium tumefaciens-mediated transformation (ATMT). This project includes (1) construction of transformants library (2) development of high throughput phenotype screening and DNA extraction systems, and (3) rescuing flanking sequences of T-DNA insertion from selected transformants. We generated 21,070 transformants and screened for 7 phenotypes as a high throughput manner; conidiation, conidial morphology, conidial germination, appressorium formation, mycelial growth, pigmentation, and pathogenicity. Over 1,000 loss of virulence and development-defective mutants were obtained from ATMT mutant library. The T-DNA tagged sequences from the selected transformants are being rescued by TAIL-PCR and sequencing and 741 unique loci are identified thus far. To verify phenotype changes by T-DNA insertion, crossing with a wild type and targeted gene knock-out are being applied. Distribution and integration patterns of T-DNA on fungal chromosomes are also analyzed. Furthermore, we developed the ATMT data management system to handle all phenomics and genomics data of these transformants.

Characterization of T-DNA Insertion Patterns and Identification of Pathogenicity-deficient Mutants

in Rice Blast Fungus Magnaporthe Oryzae

Guihua Li, Zhuangzhi Zhou, Chunhua Lin and Chaozu He*

State Key Laboratory of Plant Genomics, Institute of Microbiology, Chinese Academy of Sciences, Beijing 100101

* Corresponding author: E-mail: hecz@im.ac.cn

Agrobacterium tumefaciens-mediated transformation (ATMT) has been proven to be a powerful strategy for gene disruption in plants and fungi. Using ATMT, a T-DNA tagged population consisting of 6179 transformants of Magnaporthe oryzae was constructed. With thermal asymmetric interlaced-PCR (TAIL-PCR), 623 right border (RB) flanking sequences and 124 left border (LB) flanking sequences were generated. Analysis of these flanking sequences indicated a significant integration bias toward non-coding sequences, suggesting distribution of T-DNAs was not random. Comparing to T-DNA RB, LB was nicked inaccurately and truncated frequently during integration. Chromosomal rearrangements, such as deletion, inversion and translocation, were associated with T-DNA integration in some transformants. Our data suggest that, comparing with plant cells, T-DNA integrates into this filamentous fungus with more precise and simpler patterns. After innoculation of these transformants against a compatible rice cultivar Nipponbare, some pathogenicity-deficient mutants were identified. Of them, one mutant was identified that the insertion of T-DNA dramatically reduced the expression of MoUROD, a gene encoding uroporphyrinogen decarboxylase (UROD). UROD is an enzyme of heme biosynthesis. Mutation of MoUROD attenuated virulence of M. oryzae to compatible rice but not affected in induction of hypersensitive response in incompatible rice. Our results suggest that MoUROD plays an important role in tolerance of oxidative stress during expansion of invasive hyphae in rice cell.

Characterization on Transcriptional Circuitry Regulating Pathogenicity Using a Proteomics Approach

Thomas Mitchell^1,, Jin-Rong Xu², Cornelia Koten², Heng Zhu³, Heesool Rho³, Yeon Yee Oh⁴, Malali Gowda⁴, and Ralph Dean⁴

1 The Ohio State University, Department of Plant Pathology, Columbus OH

2 Purdue University, Department of Botany and Plant Pathology, West Lafayette, IN

3 Johns Hopkins University, School of Medicine, Baltimore, MD

4 North Carolina State University, Center for Integrated Fungal Research, Raleigh, NC

A key component to deciphering the molecular underpinnings of pathogenicity for Magnaporthe grisea is the modeling of signaling networks the fungus uses to coordinate gene expression. These signaling networks continue to be fertile ground for generating insights into fungal biology and virulence, however full characterization of the downstream transcription factors they directly regulate is lacking. Through a combined automated and manual annotation process, we have identified over 500 putative transcription factors in the M. grisea genome and confirmed the expression of > 90% using expression data from EST, RL-SAGE and MPSS studies. We cloned >80% of these transcription factors and expressed each in yeast in order to print a M. grisea transcription factor protein microarray. Protein arrays were used to assay kinase phosphorylation specificity and activity. Gene disruption and over-expression strategies are now being used to map transcription factor binding motifs and identify genes regulated by selected transcription factors using a ChIP-chip hybridization strategy. We will present results from annotation, cloning, and phosphorylation studies as well as describe ChIP-chip studies.

The MGOS (M. grisea O. sativa) Interaction Community Database

Carol Soderlund

University of Arizona Bio5 Institute, 1657 E. Helen Street, University of Arizona, Tucson AZ 85721 USA

E-mail: cari@agcol.arizona.edu

The MGOS (Magnaporthe grisea Oryza sativa, www.mgosdb.org) database was

developed to contain genomic, gene expression and mutant data fromexperiments on the interaction between Oryza sativa and Magnaporthe grisea(M. oryzae) ^(1,2). The experiments were planned by a consortium of fungal and rice geneticists, to dissect early stages of the interaction betweenhost and pathogen. The data include ESTs and RL-SAGE information from infection in both resistant and susceptible interactions; phenotypic analysis of >50,000 M. grisea mutants; and results from a dual O. sativa-M. grisea microarray. MGOS also contains the genomic sequence from both rice and M. grisea along with automated annotation. A recent grant was awarded ⁽³⁾ to make MGOS a community database, which includes community annotation, enhanced microarray submission capabilities, and literature submissions. The community annotation functions have been implemented, and work on microarray submission is near completion. The annotation information includes the gene name and description, transcript structure and location, gene ontology terms, fungal anatomy terms, publication information, and the name of the person making the annotation submission.We request the communities' active participation to make this a model community database. We want the MGOS database to greatly enhance Magnaporthe research for everyone in the community.Towards this end, your feedback and requests will be greatly appreciated.

1. NSF-PGRP #0115642, PIs: R.Dean, D.Ebbole, M.Farman, M.Orbach, C.Soderlund, G.Wang, R.Wing, J.Xu.

2. C.Soderlund, K.Haller, V.Pampanwar, D.Ebbole, M.Farman, M.Orbach,G.Wang, R.Wing, J.Xu, D.Brown, T.Mitchell, R.Dean (2006) MGOS: a resource for studying Magnaporthe grisea and Oryza Sativa interactions.Mol Plant Microbe Interact 19: 1055-1061.

3. NSF-MGS #0627159, PIs: C.Soderlund, M.Orbach, B.Valent.

Investigating the Biology of Appressorium-mediated Plant Infection by the Rice Blast Fungus Magnaporthe grisea

Nicholas J Talbot*, Richard A. Wilson, Michael J. Kershaw, Darren M. Soanes, Diane O. C. Saunders, Elise Lambeth, Thomas A. Richards, Martin J. Egan, Han-Min Wong, Zaira Caracuel-Rios, Romain Huguet, Ana-Lilia Martinez-Rocha

School of Biosciences, University of Exeter, Geoffrey Pope Building, Stocker Road, Exeter, EX4 4QD, United Kingdom

*Corresponding author:E-mail: N.J.Talbot@exeter.ac.uk

During plant infection, the rice blast fungus elaborates a specialised infection structure known as an appressorium. This unicellular, dome-shaped structure generates turgor that is translated into mechanical force to allow rupture of the rice cuticle and entry into plant tissue. We set out to explore whether the development of a functional appressorium was linked to the control of cell division. This was based on the observation that following germination of a conidium on the rice leaf surface, a single round of mitosis always occurs during germ tube elongation, prior to the formation of an appressorium. We found that blocking completion of mitosis by generation of a temperature-sensitive MgnimA mutant prevented appressorium morphogenesis. Furthermore, we found that following mitosis, conidia always undergo cell collapse and cell death, which appears to be a programmed, autophagic process. Deletion of MgATG8 prevented autophagy in M. grisea and rendered the fungus non-pathogenic. Taken together, our results indicate that appressorium morphogenesis requires genetic control by completion of mitosis and autophagic cell death of the conidium. We have also recently demonstrated the appressorium morphogenesis is accompanied by a burst of reactive oxygen species. Deletion of NOX1 or NOX2 which encode NADPH oxidases is sufficient to prevent plant infection by interfering with appressorium function. A third NOX-encoding gene NOX3 plays a role in hyphal morphogenesis. Once the M. grisea appressorium has formed, cellular turgor is generated by accumulation of osmotically-compatible solutes, notably glycerol. We have used genetic, biochemical, proteomic and, most recently, metabolomic analysis to investigate how turgor is generated and to define the key genetic determinants of appressorium function. Of particular interest is the central role of trehalose metabolim to the genetic control of fungal virulence and the interplay between sugar signaling and nitrogen source utilization and the role of peroxisomal fatty acid beta-oxidation in appressorium physiology. The appressorium brings about plant infection by elaborating a penetration hypha that differentiates further into invasive hyphae which grow rapidly within the host plant cells. One of the key challenges in understanding rice blast disease is to determine how fungal proteins are delivered to the host during plant infection and to define the mechanisms by which the fungus proliferates biotrophically within the rice leaf. Progress towards determining the secretory processes necessary for M. grisea plant tissue colonization will also be presented.

Fungal Secondary Metabolism is an Essential Component of the Complex Interplay between Rice and Magnaporthe

Marc Henri Lebrun, Jerome Collemare, Isabelle Fudal, Heidi Bohnert

UMR5240 CNRS/UCB/INSA/BCS, Functional Genomics of Plant Pathogenic Fungi, Bayer Cropscience, 14-20 rue P Baizet, 69263 Lyon Cedex 09, France

Functional analyses of fungal genomes are expanding our view of the metabolic pathways involved in the production of secondary metabolites. These genomes contains a significant number of genes encoding biosynthetic enzymes such as PKS and hybrid PKS-NRPS involved in the production of polyketides, NRPS involved in the production of peptides and TS involved in the production of terpenes. Magnaporthe grisea has a high number of such key enzymes (22 PKS, 8 NRPS, 10 PKS-NRPS, 5 TS), suggesting that this fungal species produce a large number of diverse secondary metabolites. In particular, it has the highest number of PKS-NRPS in fungi. Among them, 4 are expressed during infection (ACE1, SYN2, SYN6, SYN8). Targeted gene replacement of SYN6, expressed in mycelium, spores and infected leaves, does not impair infection. ACE1, SYN2 and SYN8 share the same pattern specific of the early stages of infection (penetration). Ace1 was shown to be expressed only in appressoria during the penetration of the fungus into host plant suggesting that the corresponding metabolite is delivered to the first colonized cells. Targeted gene replacement of ACE1 and SYN2 does not impair infection of susceptible rice cultivars suggesting a possible functional redundancy between these pathways. However, only ACE1 null mutants are able to infect resistant rice cultivars carrying Pi33 blast resistance gene and recognition of M. grisea by Pi33 resistant rice cultivars was shown to require ACE1 that behaves as a classical avirulence gene. However, fungal AVR genes are known to encode small peptides secreted into host tissues during infection. ACE1 from M. grisea differs from these fungal AVR genes as it likely controls the production of a secondary metabolite recognized by resistant rice cultivars carrying Pi33. Arguments toward this hypothesis involves the fact that the protein Ace1 is only detected in the cytoplasm of appressoria and not in infectious hyphae differentiated inside infected epidermal cells. Furthermore, Ace1-ks0, a non-functional ACE1 allele obtained by site-directed mutagenesis of an amino acid from polyketide synthase KS domain essential for its enzymatic activity, is unable to confer avirulence. According to this hypothesis, resistant plants would have evolved mechanisms to recognize microbial pathogens through the perception of the secondary metabolites they produced during infection. In order to characterize the metabolite produce by ACE1, this enzyme is currently constitutively expressed in M. grisea and under the control of an inducible promoter in Fusarium venenatum.

Regulation of Infectious Growth by the TBL1 Complex in Magnaporthe grisea

Shengli Ding and Jin-Rong Xu

Dept. of Botany and Plant Pathology, Purdue University, West Lafayette, IN47907

Magnaporthe grisea is pathogenic to economically important crops such as rice and wheat. Although appressorium formation and penetration are well characterized, there are only limited studies on infectious hyphal growth in M. grisea. We have generated gene replacement mutants for the transducin beta-like gene TBL1, which was first identified as a novel fungal pathogenicity factor in Fusarium graminearum. Similar to the F. graminearum mutant, the M. grisea tbl1 deletion mutant was defective in conidiogenesis and non-pathogenic. Appressoria formed by the tbl1 mutant were able to penetrate but failed to form secondary infectious hyphae. TBL1 is homologous to the Tbl1 nuclear receptor corepressor. In transformants expressing a TBL1-GFP fusion construct, nuclear localization of the GFP fusion protein was observed. Deletion analysis revealed that the N-terminal LisH domain that is normally involved in protein-protein interactions is essential for Tbl1 function. Homologs of TBL1 in yeast and mammalian cells are part of a conserved HDAC transcription co-repressor complex. Several components of the yeast Set3C complex, including the Set3 and Snt3 homologs, also are highly conserved in M. grisea. Mutants deleted of the MgSET3 gene had the same defects in conidiation and plant infection as the tbl1 mutant. In transformants expressing a TBL1-3Flag construct, we were able to pull down MgSnt3. These data indicate that TBL1 is a component of this well-conserved regulatory complex, which has been evolved to regulate the development and growth of infectious hyphae in M. grisea.

Surface Sensing and Signaling during Initiation

of Rice-blast Disease

Naweed Naqvi*, Ravikrishna Ramanujam, Angayarkanni Suresh, Liu Hao

Fungal Patho-Biology Group, Temasek Life Sciences Laboratory, 1 Research Link, National University of Singapore, Singapore 117604

*Corresponding author:E-mail: naweed@TLL.org.sg

Conidial germ tubes of rice-blast fungus Magnaporthe grisea must differentiate into an infection structure called the appressorium in order to penetrate its host. Apart from hydrophobicity, the other host-surface characteristics responsible for appressorium initiation are poorly understood. In a forward genetics approach, we screened for Magnaporthe mutants defective in early surface signaling events during infection. This lead us to the identification of TMT1390 mutant (disruption in the Regulator of G-protein Signaling/RGS1 locus) which was capable of forming appressoria on non-inductive surfaces both hydrophobic and hydrophilic but couldn’t form appressoria on soft surfaces. Further characterisation, molecular identification and analysis of cells lacking RGS1 function helped us identify and define a thigmotropic response as being essential for initiation of pathogenesis. Involvement of a G-protein signaling network was identified as a downstream effector module of such early surface-hardness dependent signaling. Chemical genetic studies and global transcriptome analyses related to surface hardness indicated that such thigmo-morphogenesis is initiated within two hours after conidia germination and uses calcium signaling mediated by ion channels. Biophysical analyses allowed us to estimate the critical hardness necessary for efficient initiation of infection in Magnaporthe. Our preliminary results suggest a possible role for stretch-activated ion channels and a non-canonical GPCR in hardness sensing and host infection. These will be discussed along with a function for G proteins in elaborating the extra-cellular matrix during the pathogenic development in Magnaporthe.

Organ-specificity and Pathogenesis in the Rice Blast

Fungus Magnaporthe oryzae

Sara Tucker¹, Maria Besi¹, Stephan Goetz¹, Yueh-Mei Zhou¹, Anne Osbourn² and Ane Sesma^1,*

1 Dept. of Disease and Stress Biology, John Innes Centre NR4 7UH Norwich, UK；

2 Dept. of Metabolic Biology, John Innes Centre NR4 7UH Norwich, UK

*Corresponding author: E-mail: ane.sesma@bbsrc.ac.uk

Compared to foliar fungal pathogens, very little is known about the requirements of soil-borne fungal pathogens for successful colonisation of root tissues, despite the fact that root-infecting fungi are extremely important as disease-causing agents. Recently, fungi regarded as foliar pathogens have been found to infect plant roots under field conditions, e.g. Leptosphaeria maculans and Cercospora beticola. The rice blast fungus Magnaporthe oryzae is regarded as a foliar pathogen. It belongs to the Magnaportaceae family, which contains root-infecting pathogens such as Magnaporthe poae, the turfgrass pathogen, and Gaeumannomyces graminis, the causal agent of “take-all” disease of cereals. We have previously shown that M. oryzae can also infect roots under laboratory conditions. Instead of the appressoria formed during leaf infection, simple hyphopodia structures allow M. oryzae to penetrate roots. Now we have found that M. oryzae can penetrate through the root hairs and natural openings present on the root surface. Preliminary investigations using characterised M. oryzae mutants from other laboratories have allowed us to classify mutants with leaf-specific, root-specific and general defects in the ability to cause lesion formation on leaves and/or roots. We have undertaken plant infection tests of ~1,000 T-DNA insertional transformants generated in our lab. This has enabled us to identify several M. oryzae mutants deficient in the ability to infect leaves and/or roots. Among them, we have started to functionally characterise two mutants. One of the mutants is tagged in a gene encoding for a putative RNA-binding protein which may have a role in the maturation and transport of mRNA precursors that are required for pathogenesis. The second mutant has undergone insertional inactivation in a gene encoding a predicted protein with DNA-binding functions. Both of them are defective in infection of rice roots.

At present, there is a lack of understanding of root diseases in part, due to the difficulty of studying such processes below ground, and also because of the genetic intractability of many root-infecting organisms. Therefore, understanding the root infection-related developmental process in M. oryzae will greatly enhance our understanding of fungal pathogenesis.

Telomere Instability in Magnaporthe oryzae Caused by Highly Active Telomere-targeted Retrotransposons

John H. Starnes, Cathryn J. Rehmeyer, Shouan Zhang, David Thornbury and Mark L. Farman*

Department of Plant Pathology, University of Kentucky, Lexington, KY 40546, USA

*Corresponding author:E-mail: farman@email.uky.edu

Magnaporthe oryzae isolates that infect perennial ryegrass (prg) have unusually unstable telomeres that undergo continual rearrangements in culture and in planta. By comparison, telomeres in other host-specific forms of this fungus are quite stable. Sequencing revealed that the chromosome ends of the ryegrass pathogens are organized very differently to those in a strain with stable telomeres. Specifically, the subtelomeres consist of tandem arrays of three types of repetitive elements, with canonical telomere repeats at the ends of the arrays. The elements are found only in subtelomeric locations, so we have named them M. oryzae telomere-exclusive repeats (MoTERs). MoTER1 is 4.6 kb in length and codes for a reverse transcriptase. It lacks terminal repeats and, therefore, appears to be a non-LTR retrotransposon. MoTER2 is only 1.7 kb long and potentially codes for a 204 aa protein of unknown function, while MoTER3 (4.95 kb) is a recombinant of MoTER1 and MoTER2. We have identified new insertions of MoTER1, thereby demonstrating that it is an active transposon. We are currently trying to identify new insertions of MoTER2. The structures of MoTER1 and MoTER2, and the way they are organized at the chromosome ends, are reminiscent of the retrotransposons TART and HeT-A, which are responsible for maintaining the Drosophila telomeres. Therefore, we hypothesize that M. oryzae strains from ryegrass have dual systems for telomere maintenance - one based on telomerase, and another which uses terminally-targeted retrotransposons to repair degraded ends that arise during telomere crisis. Telomere instability was observed in all of the progeny from genetic crosses between a prg pathogen and an isolate with stable telomeres, although it was largely restricted to telomeres that were inherited from the prg parent. This suggests that the presence of the MoTER elements causes the telomeres to be unstable in the first place. Currently, we are testing effects of MoTER activity on the expression of neighboring genes. Results from these studies will be discussed.

Studying the Role of Heat Shock Proteins in Pathogenicity

of Magnaporthe oryzae

Nicole M. Donofrio

152 Townsend Hall, 531 S. College Ave, Newark, DE, 19716, USA

E-mail: ndonof@udel.edu

Heat shock proteins (HSPs), or chaperones, range in cellular function from folding of newly synthesized proteins, to repair of misfolded proteins during stress. Currently, there is a paucity of data describing whether HSPs are involved in fungal virulence on plants. We hypothesize that plant pathogens must be able to cope with stressful conditions upon gaining ingress to plant hosts, and one such mechanism may employ stress-related genes, including heat shock. Our main goal is to more fully characterize and understand the suite of HSPs in the rice blast pathogen, Magnaporthe oryzae, and ultimately determine their role in pathogenicity. Towards achieving this goal, we first performed inhibition experiments using the antibiotic, geldanamycin. This chemical is known to bind the active site of the major chaperone, HSP90, rendering it function-less. We wished to determine whether GDA had effects on the disease cycle of this pathogen, including germination, appressorial development, hyphal growth, and lesion production. Our results revealed that while GDA, at varying concentrations, appeared to have no effect on germination and appressorial development, this chemical did inhibit both hyphal growth in vitro as well as lesion development in planta by up to 75% at the highest concentrations used. Controls determined that lesion inhibition was most likely a result of GDA’s effect on the fungus, rather than the plant reacting to presence of the chemical. Second, we wished to determine whether GDA had any effect on the expression of putative heat shock genes, which RT-PCR tests revealed to be the case for several genes. We then performed a microarray experiment on GDA-treated versus untreated fungal hyphae, and results will be presented. Third, we wished to genetically determine whether several chaperones had any impact on pathogenicity. To this end, we generated knock-out mutants in two genes known to be molecular co-chaperones for HSP90 in other organisms, including yeast. Results of these knock-out experiments will also be presented.

Effector Function and Secretion during Biotrophic

Invasion by the Rice Blast Fungus

Barbara Valent¹, Gloria Mosquera¹, Chang-Hyun Khang^1,2, Romain Berruyer^1,4, Prasanna Kankanala¹, Martha Giraldo¹, Kirk Czymmek³, Sook-Young Park and Seogchan Kang²

1 Department of Plant Pathology, Kansas State University, Manhattan, KS 66503, USA

2 Pennsylvania State University, University Park, PA, USA

3 University of Delaware, Newark, DE, USA

4 Current Address: Université d'Angers, 49045 Angers, France

Magnaporthe oryzae is a hemibiotropic fungus that sequentially invades living plant cells using intracellular invasive hyphae (IH) that grow from cell to cell. Using live-cell imaging of a highly compatible interaction, we reported that IH are tightly wrapped in plant-derived extra-invasive hyphal membrane (EIHM). IH appear to seek out plasmodesmata for moving into the next cell, and filamentous IH that first grow in these cells have distinctive membrane caps at their tips (Kankanala et al, The Plant Cell, February 2007). To begin to identify and characterize effector proteins secreted by IH into live plant cells to control host cellular processes, we developed a reproducible procedure for obtaining infected sheath tissue with 20% IH RNA at 36 hpi, when most IH were still growing in the first-invaded rice cell. These samples were used for hybridization of both the whole genome M. oryzae microarray and a rice microarray to identify the interaction transcriptome. Fungal genes that were induced >50-fold in IH (compared to mycelium in culture) were enriched in putative secreted proteins with unknown functions, and rice genes that were induced >50-fold in infected tissue (compared to mock-inoculated tissue) were enriched in MAP kinase kinase kinase genes and transcription factors. These results are consistent with the hypothesis that IH secrete novel effector proteins into rice cells to reprogram their gene expression. Our initial gene replacement experiments have not shown major phenotypes associated with putative effectors. To study effector secretion in planta, we fused known and putative blast effectors to green fluorescent protein. Predicted effector signal peptides directed secretion of GFP from the fungus, and fusion proteins accumulated at predictable locations inside the EIHM. Fluorescence concentrated in the EIHM caps and in previously unrecognized structures, Blast Interfacial Complexes (BICs). Fusion proteins accumulated in BICs as long as IH grew in the cell. Correlative light and electron microscopy suggest that BICs are complex membrane-rich and vesicle-rich structures between the fungal wall and EIHM. We will discuss a potential role for BICs in blast effector secretion.

Toward the Understanding of Magnaporthe-rice Interactions: a Multi-faceted Genomics Approach

Ryohei Terauchi^1,*, Joe Win², Sophien Kamoun², Hideo Matsumura¹, Saitoh¹, Hiroyuki Kanzaki¹, Kentaro Yoshida¹, Matt Shenton¹, Thomas Berberich¹, Shizuko Fujisawa¹, Akiko Ito¹, Yoshitaka Takano³, Yukio Tosa⁴

1 Iwate Biotechnology Research Center, Kitakami, Japan

2 Department of Plant Pathology, Ohio State University, OARDC, Wooster Ohio, USA

3 Laboratory of Plant Pathology, Kyoto University, Kyoto, Japan

4 Laboratory of Plant Pathology, Kobe University, Kobe, Japan

*Corresponding author: E-mail: terauchi@ibrc.or.jp

Since whole genome sequences are available for both Magnaporthe and rice, this pathogen-host system provides a unique opportunity to address microbe-plant interactions from genomics perspectives. To select candidate genes directly involved in the interactions, we are employing (1) bioinformatics approach through identification of 1,884 Magnaporthe putative secreted protein genes, (2) simultaneous gene expression analysis of Magnaporthe and rice by SuperSAGE with comparison of expression profiles between compatible and incompatible interactions, (3) population genetics analysis of the polymorphism of Magnaporthe secreted protein genes by EcoTILLING. Screens (1) – (2) allowed us to select 452 Magnaporthe effector candidate genes and 150 rice genes presumably involved in resistance. Screen (3) identified a gene under strong diversifying selection, possibly involved in direct pathogen-host interaction. Function of these genes is tested by Agrobacterium-mediated transient overexpression in Nicotiana benthamiana. Through this assay, we have so far identified several Magnaporthe cell death inducing proteins including Nep1-like protein, and several rice cell death inducing proteins encoding a novel transcription factor. These are further functionally tested in Magnaporthe and rice by gene knockout and overexpression.

Expression of Secreted Proteins from Magnaporthe grisea and Analysis of Elicitor Activity toward Rice

Hanno Wolf¹, Guodong Lu², Kiran Bhattarai¹, Cristina Filippi¹, Yue Shang¹, Dan Li¹, Daniel J. Ebbole^1,*

1 Department of Plant Pathology & Microbiology, Texas A & M University, College Station, USA 77843

2 College of Plant Protection, Fujian Agricultural and Forestry University, Fuzhou, China 350002

* Corresponding author:E-mail: d-ebbole@tamu.edu

Extracellular proteins of fungal pathogens are candidate effector molecules. We have cloned a large number (~300) of putative secreted proteins from Magnaporthe oryzae. The proteins were fused to a His(x6) tag to allow for affinity purification of the proteins. The genes were transformed into M. oryzae to test for expression. Approximately one-third of the genes could be expressed and secreted to sufficient levels to allow detection by western blot analysis of culture filtrates. In almost all cases, constitutive expression of the proteins in the transformed strains did not markedly alter their interaction with rice. Proteins were purified from culture filtrates of M. oryzae and symptoms were observed on rice leaves in only very few cases. A directed approach to examination of candidate factors is being pursued with emphasis on unique gene families and genes implicated in virulence in other systems.

Large-scale Isolation and Functional Analysis of Putative Effectors from M. oryzae Using Integrated

Genomics Approaches

Songbiao Chen¹, Pattavipha Songkumarn¹, Malali Gowda¹, Venu Reddyvari Channarayappa¹, Chan Ho Park¹, Maria Bellizzi¹, Daniel Ebbole², Guo Liang Wang^1,*

1 Department of Plant Pathology, The Ohio State University, USA
2 Department of Plant Pathology and Microbiology, Texas A&M University, USA
* Corresponding author E-mail: wang.620@osu.edu

Rice blast disease, caused by the fungus Magnaporthe oryzae, is a leading constraint to rice production and is a serious threat to food security worldwide. To elucidate the function of effector proteins from M. oryzae in pathogenesis and interaction with the host, we have performed RL-SAGE and MPSS approaches to study the gene expression profiles of M. oryzae during the interaction. One RL-SAGE library from leaf tissues after 96-h compatible blast infection, 5 MPSS libraries from leaf tissues after 3-h, 6-h, 12-h, 24-h, 48-h incompatible blast infection, and 6 MPSS libraries from leaf tissues after 3-h, 6-h, 12-h, 24-h, 48-h, and 96-h compatible blast infection have been made and analyzed. The RL-SAGE and MPSS analyses identified 3,441 and 3,004 annotated M. oryzae genes, respectively. Among them, 217 in-planta expressed putative secreted protein genes of M. oryzae were identified. We have then applied integrated functional genomic approaches to analyze the function of the identified putative effectors from M. oryzae. These approaches include a high throughput vector system for cloning and expression of M. oryzae genes in rice cells, a highly efficient rice protoplast transient expression system for cell death and defense response assays after expression of the fungal genes and a yeast secretion trap system for confirming the secretion function of signal peptide of M. oryzae secreted proteins. Through these integrated approaches, ~30 in-planta expressed effector proteins have been isolated and characterized. Among them, several cell-death inducing proteins in rice protoplasts have been identified. Further characterization of these genes and identification of their host targets will provide new insights into the interaction between rice and M. oryzae.

Acknowledgements: This project is supported by the NSF-Plant Genome Research Program (#0605017).

The MGOS (M. grisea O. sativa) Interaction Community Database Workshop

K.A. Greer and C. Soderlund

Bio5 Institute, 1657 E. Helen Street, University of Arizona, Tucson AZ 85721 USA

The MGOS (Magnaporthe grisea Oryza sativa, www.mgosdb.org) database was originally developed to contain genomic, gene expression and mutant data fromexperiments on the interaction between Oryza sativa and Magnaporthe grisea (M. oryzae) (1,2). A grant was awarded in December 2006 (3) to extend MGOS for community annotation and microarray submission, thereby, making it acommunity database. 

This workshop will provide a detailed demonstration of the functionality added to the MGOS database as part of NSF grant #0627159 (3), and will cover:

1. Adding a community annotation for a gene, including the gene name, symbol, and description of function, synonyms, transcript information (orientation, start, stop, exon boundaries, coding sequence start and stop), citations, and GO and fungal anatomy annotations.

2. Adding a citation for a journal article.

3. Entering, and viewing the results of, a microarray experiment.

4. Using BLAST and BLAT to align a nucleotide or amino acid sequence to the genomic or transcript sequences in MGOS.

5. Using the forum to communicate with the Magnaporthe grisea community.

An essential part of this workshop will be obtaining input from participants regarding improvements that they would like made to the MGOS database. The current grant will last through 2008, providing an incredible opportunity for the Magnaporthe grisea community to develop a world-class bioinformatics tool to aid in scientific research.

1. NSF-PGRP #0115642, PIs: R.Dean, D.Ebbole, M.Farman, M.Orbach, C.Soderlund, G.Wang, R.Wing, J.Xu.

2. C.Soderlund, K.Haller, V.Pampanwar, D.Ebbole, M.Farman, M.Orbach, G.Wang, R.Wing, J.Xu, D.Brown, T.Mitchell, R.Dean (2006) MGOS: a resource for studying Magnaporthe grisea and Oryza Sativa interactions. Mol Plant Microbe Interact 19: 1055-1061.

3. NSF-MGS #0627159, PIs: C.Soderlund, M.Orbach, B.Valent.

Rac GTPase-mediated Blast Resistance in Rice

Ko Shimamoto^1,*, Letian Chen¹, Nguyen Phuong Thao¹, Ayako Nakashima¹,Yoji Kawano¹, Keiko Imai¹, Hann Ling Wong¹, Kenji Umemura², Tsutomu Kawasaki¹

1 Laboratory of Plant Molecular Genetics, Nara Institute of Science and Technology (NAIST), 8916-5 Takayama, Ikoma 630-0101, Japan

2 Meiji Seika Ltd., 5-3-1 Chiyoda, Saitama 350-0289, Japan

* Corresponding author :E-mail: simamoto@bs.naist.jp

We have been studying the role of Rac GTPase in innate immunity of rice and found that it is a key molecular switch for defense response in rice. To identify components of the Rac GTPase-mediated innate immunity in rice we have taken a number of approaches including proteomics and reverse genetics and found components associated with OsRac1. They include known components such as RAR1, SGT1, and Hsp90 and novel components such as RWD and Sti1/Hop and they interact with OsRac1. Various studies on protein-protein interactions among these components indicate that at least 10 proteins form a network in the Rac GTPase-mediated innate immunity. Since most of those components are involved in both PAMP-mediated and R protein-mediated resistance in rice we propose a model in which early signaling events involved in these two types of resistance occur in essentially the same protein complex we call “defensome” at the plasma membrane. Details of individual components in the protein complex will be discussed.

Insights into the Rice Defense Response

Pamela Ronald

Department of Plant Pathology University of California, Davis, 95616 USA

The plant innate immune response includes perception of pathogen molecules as well as perception of signals released by the plants in response to pathogens. The molecular events of and connections between these processes are as yet ill defined, especially in the agriculturally paramount monocot clade. Our work focuses on three participants in rice (Oryza sativa) innate immunity, the pathogen recognition receptor, Xa21, and the proteins NH1 (NPR1 homolog1) and NRR (negative regulator of resistance) that respond to elevated levels of SA produced by the plant in response to pathogen attack. Xa21 allows the plant to perceive and defend against most strains of Xanthomonas oryzae pv. oryzae (Xoo), the causative agent of bacterial blight. NH1 and NRR play key roles in perception of salicylic acid, a signal that is elevated during the plant defense response.

Using yeast two-hybrid screening, we have identified 51 putative components of the rice innate immunity protein-protein interaction network that includes the Xa21 intracellular domains, NH1, and NRR. In addition, we have gathered microarray data comparing Xa21, NH1 over-expressing (ox), and NRR-ox plants before and after inoculation. We are using these data to support and prioritize further analyses of the members of the interaction map. We have found that another pathogen recognition receptor that confers resistance to the fungal pathogen Magnaportha grisea, Pi-d2, also interacts with at least three of the members of this network including Xa21 binding protein 15 (Xb15), a protein phosphatase 2C (PP2C). Functionaly analysis of Xb15 indicate that Xb15 is a PP2C phosphatase and may act as a negative regulator in XA21-mediated resistance.

Towards Understanding of Signal Perception and Transduction in Rice Blast Resistance

Yinong Yang

Department of Plant Pathology and Huck Institutes of Life Sciences, 405C Life Sciences Bldg., Pennsylvania State University, University Park, PA 16802, USA

E-mail: yuy3@psu.edu

A combination of molecular, biochemical, genomic and proteomic approaches have been taken to understand the early signal perception during the rice-Magnaporthe grisea interaction as well as the downstream signaling pathways leading to blast resistance and susceptibility. Using transgenic rice lines defective in salicylic acid (SA), jasmonic acid (JA), ethylene (ET) or abscisic acid (ABA) signaling, we have shown that (i) SA is not an effective signal molecule in rice but acts as a constitutive antioxidant to protect rice plants from the pathogen-induced oxidative damage; (ii) JA signal pathway is important for mediating rice defense gene expression and blast resistance; (iii) ET signal pathway appears to be very critical for rice blast resistance; and (iv) ABA interacts antagonistically with ET signaling in rice and significantly increases blast susceptibility. We also identified 17 members of rice mitogen-activated protein kinase gene (OsMPK) family and found that a half of them were induced by M. grisea infection. Transgenic analysis demonstrated that pathogen-induced OsMPKs were capable of mediating JA, ET and/or ABA pathway interactions and modulating rice blast resistance. Our preliminary data revealed that pathogen-responsive OsMPKs may interact with upstream NBS-LRR proteins and phosphorylate key components of downstream defense pathways. It is hypothesized that NBS-LRR proteins may detect the interaction between M. grisea effectors and host cell targets, and relay the early signal through the OsMPK cascades, leading to the activation of defense pathways and subsequent blast resistance or susceptibility.

Functional Dissection of a Rice Mitogen-activated Protein Kinase Kinase Kinase OsACDR1

Jung A Kim and Nam-Sooc Jwa*

Department of Molecular biology, Sejong University, Seoul 143-747, Korea

* Corresponding author: E-mail: nsjwa@sejong.ac.kr

The OsACDR1 (OsEDR1) gene of rice has been previously characterized to encode a Raf-like mitogen-activated protein kinase kinase kinase (MAPKKK), showed multistress responsive regulations. The OsACDR1 protein, as a MAPKKK, displayed autophosphorylation and kinase activity. OsACDR1 positively regulated HR-like leaf spots and its overexpression induced typical lesion mimic phenotype. Defense-related molecules including phenolic compounds and phytoalexins were accumulated around individual lesions. Pathogens-related (PR) genes were also up-regulated associated with lesion development. Lesion development in OsACDR1-OX transgenic lines increases resistance to both fungal (M. grisea) and bacterial (Xanthomonas oryzae pv. oryzae) pathogens through activation of basal resistance. Over-expression of OsACDR1 enabled transgenic lines to inhibit appressoria penetration of M. grisea on the leaf surface. Deletion mutant of OsACDR1 showed susceptible phenotype to rice blast fungus. Those results suggest that OsACDR1 is the first identified rice MAPKKK, positively regulating lesion development as well as basal resistance.

Functional Analysis of Rice WRKY89 in Rice Defense Response and UV-B Stress

Zejian Guo, Haihua Wang, Junjie Hao, Xujun Chen

Department of Plant Pathology, China Agricultural University, Beijing 100094, China

WRKY proteins are a large family of transcriptional regulators involved in a variety of biological processes in plants. In this study, we characterized a rice WRKY gene, OsWRKY89. RNA gel blot analysis indicated that OsWRKY89 was induced strongly by treatments of methyl jasmonate and UV-B radiation. Transient expression analysis using an OsWRKY89-eGFP fusion gene in onion epidermal cells revealed that the OsWRKY89 protein was targeted to nuclei. Protein fusion analysis of OsWRKY89 and its mutants with a GAL4 DNA binding domain indicated that the 67 C-terminal amino acids were required for transcriptional activation and that the leucine zipper region at the N-terminus enhanced transcriptional activity. Overexpression of OsWRKY89 led to retarded growth at the early stage and reduced internode length. Scanning electron microscopy revealed an increase in wax deposition on leaf surfaces of the overexpressing lines and a decrease in wax loading in the RNAi lines. Moreover, extractable and cell-wall-bound phenolics were decreased in the overexpressor lines, whereas the SA levels increased. Staining experiments demonstrated an increase in lignification in culms. Interestingly, overexpression of the OsWRKY89 gene enhanced resistance to rice blast fungi and white-backed plant hoppers and tolerance to UV-B tolerance. These results suggest that OsWRKY89 is involved in response to biotic and abiotic stresses.

Functional and Evolutionary Analysis of the Pi9 Resistance Gene Cluster in Rice

Bo Zhou^{1, 2}, Liangying Dai³, Xionglun Liu³, Xunbo Li, Jun Wu³, Yajun Hu³, Jinling Liu³, Shaohong Qu¹, Guifu Liu^1,, Bellizzi Maria^1,, Hajime Sakai⁴, Bin Han², and Guo-Liang Wang^1,3

1 Department of Plant Pathology, the Ohio State University, Columbus OH 43210 USA,

2 National Center for Gene Research, Chinese Academy of Sciences, Shanghai 200233, China

3 Rice Genomics Laboratory, Hunan Agricultural University, Changsha, Hunan 410128, China

4 DuPont Crop Genetics, Experimental Station, Wilmington, DE 19880, USA,

The complex Pi9 locus contains at least six resistance (R) alleles, i.e., Pi2, Pi9, Piz-t, Piz, Pigm(t) and Pi40(t), and each of them confers broad-spectrum resistance to diverse strains of Magnaporthe oryzae. To understand the molecular mechanism underlying the broad spectrum resistance mediated by these R genes, we successfully cloned the Pi2, Pi9, and Piz-t genes via map-based cloning and PCR homology cloning strategies. They all encode highly homologous proteins with a nucleotide binding site (NBS) and a leucine-rich repeat (LRR) domain and belong to a multiple NBS-LRR gene family in each donor line. The three genes share over 96% identity in amino acid sequence to each other. Sequence analysis indicated that the LRR region plays a critical role in the determination of their resistance specificities. Only eight amino acid changes distinguish the Pi2 from the Piz-t, which are exclusively confined within three consecutive LRRs. The Pi2/Pi9 chimeric gene, which is derived from the replacement of the Pi2’s LRR region with the Pi9’s, was found to confer the Pi9 resistance specificity, further suggesting that the LRR region is the major determinant of the resistance specificity. Comparative analysis of the Pi9 locus in five different rice cultivars revealed contrasting genomic dynamics at the intra- and inter-haplotype levels. To further elucidate the evolutionary mechanism of the Pi9 cluster in wild rice species, we sequenced five BAC clones spanning the Pi9 locus in AA, BB, CC and BBCC genomes. Preliminary sequence analysis indicated that the Pi9 locus within each wild rice genome have a similar genomic dynamics to that in the cultivated species, in which the NBS-LRR paralogues have the same phase and position of their first intron but show significant sequence variation with each other. In contrast, a clear orthologous relationship of the NBS-LRR genes was not observed among the five genomes. Furthermore, each genome comprises different copy numbers of NBS-LRR genes, of which the AA genome has the least number of the NBS-LRR genes. A sequence rearrangement among different NBS-LRR genes was found in the five genomes, suggesting that the Pi9 locus has undergone intergenic unequal recombination during its evolution. We also amplified 119 Pi9 homologous DNA fragments containing both NBS and LRR domains from 60 cultivated and wild rice lines. These comparative analyses of the Pi9 cluster will provide valuable insights into the genomic dynamics and evolutionary mechanism of the Pi9 NBS-LRR resistance gene complex in the Genera Oryzae.

Understanding the Coevolution of Rice Blast Resistance Hene Pi-ta and Magnaporthe oryzae

Avirulence Gene AVR-Pita

Yulin Jia

USDA-ARS Dale Bumpers National Rice Research Center Stuttgart, AR 72160

E-mail: yjia@spa.ars.usda.gov

Rice blast disease caused by the filamentous ascomycetous fungus Magnaporthe oryzae remains to be one of the most serious threats for food security globally. Using resistance (R) genes in integrated cultural practices has been the most powerful practice for rice crop protection. Genetic analysis suggests that resistance mediated by the R gene Pi-ta in rice is effective at preventing infection of fungal races containing the corresponding avirulence gene AVR-Pita. To develop effective strategies to control rice blast disease, a comprehensive study of the co-evolution of Pi-ta with AVR-Pita has been undertaken at USDA-ARS Dale Bumpers National Rice Research Center in cooperation with scientists from US, China and Colombia. A survey in the USDA rice collection of 52 Oryza species identified a total of 14 haplotypes of the Pi-ta allele. Translation of these Pi-ta haplotypes revealed 10 highly similar Pi-ta proteins. Bootstrapping and neighbor joining analysis suggest that these Pi-ta haplotypes belong to two major clades. Tijma’s D value suggests that the Pi-ta allele existed before the divergence of Oryza species, and the Pi-ta allele is under neutral mutation.

In contrast, the AVR-Pita allele was present in most fungal isolates examined. The virulent isolates, with altered AVR-Pita alleles, were detected from US, China and Colombia. Deletions and transposon insertions in the coding regions were found to be responsible for the structure variation of the AVR-Pita gene in virulent isolates. More point mutations leading to amino acid substitutions (nonsynonymous substitution, Ka) were observed than point mutation leading to silenced mutation (synonymous substitutions, Ks) in avirulent isolates suggesting that AVR-Pita is under diversified selection. These data indicate that the ratios of Ka/Ks is less than 1 for Pi-ta and greater than 1 for AVR-Pita. We suggest that Pi-ta and AVR-Pita evolve through trench warfare and will present the implications for crop protection.

The Relationship between the Pseudogenization of NBS-LRR Genes and the Functional Loss of Rice Blast Resistance Genes in Asian Cultivated Rice (Oryza sativa L.)

Jun Shang¹, Yong Tao¹, Cailin Lei², Xuewei Chen¹, Yan Zou¹, Jing Wang¹,

Meijun Zhang³, Z Jun hi Ke Lu³, Lihuang Zhu¹

1 State Key Laboratory of Plant Genomics & National Plant Gene Research Centre (Beijing), Institute of Genetics and Developmental Biology, Chinese Academy of Sciences, Beijing 100101, China

2 Institute of Crop Research, Chinese Academy of Agricultural Sciences, Beijing 100081, China

3 Beijing Institute of Genomics, Chinese Academy of Sciences, Beijing 101300, China

Rice blast, caused by Magnaporthe grisea, is one of the most devastating diseases. The two major subspecies of Asian cultivated rice (Oryza sativa L.), indica and japonica, have been shown obvious difference in rice blast resistance. We performed a genome-wide comparison of the NBS-LRR genes between the two sequenced rice genomes, 93-11 (indica) and Nipponbare (japonica). The pseudogenization of the NBS-LRR genes were found to some degree associated with the loss of the blast resistance. Using the NBS-LRR pseudogene sequences as genetic landing markers, we identified a novel blast R gene, Pid3. The allelic loci of Pid<?xml:namespace prefix = st1 ns = "urn:schemas-microsoft-com:office:smarttags" />3 in most of the tested japonica varieties were found as pseudogenes, suggesting that the pseudogenization of Pid3 in japonica is a gene loss event occurred in the evolution of cultivated rice in agreement with the “birth-and-death” model.

Development of Targeted Evolution System of Resistance Genes LRR (Leucine Rich Repeat) to Recognize

Blast Surface Protein

Shinji Kawasaki, Ken'Ichi Ikeda, Ko Hirano

NIAS (Natl. Inst. Agrobiological Sciences) Kannon-dai, Tsukuba, Ibaraki, JAPAN 305-8602

It seems that in most cases the specificity of the plant disease resistance genes may be held in their product’s LRR structure, probably often with the help of the other factors. Still, there are a few cases known that the resistance gene products interact directly with the avirulence gene products, such as that of Pita-AvrPita pair. Really, LRR is a state of art ligand recognizing system with stacked LRR units each with specific inner side amino acids, with much simpler genomic structure than that of immunoglobulins. Only the deficit of the plant disease resistance genes is that they can not keep in pace with the mutations of the ligand-coding avirulence genes of the parasites, at least in the artificial monoculture system. If an LRR of a resistance gene can be made to recognize the Achilles tendon of the parasite; the indispensable structure or domain of the vital enzyme or structural protein, this will be a very powerful and durable resistance gene. Or several kinds of such LRRs may be used as material of multi-line strategy, with no afraid of using up the resistance genes resource.

Therefore, we have developed a targeted evolution system of LRR. As a model case, we have preferred the blast surface protein MPG1 of the rice blast fungus Maganporthe grisea as the target, and the material LRR was supplied from those of Pib and FLS2. After constructing a randomly shuffled library of the LRR repeating units of these two genes in the bacterial two hybrid system, with E coli harboring the bait (target) plasmid of MPG1 gene, the best surviving clones were chosen in the selective condition. The resultant LRR showed significant affinity to the MPG1, comparable to those of Pita/AvirPita or Galacturonase/Inhibitor pairs. Although the affinity was weaker than that of the positive control of this hybrid system (bait: Gal4-LGF2/target: Gal11^P), after fine tuning with the Error-prone PCR system, finally the refined LRR was obtained, with the affinity comparable to the above control. We are now trying to confer the signal transducing ability to this LRR by swapping it to the Xa21 LRR, to induce the defense reactions in rice on infection.

Molecular Cloning and Characterization of the Rice Blast Resistance Gene Pi37 in the Well-known Cultivar St. No. 1

Fei Lin, Shen Chen, Zhiqun Que, Qinzhong Yang, Chunzhai, Lixia Hua, Xinqiong Liu, Ling Wang, and Qinghua Pan*

Laboratory of Plant Resistance and Genetics, College of Natural Resources & Environment, South China Agricultural University, Guangzhou 510642, China

* Corresponding author :E-mail: panqh@scau.edu.cn

The resistance (R) gene Pi37, present in the rice cultivar St. No. 1, was isolated by an in silico map-based cloning procedure. The equivalent genetic region in Nipponbare contains four NBS-LRR type loci. These four candidates for Pi37 (Pi37-1, -2, -3 and -4) were amplified separately from St. No. 1 via long-range PCR, and cloned into a binary vector. Each construct was individually transformed into the highly blast susceptible cultivar Q1063. The subsequent complementation analysis revealed Pi37-3 to be the functional gene, while -1, -2 and -4 are probably pseudogenes. Pi37 encodes a 1290 peptide NBS-LRR product, and the presence of substitutions at two sites in the NBS region (V239A and I247M) is associated with the resistance phenotype. Semi-quantitative expression analysis showed that in St. No. 1, Pi37 was constitutively expressed and only slightly induced by blast infection. Transient expression experiments indicated that the Pi37 product is restricted to the cytoplasm. Pi37-3 is thought to have evolved recently from -2, which in turn was derived from an ancestral -1 sequence. Pi37-4 is likely the most recently evolved member of the cluster, and probably represents a duplication of -3. The four Pi37 paralogs are more closely related to maize rp1 than to any of the currently isolated rice blast R genes Pita, Pib, Pi9, Pi2, Piz-t and Pi36.

Functional and Evolutionary Analysis of the Broad Spectrum Resistance Locus Pigm(t) to

Rice Blast Magnorpathe oryzae

Zuhua He¹, Yiwen Deng¹, Jing Xu¹, Hongqi Chen², Xudong Zhu²

1 National Key Laboratory of Plant Molecular Genetics, Institute of Plant Physiology and Ecology, Shanghai Institutes for Biological Sciences, Chinese Academy of Sciences, Shanghai, China

2 China National Rice Research Institute, Hangzhou 31006, China

Rice blast, caused by the fungal pathogen Magnorpathe oryzae, is one of the most destructive diseases of rice worldwide. The identification and utilization of broad-spectrum or durable resistance genes has been proven the most effective and economical approach to control the disease. A native Chinese variety, GM4, was identified with broad-spectrum and durable resistance and has been used in rice breeding for blast resistance for more than 20 years. Genetic and mapping analysis indicated that blast resistance to different races in GM4 is controlled by the same dominant locus designated as Pigm(t). The allelism test showed that Pi-gm(t) was allelic to Pi2 and Pi9, two known blast resistance genes. The map-based cloning strategy was employed with a large mapping population consisting of 1556 recessive individuals. Pigm(t) was finally mapped on chromosome 6 region between two markers c5483 and c0428. A BAC contig covering the Pigm(t) region was constructed and completely sequenced. An NBS-LRR gene cluster encompassing 10 NBS-LRR candidate resistance genes was identified in the 120-kb sequenced region, which contains 6 resistance genes in the Pi2/Pi9 gene cluster. Sequence comparison of the orthologous and paralogous genes in the Pigm(t) locus in both resistant and susceptible backgrounds showed that the Pigm(t) loci had undergone duplication during the evolution of the resistance cluster, suggesting that broad-spectrum disease resistance of GM4 might be conferred by a pair of duplicated R genes with sequence variation. Our results of resistance spectrum analysis showed that Pigm(t) confers a broader resistance to blast isolates from different cultivated regions than Pi9/Pi2/Piz^t/Piz, making Pigm(t) a good genetic resource for hybrid rice breeding for durable blast resistance with markers-assisted selection. Extensive genetic complement analysis of Pigm(t) is underway to dissect the Pigm(t)-mediated blast resistance.

Molecular Cloning and Gene Pyramiding of QTLs Controlling Field Resistance to Blast in Rice

Shuichi Fukuoka^1,*, Norikuni Saka³, Hironori Koga⁴, Takehiko Shimizu², Kaworu Ebana¹, Akira Takahasi¹, Hirohiko Hirochika¹, Masahiro Yano¹, Kazutoshi Okuno⁵

1 National Institute of Agrobiological Sciences, Kannondai 2-1-2 Tsukuba, Ibaraki 305-8602, Japan

2 Institute of the Society for Techno-Innovation of Agriculture, Forestry and Fisheries, Ippaizuka, Tsukuba, Ibaraki 305-0854, Japan

3 Mountainous Region Agricultural Research Institute, Aichi Agricultural Research Center, Inahasi, Toyota 441-2513, Aichi, Japan

4 Bioproduction Sciences, Ishikawa Prefectural University, Suematsu 1-308, Nonoichi-machi, Ishikawa 921-8836, Japan

5 Graduate School of Life and Environmental Sciences, University of Tsukuba, Tennodai 1-1-1, Tsukuba, Ibaraki 305–8572, Japan

* Corresponding author: E-mail: fukusan@affrc.go.jp

Field resistance in the Japanese upland rice cultivar Owarihatamochi is controlled by quantitative trait loci (QTLs) that confer durable resistance to diverse races of blast. We previously identified three QTLs and determined the map position of the QTL with the largest effect, the recessive resistance locus pi21 on chromosome 4. To characterize field resistance, we cloned the pi21 gene by a map-based strategy and used marker-assisted selection to develop experimental lines that contained multiple resistant QTLs.

A large-scale linkage analysis at the pi21 locus identified the sequence variations associated with the resistant/susceptible phenotypic difference. These variations cause two deletions at the amino acid level in a protein of unknown function encoded by the resistance allele. Transgenic plants with genomic fragments containing the Pi21 gene from a susceptible cultivar showed increased susceptibility to blast, whereas those with fragments containing the gene from a resistant cultivar showed no change in phenotype.

Backcrossed progeny lines, each carrying one of four resistant QTL (including one newly identified one) were developed in the genetic background of the susceptible cultivar, and two to five resistant QTLs were combined by marker-assisted selection. The disease severity of lines with multiple QTLs was lower than that of lines with a single QTL. In the pi21 resistant lines into which other resistance QTLs had been combined, the level of resistance was comparable to that of the original upland rice cultivar Owarihatamochi.

The results suggest that pi21 is a major QTL that belongs to a novel type of gene for plant defense. The combined effect of pi21 with some other resistance alleles may be the basis for the high level of field resistance in Japanese upland rice.

Identification, Mapping and Positional Cloning Rice Blast Resistance Gene Pi-kh from Rice Line Tetep

T.R. Sharma*, S. P. Kumar, N. Gautam, A.K. Rai, M.S. Madhav, H.C. Upreti and N. K. Singh

*Genoinformatics Laboratory National Research Centre on Plant Biotechnology

IARI, New Delhi –110012, India

* Corresponding author: E-mail: trsharma@nrcpb.org

Biotic stresses like Rice blast, Bacterial leaf blight, Sheath blight and Stem borer limiting rice productivity where ever rice is grown. Of these stresses, rice blast caused by Magnaporthe grisea (hebert) Barr is a serious constraint in rice production at global level. None of the rice cultivars possesses durable blast resistance because of the highly variable nature of the pathogen in the North-Western Himalayan region of India. Although chemical control of rice blast disease is feasible yet it remains environmentally unsafe. Developing blast resistance cultivars is important for sustainable management of the disease. Therefore, We have identified and mapped a blast resistance gene Pi-kh in rice line Tetep by using SSR markers at on long arm of rice chromosome 11. Once a skeleton genetic map was constructed using a mapping population of HP2216 (blast susceptible) and Tetep (blast resistant) rice lines, STMS markers linked to Pi-kh genes were mapped on the rice genome sequence of japonica type using sequence homology approach and a physical map having 1MB region was constructed. Since, known DNA markers were not available in the I MB region of constructed physical map, we identified simple sequence repeat (SSR) markers and designed 40 new SSR markers. Of these, only two were found flanking to the Pi-kh at 0.5 and 0.7 cM distance. The physical distance between these two markers was 142 kb. In this 142 kb genome sequence we predicted only one candidate gene for disease resistance belonging to the NBS and LRR categories of R-genes. PCR primers were designed from the Nipponbare sequence and used for amplification of the gene from blast resistant indica rice line Tetep. The Pi-kh gene was finally cloned and characterized at molecular level. Our experiment showed pathogen inducible nature of the gene. For functional validation of the gene we successfully produced transgenic Taipei 309 lines containing Pi-kh gene. Molecular analysis of these transgenics has shown stable integration of Pi-kh gene in the Taipei genome. Functional analysis of this gene using complementation test with M. grisea inoculation revealed that Pi-kh gene is expressing in the transgenic lines and imparting resistance to the pathogen. A complete story of the identification, mapping, cloning and functional validation of rice blast resistance gene Pi-kh will be presented.

Microarray-assisted Identification of Genes Associated with Blast Resistance in Rice

Yan Liu¹, Xiaoyuan Zhu², Shaohong Zhang², Menchu Bernardo¹, Jeremy Edward³, David Galbraith³, Jan Leach⁴, Hei Leung¹, Bin Liu²

1 Plant Breeding, Genetics and Biotechnology Division, IRRI, Los Banos, Philippines

2 Guangdong Academy of Agricultural Sciences, Guangdong, China

3 University of Arizona, Tucson, Arizona, USA

4 Colorado State University, Fort Collins, Colorado, USA

E-mail: lbgz_2006@yahoo.com

Previous work showed that SHZ-2, an indica cultivar grown in China, has broad spectrum resistance to multiple races of the blast pathogen. A number of QTL have been mapped using recombinant inbred lines derived from SHZ-2. We crossed SHZ-2 to TXZ-13, a blast susceptible but high-quality variety to produce two BC3 lines, BC10 and BC116. These two lines showed strong to moderate blast resistance over eight cropping seasons in the field. In particular, BC10 has been successfully used in the hybrid rice program in Guangdong Province to produce high-yielding and blast resistance hybrids.

To further dissect the QTL responsible for durable blast resistance, 451 BC4F3 lines were developed by backcrossing BC10 and BC116 to TXZ-13. The BC4F3 population consisted of 244 lines from BC10 and 207 lines from BC116. The two BC4 populations were evaluated for blast resistance in the greenhouse and the blast nursery at IRRI, Philippines. Chromosomal introgressions from SHZ-2 were identified by using a genome-wide, genotyping microarray produced at University of Arizona. This array consists of 880 oligos that detect single feature polymorphisms (insertions and deletions) in unique genes evenly spaced along the chromosomes (medium spacing about 250 kb). Hybridization was done by pairing Cy3/Cy5 labeled DNA of BC10 and BC116 with that of the recurrent parent TXZ-13 and graphical genotypes of BC10 and BC116 generated using the software of Graphical Genotyping (GGT) version 2.0. Based on data from four replicated hybridization experiments, 39 and 22 regions with SHZ-2 introgression segments were detected in BC10 and BC116, respectively. Some of the regions of SHZ-2 introgression shown by microarray data were checked by SSR markers located near the introgression regions. Of 42 SSR markers tested, 19 were consistent with the microarray data; the other SSR markers were monomorphic and hence not informative.

To determine the relationship between introgression regions and blast resistance, extreme resistant and susceptible BC4F3 lines were selected for analysis. We monitored the introgression of SHZ-2 alleles in BC4F3 lines using SSR markers and SNPs in defense genes assayed by TILLING (a procedure that detects SNP between two target sequences). Through a combination of genome-wide genotyping, SSR marker genotyping and detection of specific alleles, we found four regions were on chromosome 2, one on chromosome 6 (about 250 kb) and one on chromosome 9 (about 400 kb) associated with blast resistance in the advanced backcross lines. Within the narrow region on chromosome 6 and chromosome 9, several candidate defense genes including NB-LRR genes were found. We are testing whether coordination expression of these genes are responsible for the quantitative resistance observed in advanced backcross families.

Genomic Approaches for Development of Broad-spectrum Rice Variety for Leaf and Neck Blast

Chatchawan Jantasuriyarat¹, Pattama Sirithunya ², Tanee Sreewongchai¹, Saengchai

Sriprokhon ¹, Chanakarn Wongsaprom¹, Apichart Vanavichit ^1,3,* ,Theerayut Toojinda¹,

and Didier Tharreau ⁴

1 Rice Gene Discovery, National Center for Genetic Engineering and Biotechnology

(BIOTEC), Kasetsart University, Kamphangsaeng, Nakornpathom, 73140, Thailand

2 Rajamangala University of Technology Lanna, Science and Agricultural Technology

faculty, Chiengmai, 50000, Thailand

3 Agronomy Department, Kasetsart University, Kamphangsaeng Campus,

Nakornpathom, 73140, Thailand

4 UMR BGPI, INRA-ENSAM-CIRAD, TA73/09, 34398 Montpellier Cedex 05, France.

* Corresponding auther: E-mail: vanavichit@gmail.com

Rice blast, caused by the fungus, Magnaporthe grisea, is one of the most important diseases in rice production worldwide. Host resistance is the most effective way to control the disease. Genomic approaches including a survey of rice germplasm for new sources of resistance, a study of genetic structure of rice blast fungus, mapping of resistant and avirulence genes, and using of marker assisted selection (MAS) for gene pyramiding were used to develop broad spectrum blast resistant rice varieties. For the result, new source of rice leaf and neck blast resistant genes from Thai rice variety “Jao Hom Nin” (JHN), which shows broad-spectrum resistance to all but one group of rice blast isolates in Thailand, was mapped on rice chromosome 1 and 11. Pyramiding these QTLs for rice blast resistance from JHN with QTLs for rice blast resistance from IR64, which located on chromosome 2 and 12, result a broad-spectrum rice varieties which is resistant to all Thai blast isolates. Currently this rice variety is being used in rice breeding program in Thailand. A QTL for avirulence gene of rice blast fungus corresponding to leaf and neck blast resistant QTLs in JHN was mapped to the same region on blast chromosome 2 which can explain approximately 20 percent of phenotypic variance. Fine mapping of both resistance genes and avirulence genes are underway and positional cloning and identifying of their functions are our ultimate goal.

Regulation of Reactive Oxygen Production by the Binding of Rac GTPase to the N-terminal Extension of NADPH Oxidase of Rice

Hann Ling Wong¹, Reinhard Pinontoan¹, Kokoro Hayashi², Ryo Tabata², Takashi Yaeno³, Kana Hasegawa¹, Chojiro Kojima², Hirofumi Yoshioka⁴, Koh Iba³, Tsutomu Kawasaki¹, and Ko Shimamoto¹

1 Laboratory of Plant Molecular Genetics, Nara Institute of Science and Technology (NAIST), Ikoma, Japan

2 Laboratory of Biophysics, Nara Institute of Science and Technology (NAIST), Ikoma, Japan

3 Department of Biology, Kyushu University, Fukuoka, Japan. 4Laboratory of Defense in Plant-Pathogen Interactions, Nagoya University, Nagoya, Japan

Reactive oxygen species (ROS) produced by NADPH oxidase play critical roles in various cellular activities including defense against pathogens by plants. In contrast to the large multi-protein NADPH oxidase complex of phagocytes, in plants only the homologues of the catalytic subunit gp91phox and the cytosolic regulator small GTPase Rac are found. Plant homologues of the gp91phox subunit are known as Rboh (respiratory burst oxidase homologue). Although numerous Rboh have been isolated in plants, the regulation of enzymatic activity remains unknown. All rboh genes identified to date possess a conserved N-terminal extension that contains two Ca²⁺-binding EF-hand motifs. The involvement of Ca²⁺ in plant immune response has been implicated in many studies, but its exact role in ROS signaling remains unclear. Previously, we showed that a small GTPase Rac (OsRac1) enhanced elicitor-induced ROS production and resistance to pathogens in rice. In this study, using yeast two-hybrid assay, we found that interaction between Rac GTPases and the N-terminal extension is ubiquitous and that a substantial part of the N-terminal region of Rboh, including the two EF-hand motifs, is required for the interaction. The direct Rac-Rboh interaction was confirmed in further studies using in vitro pull-down assay, and in vivo fluorescence resonance energy transfer (FRET) microscopy. The FRET microscopy and in vitro NADPH oxidase assay also suggest that the direct Rac-Rboh interaction and the activation of NADPH oxidase activity are modulated by cytosolic Ca²⁺ concentration. Therefore, our study provides evidence that Ca²⁺ and Rac GTPase coordinately regulate ROS production in plant immune response.

Host Active Defense Responses Occur within 24 Hours after Pathogen Inoculation in the Rice Blast System

Zhonghua Wang¹, Yulin Jia², Hui Lin³, Barbara Valent⁴, J. Neil Rutger⁵

1 Institute of Biotechnology, Zhejiang Wanli University, Ningbo 315100, P. R. China

2 USDA-ARS Dale Bumpers National Rice Research Center, P. O. Box 1090, Stuttgart, AR 72160, USA

3 Adair Intern, Department of Plant Pathology, University of Arkansas, Fayetteville, AR 72701, USA

4 Department of Plant Pathology, Kansas State University, 4024 Throckmorton Plant Science Center, Manhattan, KS 66506-5502, USA

5 USDA-ARS Dale Bumpers National Rice Research Center, P. O. Box 1090, Stuttgart, AR 72160, USA

Phenotypical, cytological and molecular responses of rice to the fungus M. grisea were studied using rice cultivars and lesion mimic plants. Cultivar Katy was susceptible to several virulent Magnaporthe grisea isolates, and a Sekiguchi like-lesion mimic mutant of Katy (LmmKaty) was shown enhanced resistance to these isolates. Lesion mimic phenotype of LmmKaty was rapidly induced by virulent M. grisea isolates or by avirulent isolates only at high levels of inoculum. Autofluorescence (a sign of an active defense response) was visible under ultraviolet light 24 h after localized inoculation in the incompatible interaction whereas, autofluorescence was not evident in the compatible interaction. Autofluorescence was also observed in LmmKaty 20 h after pathogen inoculation, thus indicating that rapid cell death is a mechanism of LmmKaty to restrict pathogen invasion. Rapid accumulation of defense related (DR) gene transcripts, phenylalanine ammonia lyase and ß-glucanase, was observed beginning at 6 h and was obvious at 16 h and 24 h in an incompatible interaction. Rapid transcript accumulation of PR-1 and chitinase had occurred by 24 h after inoculation in an incompatible interaction. Accumulation of these transcripts was delayed in a compatible interaction. These results indicate that host active defense responses occur 24 h after pathogen inoculation and that LmmKaty exhibits enhanced resistance to M. grisea. We suggest that the autofluorescence and expression of the DR genes after heavy inoculation be important cytological and molecular markers respectively for early determination of the host response to M. grisea in the rice blast system.

RacGEFs Function as Activators of Small GTPase OsRac1 in Innate Immunity of Rice

Tsutomu Kawasaki, Keiko Imai, Hann Ling Wong, Yoji Kawano, Keita Nishide, Jun Okuda and Ko Shimamoto

Laboratory of Plant Molecular Genetics, Nara Institute of Science and Technology, Ikoma, Japan.

E-mail: kawasaki@bs.naist.jp

We have demonstrated that the small GTPase OsRac1 is a key regulator for induction of immune responses in rice. Activation of OsRac1 induces NADPH oxidase-mediated ROS production, PR gene expression, production of antimicrobial compounds, and lignification. Transgenic rice plants expressing constitutively active mutant of OsRac1 showed enhanced resistance of virulent races of Magnaporthe grisea and Xanthomonas oryzae, whereas hypersensitive response induced by avirulent M. grisea is suppressed by dominant-negative form of OsRac1. These results indicate that OsRac1 plays important roles in R gene-mediated resistance and basal resistance. In addition, OsRac1 was found to interact with important immune components including NBS-LRR, RAR1 and HSP90. However, how OsRac1 is activated during immune responses remains to be identified.

Recently, a new type of GDP-GTP exchange factor (GEF) for Rac/Rop has been found in plants, termed PRONE-type GEF. We found twelve PRONE-type RacGEFs in rice by data base search. All of them are constitutively expressed in most of rice tissues, suggesting that these RacGEFs may be activated at the post-translational level. To identify the RacGEFs that activate OsRac1, we purified all of the recombinant RacGEF proteins and examined the in vitro GEF activities against OsRac1. Four of them exhibited strong GEF activities for OsRac1, indicating that these RacGEF genes are possible candidates that regulate activation of OsRac1 in immune responses. Recently, the PRONE-type GEF was found to interact with LRR-type protein kinase similar to PAMPs receptors such as FLS2 and EFR. Therefore, the RacGEFs may regulate both PAMPs- and R gene-mediated disease resistances through activation of OsRac1 in rice.

Identification and Application of a Novel Bi-directional Promoter that Drives High Level Expression

of Genes in Monocot and Dicot Plants

Wensheng Zhao, Junhua Liu, Kezhen Yang, De Ye and You-Liang Peng

The State Key Laboratory for Agrobiotechnology, The MOA Key Laboratory for Molecular Plant Pathology and Department of Plant Pathology,

China Agricultural University.

In order to isolate a pathogen-inducible bi-directional promoter for engineering plant disease resistance, the authors firstly carried out large-scale identification of genes that were induced in rice leaves by infection of Magnapothe oryzae. A cDNA library was constructed using mRNA from the pathogen-infected leaves. Out of about 30000 recombinants from the library, about 2000 clones were identified as the induced cDNA and are subjected to sequencing. Sequence analysis showed that the cDNA clones were originated from 1075 rice genes and that two genes, named OsSCI2 and OsSCI3 occupying 21 and 12 copies of the cDNA clones were linked head by head in the genome. Northern blot analysis confirmed that these two genes are highly induced in rice leaves during blast infection. The above results suggest that the intergenic region between OsSCI2 and OsSCI3 is putative a bi-directional promoter.

To test the hypothesis, a DNA segment with 1152 bp length upstream of the ATG codon of the two genes was fused with GUS gene in opposite directions respectively. The fusion constructs were introduced into rice plants by Agrobaterium-mediated transformation. Analyses on T1 transgenic plants showed that the DNA segment drove the expression of a GUS gene in both directions, but the expressions were constitutive in roots and induced in leaves. The induced expressions were very high, 3 times and 1.2 times, and 28 and 12 times as high as that of CaMV 35S or rice ACTIN1 promoter, respectively. Transgenic rice lines were also generated containing a construct in which the GUS and LUC genes were fused to each of the ends respectively. Northern blot analysis showed that all tested T1 transgenic plants expressed the GUS and LUC genes at a high level in the leaves in response to blast infection. These results proved that the intergenic region between OsSCI2 and OsSCI3 could drive simultaneous expression of two genes.

In addition, transgenic lines of Arabidop