Genetic mapping showed that the rice blast avirulence gene AVR-Pita is tightly linked to a telomere on chromosome 3 in the plant pathogenic fungus Magnaporthe grisea . AVR-Pita corresponds in gene-for-gene fashion to the disease resistance ( R ) gene Pi-ta . Analysis of spontaneous avr-pita Ϫ mutants indicated that the gene is located in a telomeric 6.5-kb BglII restriction fragment. Cloning and DNA sequencing led to the identification of a candidate gene with features typical of metalloproteases. This gene is located entirely within the most distal 1.5 kb of the chromosome. When introduced into virulent rice pathogens, the cloned gene specifically confers avirulence toward rice cultivars that contain Pi-ta . Frequent spontaneous loss of AVR-Pita appears to be the result of its telomeric location. Diverse mutations in AVR-Pita , including point mutations, insertions, and deletions, permit the fungus to avoid triggering resistance responses mediated by Pi-ta . A point mutation in the protease consensus sequence abolishes the AVR-Pita avirulence function. INTRODUCTIONRice blast disease, caused by the fungus Magnaporthe grisea (Hebert) Barr (Rossman et al., 1990), remains one of the most important constraints to rice production worldwide, despite the availability of dozens of "major" disease resistance ( R ) genes, which are known as Pi genes (Yamada et al., 1976; Zeigler et al., 1994). Resistance mediated by Pi genes has not been durable, typically failing under field conditions soon after new cultivars are introduced to the field as new races, or pathotypes, of the pathogen appear with the ability to attack the previously resistant rice cultivars. Sasaki first distinguished differences in cultivar specificity between rice blast field isolates in 1922 (Yamada, 1985); since then, hundreds of pathogen races have been identified based on their infection spectra on differential rice cultivars. Recently, DNA fingerprinting analyses have shown that rice blast fungal populations are asexual and consist of dozens of families (discretely separable clonal lineages) having apparently predictable responses toward specific R genes (Levy et al., 1993; Zeigler et al., 1994). Lineage exclusion strategies are currently being tested as an aid for deploying R genes for genetic control of rice blast disease.The rice blast system is a classical gene-for-gene system (Flor, 1971) in which avirulence ( AVR ) genes in the pathogen show a functional correspondence with particular R genes in rice (Silué et al., 1992; Zeigler et al., 1994). AVR genes may encode pathogen molecules that are themselves recognized, directly or indirectly, by the corresponding R gene product, or they may encode enzymes involved in production of small molecule ligands that serve as recognition factors. Pathogen recognition triggers host resistance responses and stops infection. Field efficacy of any R gene thus depends on the biology of its associated AVR gene. Specifically, R gene efficacy depends on the nature of any function that an AVR gene may provide for th...

The rice blast resistance (R) gene Pi-ta mediates gene-for-gene resistance against strains of the fungus Magnaporthe grisea that express avirulent alleles of AVR-Pita. Using a map-based cloning strategy, we cloned Pi-ta, which is linked to the centromere of chromosome 12. Pi-ta encodes a predicted 928-amino acid cytoplasmic receptor with a centrally localized nucleotide binding site. A single-copy gene, Pi-ta shows low constitutive expression in both resistant and susceptible rice. Susceptible rice varieties contain pi-ta(-) alleles encoding predicted proteins that share a single amino acid difference relative to the Pi-ta resistance protein: serine instead of alanine at position 918. Transient expression in rice cells of a Pi-ta(+) R gene together with AVR-Pita(+) induces a resistance response. No resistance response is induced in transient assays that use a naturally occurring pi-ta(-) allele differing only by the serine at position 918. Rice varieties reported to have the linked Pi-ta(2) gene contain Pi-ta plus at least one other R gene, potentially explaining the broadened resistance spectrum of Pi-ta(2) relative to Pi-ta. Molecular cloning of the AVR-Pita and Pi-ta genes will aid in deployment of R genes for effective genetic control of rice blast disease.

We report a new fungal selectable marker that confers resistance to chlorimuron ethyl, a sulfonylurea herbicide. This gene as well as genes that confer resistance to hygromycin and bialaphos have been engineered to be compact and to eliminate sites for most common restriction enzymes. These three selectable markers have been used to construct a series of vectors for fungal transformation.

We have initiated a mutational analysis of pathogenicity in the rice blast fungus, Magnaporthe grisea, in which hygromycin-resistant transformants, most generated by restriction enzyme-mediated integration (REMI), were screened for the ability to infect plants. A rapid primary infection assay facilitated screening of 5,538 transformants. Twenty-seven mutants were obtained that showed a reproducible pathogenicity defect, and 18 of these contained mutations that cosegregated with the hygromycin resistance marker. Analysis of eight mutants has resulted in the cloning of seven PTH genes that play a role in pathogenicity on barley, weeping lovegrass, and rice. Two independent mutants identified the same gene, PTH2, suggesting nonrandom insertion of the transforming DNA. These first 7 cloned PTH genes are described.