Search

Development of resistant cultivars has been an effective method for controlling rice blast disease caused by Magnaporthe oryzae. Quantitative blast resistance genes may offer durable resistance because the selection pressure on M. oryzae to overcome resistance is low as a result of the genes’ moderate susceptibility. Because the effects of individual resistance genes are relatively small, pyramiding these genes in rice cultivars is a promising strategy. Here, we used near-isogenic and backcross lines of rice cultivar Koshihikari with single- or two-gene combinations of blast resistance genes (pi21, Pi34, and Pi35) to evaluate the suppression of leaf blast. The severity of the disease was assessed throughout the infection process. Resistance varied among the lines: Pi35 conferred the strongest resistance, while Pi34 showed the weakest effects. Two types of combined-gene interactions were observed, and they varied on the basis of gene combination and characteristic of the infection: (i) the combination of two resistance genes was more effective than either of the genes individually or (ii) the combination of two resistance genes was similar to the level of the most effective resistance gene in the pair. The most effective gene combination for the suppression of leaf blast was pi21 + Pi35.

The rice cultivar ‘Chumroo’ is commonly cultivated in the mid- and high-altitude areas of Bhutan. This cultivar has shown durable blast resistance in that area, without evidence of breakdown, for over 20 years. Chumroo was inoculated with 22 blast isolates selected from the race differential standard set of Japan. The cultivar showed resistance to all the isolates. To identify the resistance gene(s), Chumroo was crossed with a susceptible rice cultivar, Koshihikari. The F1 plants of the cross showed resistance. Segregation analyses of 300 F3 family lines fitted the segregation ratio of 1:2:1 and indicated that a single dominant gene controls the resistance to a blast isolate Ao 92-06-2 (race 337.1). The Chumroo resistance locus (termed Pi46(t)) was mapped between two SSR markers, RM6748 and RM5473, on the terminal region of the long arm of chromosome 4, using linkage analysis with SSR markers. The nearest marker, RM5473, was linked to the putative resistance locus at a map distance of 3.2 cM. At the chromosomal region, no true resistance genes were identified, whereas two field resistance genes were present. Therefore, we designated Pi46(t) as a novel blast resistance locus.

The vertical distribution of leaf blast lesions caused by the fungus Pyricularia grisea was studied to estimate the degree of leaf blast suppression in rice multilines in experimental paddy fields for 4 years. Leaf blast in 1 : 1 and 1 : 3 mixtures of susceptible rice cultivar Sasanishiki and its resistant near-isogenic line, Sasanishiki BL7, developed slower than that in pure stands of Sasanishiki. The average distance of lesions on leaves from the ground in the 1 : 3 mixtures was significantly lower than that in the pure stands at the end of leaf blast epidemics (at booting stage). This result shows that the distribution of leaf blast lesions in the upper layer differs between the susceptible pure stands and the 1 : 3 mixtures at the end of leaf blast epidemics.

To investigate the frequency distribution of purple seed stain of soybean caused by Cercospora kikuchii in two experimental fields in 2004, we set up rows 75 cm apart and sowed two asymptomatic seeds at each of positions 20 cm apart in each row. We sowed purple-stained seeds infected with the pathogen as inocula at four points instead of asymptomatic seeds in each field. We assessed disease incidence in harvested seeds by counting the numbers of purple-stained and asymptomatic seeds. To determine the spatial distribution of the disease, we grouped the field points into analytical units of several sizes. Beta-binomial and binomial distributions described the distribution patterns of purple-stained seeds. The smallest value of α, a beta-binomial parameter, occurred with analytical units that contained three or nine points next to each other within a single row, suggesting that these units showed the most aggregated distribution of the disease, each of the patches of seeds infected with C. kikuchii could be defined approximately by the area covered by three or nine points (75 × 60 or 75 × 180 cm), and the disease tended to infect plants next to each other within rows.

Leaf blast suppression in multilines was evaluated based on the number of susceptible lesions observed in a pure stand of susceptible rice cultivar Sasanishiki, and in 1 : 1 and 1 : 3 mixtures of Sasanishiki and a resistant near-isogenic line, Sasanishiki BL4 or BL7, from 1998 to 2001. The number of lesions first observed in fields in the 1 : 1 and 1 : 3 mixtures were close to theoretical numbers calculated using the number of lesions observed in the pure stands and the ratios of the susceptible Sasanishiki in the mixtures. The ratio of the number of lesions in the 1 : 1 and 1 : 3 mixtures to the number in the pure stand was 0.29 and 0.09, respectively. The relationship between these ratios and the ratios of susceptible Sasanishiki in mixtures was defined in an equation to estimate the degree of leaf blast suppression. Validation studies for the ratios of the number of lesions in the 1 : 1 and 1 : 3 mixtures to the number in the pure stand were conducted in two different locations and showed that the ratios are almost acceptable. The calculated autoinfection to alloinfection ratio was 1.3 and 1.4 in the 1 : 1 and 1 : 3 mixtures, respectively, suggesting that the calculated ratio will affect the degree of leaf blast suppression. Thus, predictors were obtained to estimate leaf blast suppression for effective blast control in multilines.

Azoxystrobin was recently registered in Japan for the control of purple seed stain of soybean caused by Cercospora kikuchii, because the pathogen has developed resistance to thiophanate-methyl. To investigate the effects of these fungicides on the frequency of C. kikuchii strains resistant to thiophanate-methyl and on the genotype structure of the population, we sowed purple-stained seeds, approximately 40% of which were infected with resistant strains, as inocula with asymptomatic seeds and applied thiophanate-methyl and azoxystrobin during the reproductive growth of soybeans. The isolation frequency of resistant strains increased more than 99% by thiophanate-methyl but was not significantly increased by azoxystrobin. In amplified fragment length polymorphism (AFLP) DNA fingerprinting, genotypic diversity was significantly decreased by thiophanate-methyl but was not affected by azoxystrobin. In addition, the similarity of the AFLP genotype structure was increased by thiophanate-methyl but not by azoxystrobin. These results suggest that thiophanate-methyl selectively inhibited the proliferation of sensitive strains, which resulted in a small number of genotypes, most of which were resistant strains. Azoxystrobin was found to nonselectively inhibit proliferation of the pathogen, which retained a large number of genotypes including thiophanate-methyl-sensitive or thiophanate-methyl-resistant strains or both.

We collected 247 isolates of Cercospora kikuchii from soybean seeds with typical purple stain symptoms from 15 prefectures in Japan. Of the 247 isolates, 93 were sensitive to thiophanate-methyl, a benzimidazole used to control this soybean disease; the remaining 154 were highly resistant to the fungicide. To examine genetic variability among the population of 247 isolates, we developed amplified fragment length polymorphism (AFLP) markers. An AFLP primer pair generated DNA fingerprint polymorphisms among the sample isolates, and with the unweighted pair-grouping method to cluster arithmetic means of the similarity coefficients among all pairs of the fingerprint patterns, the isolates were divided into four lineages (I to IV). Of the 247 isolates, 225 belonged to lineage I, including all isolates that were resistant to thiophanate-methyl. To determine whether the resistance of these isolates was related to mutations in the β-tubulin gene, we amplified partial nucleotide sequences of the gene from 29 representative isolates, including 12 that were resistant to thiophanate-methyl, by means of the polymerase chain reaction. The resistant isolates had identical nucleotide sequence with a one-step change at codon 198, in which the amino acid glutamic acid had been replaced by alanine. The evidence thus suggests that thiophanate-methyl resistance might have arisen in lineage I, the largest of the four lineages.

The japonica rice (Oryza sativa) cultivar Chubu 32 has a high level of partial resistance to blast, which is mainly controlled by a dominant resistance gene located on chromosome 11. The partial resistance to the rice blast fungus (Magnaporthe grisea) in Chubu 32 has isolate specificity; isolate IBOS8-1-1 is more aggressive on Chubu 32 than are other isolates. We hypothesized that the gene-for-gene relationship fits this case of a partial resistance gene in Chubu 32 against the avirulence gene in the pathogen. The partial resistance gene in Chubu 32 was mapped between DNA markers C1172 (and three other co-segregated markers) and E2021 and was designated Pi34. In the 32 F3 lines from the cross between a chromosome segment substitution line (Pi34−) from Koshihikari/Kasalath and Chubu 32, the lines with high levels of partial resistance to the M. grisea isolate Y93-245c-2 corresponded to the presence of Pi34 estimated by graphic genotyping. This indicated that Pi34 has partial resistance to isolate Y93-245c-2 in compatible interactions. The 69 blast isolates from the F1 progeny produced by the cross between Y93-245c-2 and IBOS8-1-1 were tested for aggressiveness on Chubu 32 and rice cultivar Koshihikari (Pi34−). The progeny segregated at a 1 : 1 ratio for strong to weak aggressiveness on Chubu 32. The results suggested that Y93-245c-2 has one gene encoding avirulence to Pi34 (AVRPi34), and IBOS8-1-1 is extremely aggressive on Chubu 32 because of the absence of AVRPi34. This is the first report of a gene-for-gene relationship between a fungal disease resistance gene associated with severity of disease and pathogen aggressiveness.

Background Rice blast is a destructive disease caused by Magnaporthe oryzae, and it has a large impact on rice production worldwide. Compared with leaf blast resistance, our understanding of panicle blast resistance is limited, with only one panicle blast resistance gene, Pb1, isolated so far. The japonica cultivar Miyazakimochi shows resistance to panicle blast, yet the genetic components accounting for this resistance remain to be determined. Results In this study, we evaluated the panicle blast resistance of populations derived from a cross between Miyazakimochi and the Bikei 22 cultivar, which is susceptible to both leaf and panicle blast. The phenotypic analyses revealed no correlation between panicle blast resistance and leaf blast resistance. Quantitative trait locus (QTL) analysis of 158 recombinant inbred lines using 112 developed genome-wide and 35 previously reported polymerase chain reaction (PCR) markers revealed the presence of two QTLs conferring panicle blast resistance in Miyazakimochi: a major QTL, qPbm11, on chromosome 11; and a minor QTL, qPbm9, on chromosome 9. To clarify the contribution of these QTLs to panicle blast resistance, 24 lines homozygous for each QTL were selected from 2,818 progeny of a BC2F7 backcrossed population, and characterized for disease phenotypes. The panicle blast resistance of the lines harboring qPbm11 was very similar to the resistant donor parental cultivar Miyazakimochi, whereas the contribution of qPbm9 to the resistance was small. Genotyping of the BC2F7 individuals highlighted the overlap between the qPbm11 region and a locus of the panicle blast resistance gene, Pb1. Reverse transcriptase PCR analysis revealed that the Pb1 transcript was absent in the panicles of Miyazakimochi, demonstrating that qPbm11 is a novel genetic component of panicle blast resistance. Conclusions This study revealed that Miyazakimochi harbors a novel panicle blast resistance controlled mainly by the major QTL qPbm11. qPbm11 is distinct from Pb1 and could be a genetic source for breeding panicle blast resistance, and will improve understanding of the molecular basis of host resistance to panicle blast.

Multiple functional polymorphisms in a single disease resistance gene in rice enhance durable resistance to blast

Effect of field resistance on the development of leaf blast in mixtures of susceptible and resistant rice cultivars was studied in three field experiments (one in the upland nursery and two in the paddy fields) by using two susceptible rice cultivars with different (low and intermediate) levels of field resistance to the disease. In the upland nursery experiment where infected leaves with blast fungus were strewed uniformly in the plots and in the paddy experiment in 1982 where the disease occurred naturally, the disease severity on respective susceptible cultivars decreased with the reduction in the proportion of the susceptible cultivars in the plots, and the decrease for the cultivar with a low level of field resistance (LFC) was greater than that for the cultivar with an intermediate level of the resistance (MFC). Similar results were obtained in the paddy experiment in 1981 where an inoculum source was placed at the center of each plot. Since leaf blast severity in each experiment was smaller for MFC than for LFC and the disease gradient on the susceptible cultivar from the inoculum source was steeper for MFC than for LFC in the paddy experiment in 1981, it was assumed that less spread of the disease between plants on MFC reduced the effectiveness of the mixtures for the control of the disease.

Differential systems of blast races based on the gene-for-gene relationship between host and pathogen were established in Japan by Yamada et al. (1976) and Kiyosawa (1984), and blast race monitoring was conducted nationwide in Japan in 1976, 1980, 1994, and 2001 using these systems. Those monitoring results showed that the frequencies of complete resistance genes in cultivated rice cultivars affected the prevalence of the corresponding avirulence genes in blast races, and such monitoring results have contributed to decisions regarding component proportions in three released Japanese multilines, as well as the development of rice cultivars with high levels of partial resistance. However, the optimal way for stable use of blast resistance in rice is not yet fully understood, because sufficiently quantitative analyses of blast race increases in host populations with different resistance genes, which are required to understand the optimal way, have not been conducted. For conducting the analyses, development of simple methods is necessary for monitoring blast races relative to resistance genotypes of the rice cultivars. These methods will facilitate accumulation of quantitative data on the increases of blast races in host populations with different resistance genes and such data will contribute to constructing reliable epidemiological simulation models to simulate blast race increases in the host populations.

The mating type locus (MAT1) of Magnaporthe oryzae has similar structural organization to MAT in other ascomycetes and encodes the mating type genes MAT1-1-1 with an alpha-box motif and MAT1-2-1 with an HMG-box motif in the MAT1-1 and MAT1-2 idiomorphs, respectively. Sequence and expression analyses of the MAT1 locus indicated a second open reading frame (ORF), MAT1-1-2, in the MAT1-1 idiomorph, and novel mating-type dependent ORFs (MAT1-1-3 and MAT1-2-2) at the locus. The MAT1-1-3 ORF initiated within the MAT1-1 idiomorph while the MAT1-2-2 ORF initiated at the border of the MAT1-2 idiomorph with both ORFs sharing most of their reading frames in the MAT1 flanking region. This suggests that the encoded proteins (MAT1-1-3 and MAT1-2-2) should be similar in their primary structures but can be distinguished by distinct N-termini with amino acids of 1 and 32, respectively, in each mating type. A CT dinucleotide repeat, (CT)n, present in the upstream region of MAT1-1-3, was polymorphic among the isolates.

Fusarium head blight (FHB) is a devastating disease of small grain cereal crops caused by the necrotrophic pathogen Fusarium graminearum and Fusarium culmorum. These fungi produce the trichothecene mycotoxin deoxynivalenol (DON) and its derivatives, which enhance the disease development during their interactions with host plants. For the self-protection, the trichothecene producer Fusarium species have Tri101 encoding trichothecene 3-O-acetyltransferase. Although transgenic expression of Tri101 significantly reduced inhibitory action of DON on tobacco plants, there are several conflicting observations regarding the phytotoxicity of 3-acetyldeoxynivalenol (3-ADON) to cereal plants; 3-ADON was reported to be highly phytotoxic to wheat at low concentrations. To examine whether cereal plants show sufficient resistance to 3-ADON, we generated transgenic rice plants with stable expression and inheritance of Tri101. While root growth of wild-type rice plants was severely inhibited by DON in the medium, this fungal toxin was not phytotoxic to the transgenic lines that showed trichothecene 3-O-acetylation activity. This is the first report demonstrating the DON acetylase activity and DON-resistant phenotype of cereal plants expressing the fungal gene.

For rice blast control with multilines, nine near-isogenic lines (NILs) with different complete blast-resistance genes were developed from a leading rice cv. Sasanishiki at Miyagi Prefectural Furukawa Agricultural Experiment Station in Japan. Four of these nine NILs, namely, the Pik, Pik-m, Piz and Piz-t lines were first registered as multilines named Sasanishiki BL (BL) in 1994, and the Pita-2, Pita and Pib lines were added to the registered NILs in 1997, 1998 and 2001, respectively. BL has been cultivated in farmers’ fields in Miyagi Prefecture from 1995, was planted on over 5,000 hectares in 1997. However, its cultivated area is now less than 800 hectares, since Sasanishiki has become unpopular. Resistance of the Sasanishiki multilines to blast has not been broken down for 7 years, and other NILs are also being developed from other leading rice cultivars in our experiment station. New rice multilines are expected to be cultivated for blast control.

Rice leaf blast disease incidence was investigated in mixed plantings of susceptible and resistant rice cultivars in the nursery and paddy field. In the blast nursery experiment, increase of rice leaf blast disease could be described by the equation dx/dt=rx(1-x), where x is the proportion of diseased area per plant of susceptible cultivar, t is the time in days and r is the apparent infection rate. The r value in mixed plantings of susceptible and resistant cultivars (rm) was in accordance with the equation rm=rs+clogem, where rs is r in single planting of susceptible cultivar, m is the proportion of susceptible cultivar in mixed plantings and c is a constant. In the paddy field experiment, the gradient of leaf blast disease incidence from inoculum sources which were placed in the center of plots was given by the equation log10y=A+Blog10x, where y is the number of acute-type lesions per hill of susceptible cultivar, x is the number of hills from inoculum source and A and B are parameters. The B values of plots, which represented leaf blast disease gradient from inoculum source, increased with both increasing ratio of susceptible cultivar in mixed plantings and with time. The presumed total number of acute-type lesions in plots was calculated by the equation log10y=A+Blog10x, using the A and B values for each plot. In order to compare rice leaf blast increase of each plot directly, The recalculated total number of acute-type lesions was obtained by dividing the presumed total number of acute-type lesions by the ratio of susceptible cultivar in the mixture. The recalculated total number of acute-type lesions in mixed plantings was smaller than that of single planting of susceptible cultivar. Infection rates between each investigated date which was calculated by the exponential growth model using the recalculated total number of acute-type lesions approximated Leonard's equation rm=rs+clogem.