Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A Recessive Mutation in Rice Conferring Non-Race-Specific Resistance to Bacterial Blight and Blast

A Recessive Mutation in Rice Conferring Non-Race-Specific Resistance to Bacterial Blight and Blast Rice (2009) 2:104–114 DOI 10.1007/s12284-009-9027-x A Recessive Mutation in Rice Conferring Non-Race-Specific Resistance to Bacterial Blight and Blast Ma. Reina Suzette Madamba & Nobuko Sugiyama & Alicia Bordeos & Ramil Mauleon & Kouji Satoh & Marietta Baraoidan & Shoshi Kikuchi & Ko Shimamoto & Hei Leung Received: 17 April 2009 /Accepted: 8 June 2009 /Published online: 26 June 2009 Springer Science + Business Media, LLC 2009 Abstract To understand the basis of broad-spectrum phenotype. The mutant provides a useful tool for investi- disease resistance in rice, we isolated a gamma-ray- gating the important genes responsible for non-race specific induced IR64 mutant G978 that showed enhanced resis- resistance to two distinct diseases. tance to blast and bacterial blight. The resistance is quantitative and non-race specific against the bacterial and Keywords Broad-spectrum disease resistance Gamma-ray . . . fungal pathogens. The mutation is inherited as a single mutagenesis Gene expression profiling IR64 recessive gene, designated as Bsdr1 and causes shorter QTL mapping stature relative to the wild type; however, it does not show lesion mimics phenotype under the conditions tested. The mutation was mapped as a quantitative trait locus to a 3.8- Introduction Mb region on chromosome 12. By comparing the gene expression profiles of the mutant and wild type, we Major resistance (R) genes have been used for disease identified a candidate gene encoding a U-box domain- control in many crops to provide resistance to specific containing protein. The disrupted gene showed a loss of pathogens. The R gene-mediated defenses typically involve expression in the mutant and co-segregated with mutant a rapid, localized cell death, or hypersensitive response (HR), and formation of antimicrobial chemicals that restrict pathogen invasion at the site of infection (Hammond-Kosack Ma. Reina Suzette Madamba, Nobuko Sugiyama, Alicia Bordeos contributed equally to this work. and Jones 1996; Keen 1990; Martin 1999). Some of these genes, however, lose effectiveness rapidly in cases where Electronic supplementary material The online version of this article the pathogens are capable of evading the recognition by (doi:10.1007/s12284-009-9027-x) contains supplementary material, which is available to authorized users. changing the corresponding effector (also called avirulence : : : : or Avr) genes. This has prompted an increasing interest in M. R. S. Madamba N. Sugiyama A. Bordeos R. Mauleon M. Baraoidan H. Leung (*) host defense genes which are not directly involved in the Division of Plant Breeding, Genetics and Biotechnology, recognition of specific strains of pathogens. These defense- International Rice Research Institute, response (DR) genes are implicated to contribute to DAPO 7777 Metro Manila, Philippines quantitative and broad-spectrum resistance to pathogens e-mail: h.leung@cgiar.org (Faris et al. 1999; Freialdenhoven et al. 1996; Liu et al. K. Satoh S. Kikuchi 2004; Lu et al. 2004; Ramalingam et al. 2003; Wen et al. Unit of Plant Genome Research, 2003). Broad-spectrum disease resistance is desirable National Institute of Agrobiological Sciences, because of its potential for conferring durable resistance Tsukuba, Japan against diverse races of a pathogen or multiple pathogens. K. Shimamoto Mutational analysis provides an effective means for Laboratory of Plant Molecular Genetics, identifying genes involved in different stages or compo- Nara Institute of Science and Technology, nents of resistance expressed by a plant. Mutants with 8916-5 Takayama, altered response to pathogens, either gain or loss of Ikoma 630-0101, Japan Rice (2009) 2:104–114 105 resistance, are particularly useful for dissecting defense In this study, we investigated the phenotypic character- mechanisms. In the case of gain-of-resistance mutants, istics and genetic control of a gain-of-resistance mutant of several studies have provided useful insights into the IR64, a widely grown indica rice variety. This mutant, regulation of defense (Feys and Parker 2000; Frye and G978, showed broad-spectrum resistance to BB and blast Innes 1998; Yu et al. 2000; Thatcher et al. 2005). In but did not exhibit LM under normal greenhouse con- Arabidopsis thaliana, a recessive mutant edr1 (enhanced ditions. We characterized the level of resistance of this disease resistance 1) presents a high level of resistance to mutant by artificial inoculation and by field evaluation at some bacterial and fungal pathogens (Frye and Innes 1998). both vegetative and reproductive stages. Genetic analyses EDR1 encodes a raf-like mitogen-activated protein kinase showed that the mutation is controlled by a recessive gene. kinase kinase that may function as a negative regulator of By genotyping the phenotypic extremes of a recombinant disease resistance (Frye et al. 2001). The recessive mlo population, we mapped the G978 mutation onto a short mutation in barley confers broad-spectrum resistance to segment on chromosome 12. Whole-genome gene expres- diverse isolates of the fungus Blumeria graminis f. sp. sion analysis of the mutant revealed 12 down-regulated hordei. The Mlo gene encodes a novel class of plant- genes within the mapped interval. Gene expression and co- specific integral membrane proteins anchored in the plasma segregation analysis further delimit the region, leading to membrane by seven transmembrane domains (Buschges et identification of a candidate gene that co-segregated with al. 1997; Devoto et al. 1999, 2003a). Mlo is found to be a the gain-of-resistance phenotype. The genetically and negative regulator of defense-responses and/or cell death phenotypically defined mutant provides a vehicle for such that the null mlo alleles mediate resistance by allowing investigating the important genes responsible for non-race- an elevated level of defense responses to occur constitu- specific resistance to two distinct diseases. tively during infection by pathogen, a function that is dependent on Ror1 and Ror2 (required for mlo-specified resistance)(Anderson et al. 2005; Peterhansel et al. 1997). Results In rice, bacterial blight (BB; caused by Xanthomonas oryzae pv. oryzae, Xoo) and blast (caused by Magnaporthe G978-18 exhibited non-race specific resistance oryzae) are the most important diseases worldwide and they to the bacterial blight and blast pathogens also serve as model systems for investigations of plant– pathogen interactions in monocots. Rice mutants with To determine its BB resistance spectra, the gain-of- altered response to diseases have been found (Campbell resistance IR64 mutant G978-18, isolated from a screen and Ronald 2005; Wang et al. 2004). Lesion mimic (LM) using Philippine Xoo strain PXO87, was tested against mutants showing broad-spectrum resistance to blast or representative strains of the ten Xoo races from the bacterial blight have been identified (Arase et al. 2000; Philippines. The two G978-18 sister lines showed highly Mizobuchi et al. 2002; Takahashi et al. 1999; Wu et al. enhanced resistance to all ten Xoo races, comparable to that 2008; Yin et al. 2000; Zeng et al. 2002). These mutants conferred by major R genes (Fig. 1a and b). The resistant develop spontaneous lesions that resemble disease symptoms mutants had mean lesion lengths ranging from 0.56– in the absence of pathogen attack, some of which have been 5.41 cm which were five to ten times shorter than those shown tobeinvolvedinprogrammedcelldeath,leading to in the susceptible sister line G978-12 and IR64. Even enhanced disease resistance to multiple pathogens. For though IR64 is incompatible to six Xoo races (races 1, 5, 7, example, the spl11 (spotted leaf) mutant confers non-race- 8, 9, and 10), the two G978-18 sister lines still had shorter specific resistance to blast and bacterial blight and the level lesions compared to the wild type (Fig. 1a). of resistance in this mutant correlates with the defense- G978-18 also showed partially enhanced resistance to blast related gene expression and lesion development on the when inoculated with IR64-compatible blast isolates PO6-6 leaves (Yin et al. 2000). Zeng et al. (2004)showed that Spl11 and Ca89 at seedling stage (Fig. 1c,Table 1). G978-18 showed encodes an E3 ubiquitin ligase which functions as a negative significant reductions in mean lesion density and percent regulator of cell death and defense through its role in protein diseased leaf area (% DLA) at low and high-inoculum quality control. Campbell and Ronald (2005) screened a fast- concentrations (25 and 75×10 spores/ml) compared to the neutron mutagenized rice (var. CO39) population and susceptible sister line G978-12-16 and IR64 (data for 75×10 identified three mutants—ebr1, ebr2,and ebr3 (enhanced spores/ml shown in Table 1). Consistent with the results of blast resistance) that display enhanced resistance to M. greenhouse inoculations, G978-18 exhibited enhanced resis- oryzae.Both ebr1 and ebr3 also confer enhanced resistance tance under natural infection in the blast nursery. In two to the BB pathogen. ebr3 has an LM phenotype upon independent tests, the two G978-18 sister lines showed less inoculation with M. oryzae and Xoo and the phenotype is disease, measured as % DLA at five time points up to 26 days also induced by a shift in environmental conditions. after sowing, than IR64 or G978-12-16 (Fig. 2). 106 Rice (2009) 2:104–114 Fig. 1 Enhanced resistance to bacterial blight and blast in IR64 Lesion length (cm) mutant G978-18. a Resistant lines G978-18-8 and G978-18- 30 16 compared with susceptible sister line G978-12-16 and IR64 (wild type) in inoculation tests against ten races of Xanthomo- nas oryzae pv. oryzae (Xoo) from the Philippines. Lesion length represents mean values of three replications. R resistant, S 5 susceptible. Error bars indicate ±SE. b Disease reaction of 12 34 5 6 7 8 9 10 G978-18-8, G978-12-16, and Xoo race IR64 to Xoo strain PXO87 (race G978-18-8 (R) G978-18-16 (R) G978-12-16 (S) IR64 3) and c enhanced resistance to blast isolate, Ca89 in G978-18- 8 compared to wild-type IR64. bc CO39 is a susceptible check. CO39 IR64 G978-18 (Control) (R) Blast (Ca89) G978-18-8 G978-12-16 IR64 (R) (S) (Control) Bacterial blight (PXO87) To determine the reaction of G978-18 to panicle blast, the mutant was grown to maturity under natural conditions in Cavinti, Laguna, Philippines, known to be a “hotspot” for blast. The two G978-18 mutant sister lines showed a four-fold reduction in disease compared to the susceptible Table 1 Mean Lesion Density and Percent Diseased Leaf Area (% sister line G978-12-16 and IR64 as measured by the disease DLA) of IR64-Derived Mutant Lines at Seedling Stage Using Two Blast Isolates, PO6-6 and Ca89 severity index (Electronic Supplementary Material Fig. S1). Thus, phenotypic characterization in greenhouse inocu- a a Line PO6-6 Ca89 lations and field tests supported the conclusion that the G978-18 mutant has broad-spectrum resistance against both Lesion % Lesion % density DLA density DLA bacterial blight and blast. Furthermore, resistance to panicle blast was effective in the field, an important consideration b b b b G978-18-8 2.44 5.62 0.24 0.67 for blast resistance. a a a a G978-12-16 5.12 32.47 2.92 14.30 a a a a IR64 4.05 27.47 3.84 17.90 Genetic analysis of resistance in G978 Inoculum concentration = 75 spores/ml The inheritance of resistance in G978 was studied using M 4 Means followed by a common letter are not significantly different at the 5% level by LSD and BC F segregation data. Table 2 shows the segregation 1 2 Rice (2009) 2:104–114 107 the inheritance pattern, we generated BC F progeny of 1 2 Trial 1 G978 backcrossed to IR64. The ratio of S and R plants for IR64 G978-derived F fitted the expected ratio of 3S:1R, G978-18-8 R consistent with the segregation in the M The frequency G978-12-16 S 4. distribution of lesion lengths in the F progeny after G978-18-16 R 2 inoculation with PXO87 was bimodal (ESM Fig. S2). Agronomically, the mutant was significantly shorter than IR64 (ESM Fig. S3). The number of tillers of G978-18 was reduced though reduction was not statistically significant (ESM Table S1). The pleiotropic effect of the mutation in G978 was confirmed by the co-segregation between disease resistance phenotype and short stature in the BC F progeny. 1 2 Together, the results of M segregation and backcross Trial 2 analysis support the conclusion that the mutation in G978- 18 is controlled by a single, recessive gene. Allelism tests between the mutation in G978 and the previously reported gain-of-resistance genes ebr1 and ebr3 (Campbell and Ronald 2005) were conducted. Under the conditions tested, the ebr1 × G978 F s(n=60), ebr3 × G978 F s(n=59) and 1 1 the CO39-derived ebr mutants were as susceptible to Xoo strain PXO99 as CO39 (data not shown). G978, on the other hand, showed enhanced resistance and had shorter lesions than the F s. These results suggest that the G978 mutation is non-allelic to either ebr1 or ebr3. We designated the mutation in G978-18 as Bsdr1 for Broad-spectrum 14 17 20 23 26 disease resistance because the mutation confers resistance Days after sowing to multiple strains for at least two pathogens tested. Fig. 2 Disease progress of leaf blast in two resistant G978-18 mutant lines compared to that in susceptible sister lines and IR64 in two Mapping of the Bsdr1 mutation independent field trials at IRRI blast nursery. Data points represent mean values of three replications. Error bars indicate ±SE. A total of 977 F progeny derived from the cross between for BB resistance among progenies of six susceptible M G978-18 and Azucena were inoculated with PXO87. The plants.Ineachofthe families,a 3susceptible (S):1 frequency distribution for disease reaction to PXO87 is resistant (R) ratio was observed when plants were presented in Fig. 3a. Segregation of F progeny fitted a inoculated with PXO87, suggesting that the mutation 3S:1R ratio (747S:230R, χ =1.11, P=0.29), consistent was conditioned by a single recessive gene. To confirm with the BC F segregation (Table 2) 1 2 . Table 2 Inheritance Pattern of Gain-Of-Resistance Phenotype in G978 M and Backcross Families 2 a Category Parent (S phenotype) Number of progeny Observed segregation ratio χ P value Susceptible Resistant M segregating family G978-12-2 20 15 5 0 1.0 G978-12-4 20 17 3 1.07 0.30 G978-12-5 20 12 8 2.40 0.12 G978-12-6 20 16 4 0.27 0.61 G978-12-7 19 15 4 0.16 0.69 G978-12-8 20 12 8 2.40 0.12 Pooled M families 119 89 30 0.003 0.96 BC F G978-18 × IR64 98 81 17 3.061 0.08 1 2 P-value≥0.05 indicates non-significant deviation from the 3 susceptible:1 resistance ratio based on Chi-square test Expected segregation ratio is 3S:1R if character is controlled by a single, recessive gene Percent diseased leaf area 108 Rice (2009) 2:104–114 Moderately Moderately Resistant Susceptible Resistant Susceptible IR64 G978-18-8 Azucena 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 Lesion length (cm) bc cM Marker Name Chrom 12 d Chrom 12 28.2 31.8 4.3 3.8 10.2 9.0 LOD score LOD score Fig. 3 Mapping of enhanced resistance phenotype using F progeny done using MAPManager QTX (Meer et al. 2002) c QTL region of G978-18 × Azucena. a Frequency distribution of lesion lengths defined by interval mapping with high confidence (based on bootstrap (cM) in F progeny inoculated with Xoo strain PXO87. b Five SSR and LOD=31.8) using mean lesion length. LOD score was based on markers (encircled in red) on chromosome 12 associated with BB 500 permutations. d QTL mapping using ratio of mean lesion length resistance phenotype (lesion length). The numbers on the left indicate over plant height as trait values tested for association with marker loci genetic distance (cM) in Kosambi units. Single-marker analysis was used (LOD=28.2). To determine the chromosomal location of the G978-18 Regression analysis of additional simple sequence repeat mutation, 73 polymorphic SSR markers from 12 chromo- (SSR) markers on chromosome 12 showed RM2529, somes were used to genotype 200 progeny representing the RM101, RM1036 and RM2935 to be highly associated phenotypic extremes of the F population. Six markers near with the resistance phenotype, with logarithm of odds the centromere on chromosome 12 showed highly signif- (LOD) scores of 26.6, 25.9, 25.1, and 23.9, respectively, icant association with the resistance phenotype based on the accounting for 42–46% of the variance (ESM Table S3). P value. Some SSR markers from chromosomes 3 and 4 However, RM1047 was considered not informative by also showed association with resistance but their effects marker regression function of MAPManager QTX, possibly were much smaller than those of the markers on chromo- because of segregation distortion in the progeny used. some 12 (ESM Table S2). They likely resulted from Thus, interval mapping was used to locate the mutation by segregation of QTL effects between IR64 and Azucena analyzing the quantitative effects of the mutation (MAP- (Ramalingam et al. 2003). Manager QTX; Meer et al. 2002). The map positions of the No. of F Plants 2 Rice (2009) 2:104–114 109 cDNA QTLs are shown in Fig. 3b. A QTL region on chromosome IR64 G978 12 was defined with high confidence based on bootstrap value. The five SSR markers detected previously by single- HOH Ca89 O Ca89 2 2 marker regression analysis were within this QTL region. It 0 24 48 24 48 0 24 48 24 48 had a LOD of 31.8 which was seven times higher than the AK102429 LOD score threshold established based on 500 permuta- tions (Fig. 3c). Because lesion length measurements might AK111521 be influenced by plant height/leaf length, we used ratios of AK064188 lesion length to plant height as trait values to test for association with marker loci. The analysis showed the same Ubiquitin markers to be significantly associated with resistance phenotype (LOD=28.2, Fig. 3d). Thus, several mapping Fig. 4 Expression analysis of three candidate genes AK102429, approaches placed the mutation within 3.8 Mb between AK11521, and AK064188 on chromosome 12. RT-PCR was performed by gene-specific primer pairs as described in the “Methods” section. positions 7.4 and 11.2 Mb (based on TIGR version 5.0 of Ubiquitin was used as a control. the rice pseudomolecules). Discussion Candidate genes identified from differentially expressed genes in chromosome 12 We have genetically defined a mutation that shows broad- spectrum resistance to multiple races of blast and BB It has been shown that genome-wide expression profiling pathogens. This study shows that mutations for quantitative could reveal candidate genes responsible for mutant phenotypes, assuming that the mutation affects the traits can be isolated from mutant lines provided that detailed phenotyping is done in replicated tests. Unlike expression of the target gene (Dybbs et al. 2005). Since major R genes, the resistance is quantitative. In G978-18, G978 was mutagenized by gamma ray, we expected that the the reduction in lesion length to BB is consistent across expression of the mutated gene could be down-regulated or multiple races of the pathogen. The degree of resistance, completely abolished. We used gene expression data of however, varied between experiments and genetic back- G978 and its wild-type parent IR64 (R. Mauleon et al, grounds. Typical of segregation of quantitative resistance, manuscript in preparation) to identify genes that were lesion length reduction varied among progeny derived from differentially expressed at 24 and 48 h after inoculation. By G978 × Azucena cross. Blast resistance in G978-18 was setting a log ratio of 0.5 threshold for difference in manifested as a reduction in lesion number rather than expression values between G978 and IR64 at either time prevention of lesion development. point, we identified 40 genes on chromosome 12 (ESM Table S4). To evaluate broad-spectrum resistance under realistic agronomic setting, it is important to have mutants without Based on the criterion of low expression at two time severe pleiotropic effects. The Bsdr1 mutation renders the points, seven genes were selected. Among them, three plant shorter with fewer tillers, but not severely abnormal as genes were located inside or relatively close to the is common with many gain-of-resistance mutants in mapped locus. These included AK102429, annotated as Arabidopsis (Shirano et al. 2002; Vogel and Somerville coding for a U-box and Arm-repeated protein, AK111521, 2000). The minor pleiotropic effect makes this mutant a good RNA-binding protein, and AK064188, unknown protein. candidate for more detailed phenotypic characterization. We Based on RT-PCR, we detected transcript for AK111521 were able to grow G978 to maturity under field conditions to with no indication of down regulation (Fig. 4). For determine that the Bsdr1 mutation conferred resistance to AK064188, we were able to detect a faint fragment with panicle blast. This result helps clarify the uncertainty of cDNA of G978. In contrast, no transcript was detected for AK102429 at all conditions tested, suggesting that the whether genes controlling leaf blast resistance are effective against panicle blast. Under our experimental conditions, AK102429 gene was disrupted. AK102429 is located resistance to leaf and neck blast was expressed in the field, within the mapped region between positions 7.4 and indicating that resistance in different tissue types can be 11.2 Mb whereas AK064188 was located outside the conferred by the same genetic mechanism. Identification of a mapped region. Southern blot analysis of segregating genetic factor responsible for reducing panicle blast is BC F progeny of G978 × IR64 further confirmed that 2 2 particularly valuable as many varieties tend to have strong the polymorphism in AK102429 co-segregated with gain- resistance to leaf blast yet succumb to panicle blast. of-resistance phenotype (Fig. 5b). Together, these results The gain-of-resistance mutation does not fall into the suggested that AK102429 is a promising candidate gene for common categories of lesion mimics mutants which the mutation. 110 Rice (2009) 2:104–114 Fig. 5 Co-segregation of resis- tance phenotype and AK102429 gene in G978 × IR64 BC F 2 2 centromere BamHI BamHI progeny. a BamHI restriction Genome (Chr12) sites in relation to the gene structure of AK102429 on chro- exon 5’ 3’ mosome 12 and the position of Probe probe used for Southern blot. b Southern blot analysis of G978 × IR64 BC F progeny (17 resis- 2 2 BC F Gain of Resistance plants tant and seven susceptible) using 2 2 AK102429 gene probe defined by the primer pair: UboxMF- 8kb TCCGTTCCACTT GATCACAA, UboxMR- GCCGAACCATTGTCACTT TT. Resistance to bacterial blight 3kb was measured as lesion length (cM) and indicated at the bottom of each lane.All resistantlines are homozygous for the mutant BC F Susceptible plants 2 2 allele, whereas the susceptible 8kb lines include both heterozygous and homozygous genotypes. 3kb involve initiation or propagation of cell death (Lorrain et al. Harushima et al. 1998). The large genetic distance could be 2003). At the macroscopic level, we did not observe leaf caused by a lack of recombination in the region, which may spots under the greenhouse conditions tested. Thus, the hinder positional cloning based on fine mapping. resistance does not appear to require cell death as a To overcome the potential problem of limited recombi- precondition for restricting disease spread. However, when nation around the mutation locus, we applied expression G978-18 was grown to maturity in the field, we observed analysis to short list candidate genes for more detailed darkening of the leaf sheath and of the panicle at heading. investigation. The physical boundaries defined by the Whether it is a result of cell death or accumulation of genetic markers are approximate due to expected differ- pigment under a relatively more stressful environment ences in the japonica versus indica genomes. Thus, without needs further investigation. Thus, we cannot rule out the additional filtering, it would be virtually impossible to possibility that lesion mimics could be formed under extract promising candidate genes within the 3.8-Mb stressed or growth-limiting conditions. Campbell and region. We reasoned that if the mutation had caused a Ronald (2005) reported three mutations ebr1, ebr2, and down expression of the mutated gene, it could be detected ebr3 with enhanced resistance to blast. The ebr1 and ebr3 by examining the expression of a subset of genes in the mutants also showed enhanced resistance to bacterial chromosomal region delimited by mapping (Fig. 3). Since blight. In addition, ebr3 showed lesion mimic phenotype the location of the mutation is known, we took a liberal when challenged by the pathogens. We tested for allelism approach to identify candidate genes in the target region to between the G978 and the ebr1 and ebr3 mutations. In our avoid making a false negative declaration (i.e., rather test, ebr1, ebr3, and their F s with G978 were all as committing a Type I than Type II error). In total, 40 down-expressed genes (mutant vs. wild type) were found susceptible to PXO99 as CO39. Based on allelism tests and the phenotypic descriptions of the ebr mutants, Bsdr1 from the genotypic contrasts. Of these, the putative U-box appears distinct from the ebr1, ebr2, and ebr3 mutations. gene showed null transcript and co-localized within the Using a genetic cross with Azucena, we mapped Bsdr1 3.8 Mb defined by QTL mapping. Co-segregation analysis onto a 30.8 cM region on chromosome 12. The LOD score is further supported the hypothesis that mutation in this U-box highly significant (P<0.0001) indicating a major phenotypic gene is causally related to the phenotype. effect of Bsdr1. Based on the current version of the Although the causal relationship between the U-box pseudomolecule, the physical distance defined by the gene and mutant phenotype remains to be verified, this flanking SSR markers is about 3.8 Mb, which is consider- candidate gene holds considerable promise based on its ably less than that predicted by the average physical distance predicted biological functions. Many U-Box genes function of 300 kb per cM (Arumuganathan and Earle 1991; in the ubiquitination protein degradation pathway that G978 G978 IR64 IR64 30.74 29.44 32.63 29.64 28.93 29.30 29.34 34.60 14.38 34.60 14.38 16.36 11.10 13.54 14.02 18.31 19.74 18.16 21.23 20.80 19.30 17.73 21.80 17.02 18.89 18.87 18.80 20.34 Rice (2009) 2:104–114 111 regulates cellular function through control of protein quality clipping method (Kauffman et al. 1973). Progeny from and turnovers (Devoto et al. 2003b; Moon et al. 2004). One putative gain-of-resistance mutants and “sister” M plants such U-Box gene in rice is encoded by the Spl11 locus (a total of six M lines per M ) were inoculated with 2 1 where a mutation gives enhanced resistance to blast and PXO87 to confirm the mutation. Selected BB gain-of- blight (Zeng et al. 2002, 2004). Spl11 encodes an E3 resistance mutants were later tested for blast resistance as ubiquitin ligase that functions as a negative regulator of described below. From screening approximately 10,000 M programmed cell death. It appears that mutation at Spl11 families, two gain-of-resistance mutants were identified but results in a deregulation of defense response leading to cell only analysis of G978 was considered in this study. death and an elevated level of resistance. Expression of resistance and the degree of lesion mimics are correlated in Characterization of resistance to bacterial blight spl11 mutant. Although we cannot conclude that Bsdr1 mutant is free of lesion mimics under all conditions, The gain-of-resistance mutant lines G978-18-8 and G978- resistance can be observed at stages with no apparent 18-16, together with susceptible wild-type sister line G978- lesion mimics. Thus, resistance conferred by Bsdr1 seems 12-16, were tested against representative strains of 10 Xoo to be temporally and spatially uncoupled from cell death. races from the Philippines in the greenhouse. The experi- The Bsdr1 mutation identified in this study provides an ment was done in three replications using a split-plot design entry point to identify critical steps in known or yet to be (four plants per line per isolate per replicate) using IR64 as discovered defense pathways. control. Plants were sown in pairs in 31×23 cm plastic In conclusion, through systematic screening, we have containers. The upper and lower leaves of three main tillers identified a rare mutant with gain-of-resistance to blast and of each plant were clip-inoculated at 45 das. Lesion lengths bacterial blight without serious pleiotropic effects. So far, were scored 14 days after inoculation. The representative enhanced resistance against multiple diseases in rice has strains of the ten races were PXO61 (race 1), PXO86 (race been reported mostly in lesion mimic mutants (Mizobuchi 2), PXO340 (race 3), PXO71 (race 4), PXO112 (race 5), et al. 2002; Wu et al. 2008), thus, determining how a PXO99 (race 6), PXO145 (race7), PXO280 (race 8), mutation with no pronounced cell death (lesion mimics) can PXO339 (race 9), and PXO341 (race 10). mediate multiple disease resistance is of practical interest. By combining mapping and gene expression profiling, we Characterization of resistance to blast narrowed down to a candidate gene responsible for the mutation. By examining transcripts rather than genomic The BB resistant G978-18-8 line was tested against blast DNA, this approach can be generally applicable to fungal isolates Ca89 and PO6-6 from the Philippines at 4 4 detecting mutations caused by deletions as well as other inoculum concentrations of 25×10 conidia/ml and 75×10 physical changes (e.g., point mutations or chromosomal conidia/ml. The susceptible sister line G978-12-16 and IR64 rearrangements) that abolish gene expression. The power of were included as controls. A randomized complete block this approach can be increased considerably through the use design with three replicates was used for each isolate and of the 44-K oligoarray (Agilent) or other chips with near- inoculum concentration. Fifteen seeds per line per replicate complete genome coverage. were sown in 35×28×11-cm plastic trays and grown in the greenhouse. Fourteen-day-old seedlings were sprayed with 100 ml of the spore suspension for each tray. Plants were Methods incubated in a chamber with 100% relative humidity for 24 h at 26°C and then transferred to an air-conditioned glasshouse Screening IR64 mutants for gain-of resistance room at 25–28°C. The seedlings were misted with water four times a day at daytime until scoring. Lesion density and Gamma-ray mutagenesis was conducted at the International percent diseased leaf area (DLA) were measured as Atomic Energy Agency, Austria, whereas the mutagenized described by Roumen (1993) and Notteghem et al. (1981) seeds were planted in the greenhouse at the International were scored 5–7 days after inoculation (dai). Rice Research Institute (IRRI), Philippines. Details of the production and maintenance of the IR64 mutants are Field evaluation under natural infection described in Wu et al. (2005). Because of the ease of using lesion length as a Two G978-18 lines, susceptible sister line G978-12-16, quantitative trait, our initial screen for gain-of-resistance IR64, and CO39 (susceptible check) were exposed to mutants was based on reaction to bacterial blight. Ten natural pathogen population in the blast nursery at IRRI, plants per M family were inoculated at 45 days after Philippines during the wet seasons of 2001 and 2003. Fifty sowing (das) with PXO87, a virulent Xoo isolate, by leaf- seeds per line were planted in half row in a 10×1.2-m 112 Rice (2009) 2:104–114 concrete bed. Five replicates and a randomized complete DNA extraction and microsatellite analysis block design were used. The plot was surrounded by spreader rows of susceptible cultivars IR50 and IR72 to DNA from individual G978-18 × Azucena F plants multiply the natural inocula. Percent DLA and lesion selected for mapping was isolated from fresh leaves using number was recorded at 14, 17, 20, 23, and 26 das. the CTAB method (Murray and Thompson, 1980). A total To evaluate the reaction of the mutant against panicle of 73 Rice Microsatellite (RM) primer pairs (Research blast, the same lines tested in the IRRI blast nursery were Genetics, Inc., Huntsville, Alabama) were used to genotype grown to maturity in a farmer’s field in Cavinti, Laguna, the selected 200 F progeny. Genomic DNA was used as a Philippines (a non-irrigated, upland rice system). A main template for PCR amplification with cycling conditions of plot of 4×21 m was established and divided into three 94°C for 5 min as hot start; 35 cycles of 94°C at 1 min for subplots to serve as three replicates. Each subplot (4×8 m ) denaturation; 55–60°C annealing for 1 min; 72°C extension was divided into four small plots. The five test materials for 2 min; and a 7 min final extension at 72°C. The PCR (including the control CO39) were planted in rows (22 hills products were run on a 40-cm long 4% denaturing per row per line with three seeds per hill) in each of the polyacrylamide gels for 1.5 h at 80 constant watts and small plots in a completely randomized design. Each of the silver stained according to the manufacturer’s instructions small plots was surrounded with spreader rows consisting (Promega, USA). of IR50 and IR72. Disease severity at reproductive stage The resulting bands were visually scored by comparing was assessed by scoring five panicles per hill and ten hills them to the bands of the resistant (G978-18) and suscep- per line per small plot per replicate using a 0–5 scale. tible (Azucena) parents. Linkage relationships were ana- Disease severity index (0–100) was computed based on a lyzed by Quantitative Trait Loci (QTL)/Simple Interval modified formula of Hildebrand et al. (2001) where n is the Mapping and Marker regression analysis with the MAP- number of panicles for ten hills in disease rating i, for i=0 Manager program (Meer et al. 2002) using the Kosambi to 5. Data were analyzed with ANOVA in SAS Program, mapping function (Kosambi 1944). VERSION 6.12 (SAS Institute, 1989). Microarray analysis to identify differentially expressed Genetic analysis and mapping genes in target region Inheritance of the mutation in G978 was inferred by Gene expression profiles of 21-day-old G978 at 24 h and observing the segregation of reaction to Xoo strain 48 h after blast inoculation using isolate Ca89 (100×10 PXO87 among M progeny of six susceptible M plants. conidia/ml) were obtained using the Agilent Technologies 4 3 To further confirm the inheritance, the mutant, G978-18 22K rice oligoarray, which represents about 50% of the was backcrossed to IR64 and the BC F (n=98) inoculated expressed genes in rice. Total RNA was isolated from 1 2 with PXO87 for segregation analysis. To test for allelism to whole plants of G978 and IR64 in each sampled time point previously reported gain-of-resistance mutations ebr1 and using Trizol (Life Technologies Inc., Gaithersburg, MD, ebr3 which also confer resistance to both BB and blast USA) protocol. For each array, blast-inoculated and water- (Campbell and Ronald 2005), a resistant G978 line was treated RNA samples from the same variety and same crossed to the two mutants. The F s, parents, and wild-type sampling time point were hybridized and compared (G978 CO39 and IR64 were inoculated with Xoo strain PXO99 at inoculated vs. water, IR64 inoculated vs. water) with three 45 das, and lesion lengths were measured at 17 dai. Allelic biological replicates per experiment. The experiments were test between the G978 mutation and ebr2, which showed conducted at the National Institute of Agrobiological enhanced resistance only to blast (Campbell and Ronald Sciences, Tsukuba, Japan. 2005), was not done. To identify differentially expressed genes in the chro- To locate the G978 mutation, we crossed G978-18 mosomal region where the mutation was mapped, we used mutant with Azucena, a traditional japonica with a high independent t tests (Pan 2002). We adopted a less stringent level of polymorphism relative to IR64. About 1,000 F test as we were primarily interested in finding candidates in plants were advanced to F .The F population was the target region and less concerned with having false 3 2 phenotyped for BB reaction using isolates PXO61 and positives. The independent t test for each experiment was PXO87. Because the resistance phenotype is quantitative, able to detect genes that were significantly expressed even F families were evaluated to confirm F phenotypes. Based with low expression ratios because unknown error variance 3 2 on the F and F phenotypes, 60 F individuals were components were estimated based on a gene-specific 2 3 2 classified as highly resistant (R), 60 as highly susceptible variance for each gene. For the independent t tests, the (S) and 80 as moderately susceptible. These 200 F plants significant differentially expressed genes (DEGs) were first were used for genotyping to map the mutation. determined for each rice entry-treatment comparison (e.g., Rice (2009) 2:104–114 113 G978 inoculated vs. water, three bioreps) using one-class Detection Systems (Amersham Biosciences) were used for t test for each gene. The mean of gene expression (log ratio) detection of hybridization signal. for three biological replicates was tested for significant Acknowledgments We thank P. Donini, F. Zapata, and R. Afza of difference against zero (no difference in gene expression International Atomic Energy Agency, Vienna, for conducting gamma- between samples compared). Significant DEGs between the ray mutagenesis from which the mutants understudied were identified. resistant and susceptible entries were then tested using two- This work is supported in part by Swiss Agency for Development and class t test, with G978 as class A and IR64 as class B, using Cooperation (SDC). three replicates. Each t test generated a list of significant and non-significant genes. At each time point, from the DEGs set, the subsets of genes that were significantly differentially References induced (DI genes) by blast inoculation in G978 were determined by the Boolean join. All t tests were done using Anderson JP, Thatcher L, Singh K. Plant defence responses: significance α=0.01 with adjusted Bonferroni correction, conservation between models and crops. Func Plant Biol. 2005;32:21–34. and p values based on 1,000 times permutation, as Arase S, Zhao C-M, Akimitsu K, Yamamoto M, Ichii M. A recessive implemented by TIGR Multiple Experiment Viewer v 3.1, lesion mimic mutant of rice with elevated resistance to fungal available at http://www.tm4.org/. pathogens. J Gen Plant Pathol. 2000;66:109–16. Arumuganathan K, Earle ED. Nuclear DNA content of some important plant species. Plant Mol Biol Rep. 1991;9:229–41. RT-PCR and Southern blot analyses Buschges R, Holricher K, Panstruga R, Simons G, Wolter M, Frijters A, et al. The barley Mlo gene: a novel control element of plant RT-PCR was used for RNA expression analysis of the three pathogen resistance. Cell. 1997;88:695–705. down-regulated genes (AK102429, AK111521, and Campbell MA, Ronald PC. Characterization of four rice mutants with alteration in the defense response pathway. Mol Plant Pathol. AK064188) from microarray result using the gain-of- 2005;6:11–21. resistance mutant and the wild-type IR64 at different time Devoto A, Piffanelli P, Nilsson I, Wallin E, Panstruga R, von Heijne G, et points after infection. A resistant BC F G978 line was 1 3 al. Topology, subcellular localization, and sequence diversity of the grown together with IR64 in the greenhouse. Plants were Mlo family in plants. J Biol Chem. 1999;274:34993–5004. Devoto A, Hartmann HA, Piffanelli P, Elliot C, Simmons C, inoculated with blast isolate Ca89 and with H O as control Taramino G, et al. Molecular phylogeny and evolution of the at 17 das. Leaves were collected at 0, 24, and 48 h after plant-specific seven transmembrane MLO family. J Mol Evol. inoculation, same time points for microarray expression 2003a;56:77–88. Devoto A, Muskett PR, Shirasu K. Role of ubiquitination in the analysis. Total RNA was isolated by TRIZOL Reagent regulation of plant defence against pathogens. Curr Opin Plant (Invitrogen). After DNase treatment by deoxyribonuclease I Biol. 2003b;6:307–11. amplification grade (Invitrogen), first-strand cDNA was Dybbs M, Ngai J, Kaplan JM. Using microarrays to facilitate synthesized with the Superscript™ II Reverse Transcriptase positional cloning:Identification of Tomosyn as an inhibitor of (Invitrogen). TAKARA Ex Taq (TAKARA Bio Inc.) was neurosecretion. PLoS Genet. 2005;1:e2. Faris J, Li WL, Liu DJ, Chen PD, Gill BS. Candidate gene analysis of used for PCR. The PCR condition used for these genes was quantitative disease resistance in wheat. Theor Appl Genet. 95°C for 30 s, 58°C for 30 s, 72°C for 60 s for 32 cycles. 1999;98:219–25. UboxMF-TCCGTTCCACTTGATCACAA, UboxMR- Feys BJ, Parker JE. Interplay of signaling pathways in plant disease GCCGAACCATTGTCACTTTT for AK102429, M10F- resistance. Trends Genet. 2000;16:449–55. Freialdenhoven A, Peterhansel C, Kurth J, Kreuzaler F, Schulze-Lefert CTCAAGCACAGGGAGAGGAG, M10R-CAAGGAGC P. Identification of genes required for the function of non-race- CTGGACAAGAAG for AK111521, and C04(144)F-CAG specific mlo resistance to powdery mildew in barley. Plant Cell. AGGCAGAGGCTTTTGAG, C04(413)R-GGACGGT 1996;8:5–14. CAAATGCAAGAAT for AK064188. Frye CA, Innes RW. An Arabidopsis mutant with enhanced resistance to powdery mildew. The Plant Cell. 1998;10:947–56. Southern blot analysis was used to determine co- Frye CA, Tang D, Innes RW. Negative regulation of defenses response segregation of BB resistance (using PXO87) and candidate in plants by a conserved MAPKK kinase. Proc Nat Acad Sci gene AK102429 in the BC F of G978 and IR64. 2 2 USA. 2001;98:784–6. AK102429 gene probe was defined by the primer pair: Hammond-Kosack KE, Jones JDG. Resistance gene-dependent plant defense responses. The Plant Cell. 1996;8:1773–91. UboxMF-TCCGTTCCACTTGATCACAA, UboxMR- Harushima Y, Yano M, Shomura A, Sato M, Shimano T, Kuboki Y, et GCCGAACCATTGTCACTTTT. DNA from selected sus- al. A high-density rice genetic linkage map with 2275 markers ceptible (n=7) and resistant (n=17) BC F was extracted as 2 2 using a single F2 population. Genetics. 1998;148:479–94. described above. Total DNA (10 µg) was digested by Hildebrand PD, McRae KB, Lu X. Factors affecting flower infection and disease severity of lowbush blueberry by Botrytis cinerea. BamHI (Amersham Biosciences), and the digested DNA Can J Plant Pathol. 2001;23:364–70. was resolved by electrophoresis using 0.9 % agarose gel. Kauffman HE, Reddy APK, Hsiek SPV, Marca SD. An improved DNA was transferred to Hybond N+ nylon membrane under technique for evaluating resistance of race varieties to Xantho- monas oryzae. Plant Dis Rep. 1973;57:537–41. alkaline condition. ECL Direct Nucleic Acid Labeling and 114 Rice (2009) 2:104–114 Keen NT. Gene-for-gene complementarity in plant–pathogen inter- site-leucine-rich repeat type R gene triggers defense responses actions. Annu Rev Genet. 1990;24:444–63. and results in enhanced disease resistance. The Plant Cell. Kosambi DD. The estimation of map distances from recombination 2002;4:3149–62. values. Ann Eugen. 1944;12:172–5. Takahashi A, Kawasaki T, Henmi K, Shii K, Kodama O, Satoh K, et Lorrain S, Vailleau F, Balague C, Roby D. Lesion mimic mutants: al. Lesion mimic mutants of rice with alterations in early keys for deciphering cell death and defense pathways in plants? signaling events of defense. Plant J. 1999;17:535–45. Trends Plant Sci. 2003;8:263–71. Thatcher L, Anderson J, Singh K. Plant defence responses: what we Liu B, Zhang S, Zhu X, Yang Q, Wu S, Mei M, et al. Candidate learnt from Arabidopsis? Functional Plant Biol. 2005;32:1–19. defense genes as predictors of quantitative blast resistance in rice. Vogel J, Somerville S. Isolation and characterization of powdery Mol Plant-Microbe Interact. 2004;17:1146–52. mildew-resistant Arabidopsis mutants. Proc Natl Acad Sci USA. Lu G, Jantasuriyarat C, Zhou B, Wang G-L. Isolation and character- 2000;97:1897–902. ization of novel defense response genes involved in compatible Wang GL, Wu C, Zeng L, He C, Baraoidan M, de Assis Goes da Silva and incompatible interactions between rice and Magnaporthe F, et al. Isolation and characterization of rice mutants compro- grisea. Theor Appl Genet. 2004;108:525–34. mised in Xa21-mediated resistance to X. oryzae pv. oryzae. Theor Martin GB. Functional analysis of resistance genes and their own Appl Genet. 2004;108:379–84. downstream effectors. Curr Opin Plant Biol. 1999;2:273–9. Wen N, Chu Z, Wang S. Three types of defense-responsive genes are Meer JM, Manly KF, Cudmore RH Jr. Map Manager QTXb14. involved in resistance to bacterial blight and fungal blast diseases Roswell Park Cancer Institute. 2002 in rice. Mol Gen Genomics. 2003;269:331–9. Mizobuchi R, Hirabayashi H, Kaji R, Nishizawa Y, Satoh H, Ogawa Wu C, Bordeos A, Madamba MRS, Baraoidan M, Ramos M, Wang T, et al. Differential expression of disease resistance in rice GL, et al. Rice lesion mimic mutants with enhanced resistance to lesion-mimic mutants. Plant Cell Rep. 2002;21:390–6. diseases. Mol Genet Genomics. 2008;279:605–19. Moon J, Parry G, Estelle M. The ubiquitin–proteoasome pathway and Wu J, Wu C, Lei C, Baraoidan M, Bordeos A, Madamba RS, et al. plant development. The Plant Cell. 2004;16:3181–95. Chemical- and irradiation-induced mutants of indica rice IR64 Notteghem JL, Chatel M, Dechanet R. In: Symposium sur la for forward and reverse genetics. Plant Mol Biol. 2005;59:85– resistance du riz a la pyriculariose. Montpellier, France. IRAT- 97. GERDAT 1981. pp 313-330. Yin Z, Chen J, Zeng L, Goh M, Leung H, Khush GS, et al. Pan W. A comparative review of statistical methods for discovering Characterizing rice lesion mimic mutants and identifying a differentially expressed genes in replicated microarray experi- mutant with broad-spectrum resistance to rice blast and bacterial ments. Bioinformatics. 2002;18(4):546–54. blight. Mol Plant-Microbe Interact. 2000;13:869–76. Peterhansel C, Freialdenhoven A, Kurth J, Kolsch R, Schulze Lefert P. Yu I, Fengler K, Clough S, Bent A. Identification of Arabidopsis Interaction analyses of genes required for resistance responses to mutants exhibiting an altered hypersensitive response in gene- powdery mildew in barley reveal distinct pathways leading to for-gene disease resistance. Mol Plant-Microbe Interact. leaf cell death. The Plant Cell. 1997;9:1397–409. 2000;13:277–86. Ramalingam J, Vera Cruz CM, Kukreja K, Chittoor JM, Wu JL, Lee Zeng L, Yin Z, Chen J, Leung H, Wang GL. Fine genetic mapping and SW, et al. Candidate resistance genes from rice, barley, and maize physical delimitation of the lesion mimic gene Spl11 to a 160-kb and their association with qualitative and quantitative resistance DNA segment of the rice genome. Mol Genet Genomics. in rice. Mol Plant-Microbe Interact. 2003;6:14–24. 2002;268:253–61. Roumen EC. Selection for partial resistance in rice to rice blast. Durability Zeng L, Qu S, Bordeos A, Yang C, Baraoidan M, Yan H, et al. Spl11, of Disease Resistance. Norwell: Kluwer; 1993. p. 195–200. a negative regulator of plant cell death and defense, encodes a U- Shirano Y, Kachroo P, Shah J, Klessig D. A gain-of-function mutation Box/ARM repeat protein endowed with E3 ubiquitin ligase in an Arabidopsis Toll Interleukin 1 receptor-nucleotide binding activity. The Plant Cell. 2004;16:2795–808. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Rice Springer Journals

A Recessive Mutation in Rice Conferring Non-Race-Specific Resistance to Bacterial Blight and Blast

Loading next page...
 
/lp/springer-journals/a-recessive-mutation-in-rice-conferring-non-race-specific-resistance-c37Cp6pAgz

References (55)

Publisher
Springer Journals
Copyright
Copyright © 2009 by Springer Science + Business Media, LLC
Subject
Life Sciences; Plant Sciences; Plant Genetics & Genomics; Plant Breeding/Biotechnology; Agriculture; Plant Ecology
ISSN
1939-8425
eISSN
1939-8433
DOI
10.1007/s12284-009-9027-x
Publisher site
See Article on Publisher Site

Abstract

Rice (2009) 2:104–114 DOI 10.1007/s12284-009-9027-x A Recessive Mutation in Rice Conferring Non-Race-Specific Resistance to Bacterial Blight and Blast Ma. Reina Suzette Madamba & Nobuko Sugiyama & Alicia Bordeos & Ramil Mauleon & Kouji Satoh & Marietta Baraoidan & Shoshi Kikuchi & Ko Shimamoto & Hei Leung Received: 17 April 2009 /Accepted: 8 June 2009 /Published online: 26 June 2009 Springer Science + Business Media, LLC 2009 Abstract To understand the basis of broad-spectrum phenotype. The mutant provides a useful tool for investi- disease resistance in rice, we isolated a gamma-ray- gating the important genes responsible for non-race specific induced IR64 mutant G978 that showed enhanced resis- resistance to two distinct diseases. tance to blast and bacterial blight. The resistance is quantitative and non-race specific against the bacterial and Keywords Broad-spectrum disease resistance Gamma-ray . . . fungal pathogens. The mutation is inherited as a single mutagenesis Gene expression profiling IR64 recessive gene, designated as Bsdr1 and causes shorter QTL mapping stature relative to the wild type; however, it does not show lesion mimics phenotype under the conditions tested. The mutation was mapped as a quantitative trait locus to a 3.8- Introduction Mb region on chromosome 12. By comparing the gene expression profiles of the mutant and wild type, we Major resistance (R) genes have been used for disease identified a candidate gene encoding a U-box domain- control in many crops to provide resistance to specific containing protein. The disrupted gene showed a loss of pathogens. The R gene-mediated defenses typically involve expression in the mutant and co-segregated with mutant a rapid, localized cell death, or hypersensitive response (HR), and formation of antimicrobial chemicals that restrict pathogen invasion at the site of infection (Hammond-Kosack Ma. Reina Suzette Madamba, Nobuko Sugiyama, Alicia Bordeos contributed equally to this work. and Jones 1996; Keen 1990; Martin 1999). Some of these genes, however, lose effectiveness rapidly in cases where Electronic supplementary material The online version of this article the pathogens are capable of evading the recognition by (doi:10.1007/s12284-009-9027-x) contains supplementary material, which is available to authorized users. changing the corresponding effector (also called avirulence : : : : or Avr) genes. This has prompted an increasing interest in M. R. S. Madamba N. Sugiyama A. Bordeos R. Mauleon M. Baraoidan H. Leung (*) host defense genes which are not directly involved in the Division of Plant Breeding, Genetics and Biotechnology, recognition of specific strains of pathogens. These defense- International Rice Research Institute, response (DR) genes are implicated to contribute to DAPO 7777 Metro Manila, Philippines quantitative and broad-spectrum resistance to pathogens e-mail: h.leung@cgiar.org (Faris et al. 1999; Freialdenhoven et al. 1996; Liu et al. K. Satoh S. Kikuchi 2004; Lu et al. 2004; Ramalingam et al. 2003; Wen et al. Unit of Plant Genome Research, 2003). Broad-spectrum disease resistance is desirable National Institute of Agrobiological Sciences, because of its potential for conferring durable resistance Tsukuba, Japan against diverse races of a pathogen or multiple pathogens. K. Shimamoto Mutational analysis provides an effective means for Laboratory of Plant Molecular Genetics, identifying genes involved in different stages or compo- Nara Institute of Science and Technology, nents of resistance expressed by a plant. Mutants with 8916-5 Takayama, altered response to pathogens, either gain or loss of Ikoma 630-0101, Japan Rice (2009) 2:104–114 105 resistance, are particularly useful for dissecting defense In this study, we investigated the phenotypic character- mechanisms. In the case of gain-of-resistance mutants, istics and genetic control of a gain-of-resistance mutant of several studies have provided useful insights into the IR64, a widely grown indica rice variety. This mutant, regulation of defense (Feys and Parker 2000; Frye and G978, showed broad-spectrum resistance to BB and blast Innes 1998; Yu et al. 2000; Thatcher et al. 2005). In but did not exhibit LM under normal greenhouse con- Arabidopsis thaliana, a recessive mutant edr1 (enhanced ditions. We characterized the level of resistance of this disease resistance 1) presents a high level of resistance to mutant by artificial inoculation and by field evaluation at some bacterial and fungal pathogens (Frye and Innes 1998). both vegetative and reproductive stages. Genetic analyses EDR1 encodes a raf-like mitogen-activated protein kinase showed that the mutation is controlled by a recessive gene. kinase kinase that may function as a negative regulator of By genotyping the phenotypic extremes of a recombinant disease resistance (Frye et al. 2001). The recessive mlo population, we mapped the G978 mutation onto a short mutation in barley confers broad-spectrum resistance to segment on chromosome 12. Whole-genome gene expres- diverse isolates of the fungus Blumeria graminis f. sp. sion analysis of the mutant revealed 12 down-regulated hordei. The Mlo gene encodes a novel class of plant- genes within the mapped interval. Gene expression and co- specific integral membrane proteins anchored in the plasma segregation analysis further delimit the region, leading to membrane by seven transmembrane domains (Buschges et identification of a candidate gene that co-segregated with al. 1997; Devoto et al. 1999, 2003a). Mlo is found to be a the gain-of-resistance phenotype. The genetically and negative regulator of defense-responses and/or cell death phenotypically defined mutant provides a vehicle for such that the null mlo alleles mediate resistance by allowing investigating the important genes responsible for non-race- an elevated level of defense responses to occur constitu- specific resistance to two distinct diseases. tively during infection by pathogen, a function that is dependent on Ror1 and Ror2 (required for mlo-specified resistance)(Anderson et al. 2005; Peterhansel et al. 1997). Results In rice, bacterial blight (BB; caused by Xanthomonas oryzae pv. oryzae, Xoo) and blast (caused by Magnaporthe G978-18 exhibited non-race specific resistance oryzae) are the most important diseases worldwide and they to the bacterial blight and blast pathogens also serve as model systems for investigations of plant– pathogen interactions in monocots. Rice mutants with To determine its BB resistance spectra, the gain-of- altered response to diseases have been found (Campbell resistance IR64 mutant G978-18, isolated from a screen and Ronald 2005; Wang et al. 2004). Lesion mimic (LM) using Philippine Xoo strain PXO87, was tested against mutants showing broad-spectrum resistance to blast or representative strains of the ten Xoo races from the bacterial blight have been identified (Arase et al. 2000; Philippines. The two G978-18 sister lines showed highly Mizobuchi et al. 2002; Takahashi et al. 1999; Wu et al. enhanced resistance to all ten Xoo races, comparable to that 2008; Yin et al. 2000; Zeng et al. 2002). These mutants conferred by major R genes (Fig. 1a and b). The resistant develop spontaneous lesions that resemble disease symptoms mutants had mean lesion lengths ranging from 0.56– in the absence of pathogen attack, some of which have been 5.41 cm which were five to ten times shorter than those shown tobeinvolvedinprogrammedcelldeath,leading to in the susceptible sister line G978-12 and IR64. Even enhanced disease resistance to multiple pathogens. For though IR64 is incompatible to six Xoo races (races 1, 5, 7, example, the spl11 (spotted leaf) mutant confers non-race- 8, 9, and 10), the two G978-18 sister lines still had shorter specific resistance to blast and bacterial blight and the level lesions compared to the wild type (Fig. 1a). of resistance in this mutant correlates with the defense- G978-18 also showed partially enhanced resistance to blast related gene expression and lesion development on the when inoculated with IR64-compatible blast isolates PO6-6 leaves (Yin et al. 2000). Zeng et al. (2004)showed that Spl11 and Ca89 at seedling stage (Fig. 1c,Table 1). G978-18 showed encodes an E3 ubiquitin ligase which functions as a negative significant reductions in mean lesion density and percent regulator of cell death and defense through its role in protein diseased leaf area (% DLA) at low and high-inoculum quality control. Campbell and Ronald (2005) screened a fast- concentrations (25 and 75×10 spores/ml) compared to the neutron mutagenized rice (var. CO39) population and susceptible sister line G978-12-16 and IR64 (data for 75×10 identified three mutants—ebr1, ebr2,and ebr3 (enhanced spores/ml shown in Table 1). Consistent with the results of blast resistance) that display enhanced resistance to M. greenhouse inoculations, G978-18 exhibited enhanced resis- oryzae.Both ebr1 and ebr3 also confer enhanced resistance tance under natural infection in the blast nursery. In two to the BB pathogen. ebr3 has an LM phenotype upon independent tests, the two G978-18 sister lines showed less inoculation with M. oryzae and Xoo and the phenotype is disease, measured as % DLA at five time points up to 26 days also induced by a shift in environmental conditions. after sowing, than IR64 or G978-12-16 (Fig. 2). 106 Rice (2009) 2:104–114 Fig. 1 Enhanced resistance to bacterial blight and blast in IR64 Lesion length (cm) mutant G978-18. a Resistant lines G978-18-8 and G978-18- 30 16 compared with susceptible sister line G978-12-16 and IR64 (wild type) in inoculation tests against ten races of Xanthomo- nas oryzae pv. oryzae (Xoo) from the Philippines. Lesion length represents mean values of three replications. R resistant, S 5 susceptible. Error bars indicate ±SE. b Disease reaction of 12 34 5 6 7 8 9 10 G978-18-8, G978-12-16, and Xoo race IR64 to Xoo strain PXO87 (race G978-18-8 (R) G978-18-16 (R) G978-12-16 (S) IR64 3) and c enhanced resistance to blast isolate, Ca89 in G978-18- 8 compared to wild-type IR64. bc CO39 is a susceptible check. CO39 IR64 G978-18 (Control) (R) Blast (Ca89) G978-18-8 G978-12-16 IR64 (R) (S) (Control) Bacterial blight (PXO87) To determine the reaction of G978-18 to panicle blast, the mutant was grown to maturity under natural conditions in Cavinti, Laguna, Philippines, known to be a “hotspot” for blast. The two G978-18 mutant sister lines showed a four-fold reduction in disease compared to the susceptible Table 1 Mean Lesion Density and Percent Diseased Leaf Area (% sister line G978-12-16 and IR64 as measured by the disease DLA) of IR64-Derived Mutant Lines at Seedling Stage Using Two Blast Isolates, PO6-6 and Ca89 severity index (Electronic Supplementary Material Fig. S1). Thus, phenotypic characterization in greenhouse inocu- a a Line PO6-6 Ca89 lations and field tests supported the conclusion that the G978-18 mutant has broad-spectrum resistance against both Lesion % Lesion % density DLA density DLA bacterial blight and blast. Furthermore, resistance to panicle blast was effective in the field, an important consideration b b b b G978-18-8 2.44 5.62 0.24 0.67 for blast resistance. a a a a G978-12-16 5.12 32.47 2.92 14.30 a a a a IR64 4.05 27.47 3.84 17.90 Genetic analysis of resistance in G978 Inoculum concentration = 75 spores/ml The inheritance of resistance in G978 was studied using M 4 Means followed by a common letter are not significantly different at the 5% level by LSD and BC F segregation data. Table 2 shows the segregation 1 2 Rice (2009) 2:104–114 107 the inheritance pattern, we generated BC F progeny of 1 2 Trial 1 G978 backcrossed to IR64. The ratio of S and R plants for IR64 G978-derived F fitted the expected ratio of 3S:1R, G978-18-8 R consistent with the segregation in the M The frequency G978-12-16 S 4. distribution of lesion lengths in the F progeny after G978-18-16 R 2 inoculation with PXO87 was bimodal (ESM Fig. S2). Agronomically, the mutant was significantly shorter than IR64 (ESM Fig. S3). The number of tillers of G978-18 was reduced though reduction was not statistically significant (ESM Table S1). The pleiotropic effect of the mutation in G978 was confirmed by the co-segregation between disease resistance phenotype and short stature in the BC F progeny. 1 2 Together, the results of M segregation and backcross Trial 2 analysis support the conclusion that the mutation in G978- 18 is controlled by a single, recessive gene. Allelism tests between the mutation in G978 and the previously reported gain-of-resistance genes ebr1 and ebr3 (Campbell and Ronald 2005) were conducted. Under the conditions tested, the ebr1 × G978 F s(n=60), ebr3 × G978 F s(n=59) and 1 1 the CO39-derived ebr mutants were as susceptible to Xoo strain PXO99 as CO39 (data not shown). G978, on the other hand, showed enhanced resistance and had shorter lesions than the F s. These results suggest that the G978 mutation is non-allelic to either ebr1 or ebr3. We designated the mutation in G978-18 as Bsdr1 for Broad-spectrum 14 17 20 23 26 disease resistance because the mutation confers resistance Days after sowing to multiple strains for at least two pathogens tested. Fig. 2 Disease progress of leaf blast in two resistant G978-18 mutant lines compared to that in susceptible sister lines and IR64 in two Mapping of the Bsdr1 mutation independent field trials at IRRI blast nursery. Data points represent mean values of three replications. Error bars indicate ±SE. A total of 977 F progeny derived from the cross between for BB resistance among progenies of six susceptible M G978-18 and Azucena were inoculated with PXO87. The plants.Ineachofthe families,a 3susceptible (S):1 frequency distribution for disease reaction to PXO87 is resistant (R) ratio was observed when plants were presented in Fig. 3a. Segregation of F progeny fitted a inoculated with PXO87, suggesting that the mutation 3S:1R ratio (747S:230R, χ =1.11, P=0.29), consistent was conditioned by a single recessive gene. To confirm with the BC F segregation (Table 2) 1 2 . Table 2 Inheritance Pattern of Gain-Of-Resistance Phenotype in G978 M and Backcross Families 2 a Category Parent (S phenotype) Number of progeny Observed segregation ratio χ P value Susceptible Resistant M segregating family G978-12-2 20 15 5 0 1.0 G978-12-4 20 17 3 1.07 0.30 G978-12-5 20 12 8 2.40 0.12 G978-12-6 20 16 4 0.27 0.61 G978-12-7 19 15 4 0.16 0.69 G978-12-8 20 12 8 2.40 0.12 Pooled M families 119 89 30 0.003 0.96 BC F G978-18 × IR64 98 81 17 3.061 0.08 1 2 P-value≥0.05 indicates non-significant deviation from the 3 susceptible:1 resistance ratio based on Chi-square test Expected segregation ratio is 3S:1R if character is controlled by a single, recessive gene Percent diseased leaf area 108 Rice (2009) 2:104–114 Moderately Moderately Resistant Susceptible Resistant Susceptible IR64 G978-18-8 Azucena 1 3 5 7 9 11 13 15 17 19 21 23 25 27 29 31 33 35 37 Lesion length (cm) bc cM Marker Name Chrom 12 d Chrom 12 28.2 31.8 4.3 3.8 10.2 9.0 LOD score LOD score Fig. 3 Mapping of enhanced resistance phenotype using F progeny done using MAPManager QTX (Meer et al. 2002) c QTL region of G978-18 × Azucena. a Frequency distribution of lesion lengths defined by interval mapping with high confidence (based on bootstrap (cM) in F progeny inoculated with Xoo strain PXO87. b Five SSR and LOD=31.8) using mean lesion length. LOD score was based on markers (encircled in red) on chromosome 12 associated with BB 500 permutations. d QTL mapping using ratio of mean lesion length resistance phenotype (lesion length). The numbers on the left indicate over plant height as trait values tested for association with marker loci genetic distance (cM) in Kosambi units. Single-marker analysis was used (LOD=28.2). To determine the chromosomal location of the G978-18 Regression analysis of additional simple sequence repeat mutation, 73 polymorphic SSR markers from 12 chromo- (SSR) markers on chromosome 12 showed RM2529, somes were used to genotype 200 progeny representing the RM101, RM1036 and RM2935 to be highly associated phenotypic extremes of the F population. Six markers near with the resistance phenotype, with logarithm of odds the centromere on chromosome 12 showed highly signif- (LOD) scores of 26.6, 25.9, 25.1, and 23.9, respectively, icant association with the resistance phenotype based on the accounting for 42–46% of the variance (ESM Table S3). P value. Some SSR markers from chromosomes 3 and 4 However, RM1047 was considered not informative by also showed association with resistance but their effects marker regression function of MAPManager QTX, possibly were much smaller than those of the markers on chromo- because of segregation distortion in the progeny used. some 12 (ESM Table S2). They likely resulted from Thus, interval mapping was used to locate the mutation by segregation of QTL effects between IR64 and Azucena analyzing the quantitative effects of the mutation (MAP- (Ramalingam et al. 2003). Manager QTX; Meer et al. 2002). The map positions of the No. of F Plants 2 Rice (2009) 2:104–114 109 cDNA QTLs are shown in Fig. 3b. A QTL region on chromosome IR64 G978 12 was defined with high confidence based on bootstrap value. The five SSR markers detected previously by single- HOH Ca89 O Ca89 2 2 marker regression analysis were within this QTL region. It 0 24 48 24 48 0 24 48 24 48 had a LOD of 31.8 which was seven times higher than the AK102429 LOD score threshold established based on 500 permuta- tions (Fig. 3c). Because lesion length measurements might AK111521 be influenced by plant height/leaf length, we used ratios of AK064188 lesion length to plant height as trait values to test for association with marker loci. The analysis showed the same Ubiquitin markers to be significantly associated with resistance phenotype (LOD=28.2, Fig. 3d). Thus, several mapping Fig. 4 Expression analysis of three candidate genes AK102429, approaches placed the mutation within 3.8 Mb between AK11521, and AK064188 on chromosome 12. RT-PCR was performed by gene-specific primer pairs as described in the “Methods” section. positions 7.4 and 11.2 Mb (based on TIGR version 5.0 of Ubiquitin was used as a control. the rice pseudomolecules). Discussion Candidate genes identified from differentially expressed genes in chromosome 12 We have genetically defined a mutation that shows broad- spectrum resistance to multiple races of blast and BB It has been shown that genome-wide expression profiling pathogens. This study shows that mutations for quantitative could reveal candidate genes responsible for mutant phenotypes, assuming that the mutation affects the traits can be isolated from mutant lines provided that detailed phenotyping is done in replicated tests. Unlike expression of the target gene (Dybbs et al. 2005). Since major R genes, the resistance is quantitative. In G978-18, G978 was mutagenized by gamma ray, we expected that the the reduction in lesion length to BB is consistent across expression of the mutated gene could be down-regulated or multiple races of the pathogen. The degree of resistance, completely abolished. We used gene expression data of however, varied between experiments and genetic back- G978 and its wild-type parent IR64 (R. Mauleon et al, grounds. Typical of segregation of quantitative resistance, manuscript in preparation) to identify genes that were lesion length reduction varied among progeny derived from differentially expressed at 24 and 48 h after inoculation. By G978 × Azucena cross. Blast resistance in G978-18 was setting a log ratio of 0.5 threshold for difference in manifested as a reduction in lesion number rather than expression values between G978 and IR64 at either time prevention of lesion development. point, we identified 40 genes on chromosome 12 (ESM Table S4). To evaluate broad-spectrum resistance under realistic agronomic setting, it is important to have mutants without Based on the criterion of low expression at two time severe pleiotropic effects. The Bsdr1 mutation renders the points, seven genes were selected. Among them, three plant shorter with fewer tillers, but not severely abnormal as genes were located inside or relatively close to the is common with many gain-of-resistance mutants in mapped locus. These included AK102429, annotated as Arabidopsis (Shirano et al. 2002; Vogel and Somerville coding for a U-box and Arm-repeated protein, AK111521, 2000). The minor pleiotropic effect makes this mutant a good RNA-binding protein, and AK064188, unknown protein. candidate for more detailed phenotypic characterization. We Based on RT-PCR, we detected transcript for AK111521 were able to grow G978 to maturity under field conditions to with no indication of down regulation (Fig. 4). For determine that the Bsdr1 mutation conferred resistance to AK064188, we were able to detect a faint fragment with panicle blast. This result helps clarify the uncertainty of cDNA of G978. In contrast, no transcript was detected for AK102429 at all conditions tested, suggesting that the whether genes controlling leaf blast resistance are effective against panicle blast. Under our experimental conditions, AK102429 gene was disrupted. AK102429 is located resistance to leaf and neck blast was expressed in the field, within the mapped region between positions 7.4 and indicating that resistance in different tissue types can be 11.2 Mb whereas AK064188 was located outside the conferred by the same genetic mechanism. Identification of a mapped region. Southern blot analysis of segregating genetic factor responsible for reducing panicle blast is BC F progeny of G978 × IR64 further confirmed that 2 2 particularly valuable as many varieties tend to have strong the polymorphism in AK102429 co-segregated with gain- resistance to leaf blast yet succumb to panicle blast. of-resistance phenotype (Fig. 5b). Together, these results The gain-of-resistance mutation does not fall into the suggested that AK102429 is a promising candidate gene for common categories of lesion mimics mutants which the mutation. 110 Rice (2009) 2:104–114 Fig. 5 Co-segregation of resis- tance phenotype and AK102429 gene in G978 × IR64 BC F 2 2 centromere BamHI BamHI progeny. a BamHI restriction Genome (Chr12) sites in relation to the gene structure of AK102429 on chro- exon 5’ 3’ mosome 12 and the position of Probe probe used for Southern blot. b Southern blot analysis of G978 × IR64 BC F progeny (17 resis- 2 2 BC F Gain of Resistance plants tant and seven susceptible) using 2 2 AK102429 gene probe defined by the primer pair: UboxMF- 8kb TCCGTTCCACTT GATCACAA, UboxMR- GCCGAACCATTGTCACTT TT. Resistance to bacterial blight 3kb was measured as lesion length (cM) and indicated at the bottom of each lane.All resistantlines are homozygous for the mutant BC F Susceptible plants 2 2 allele, whereas the susceptible 8kb lines include both heterozygous and homozygous genotypes. 3kb involve initiation or propagation of cell death (Lorrain et al. Harushima et al. 1998). The large genetic distance could be 2003). At the macroscopic level, we did not observe leaf caused by a lack of recombination in the region, which may spots under the greenhouse conditions tested. Thus, the hinder positional cloning based on fine mapping. resistance does not appear to require cell death as a To overcome the potential problem of limited recombi- precondition for restricting disease spread. However, when nation around the mutation locus, we applied expression G978-18 was grown to maturity in the field, we observed analysis to short list candidate genes for more detailed darkening of the leaf sheath and of the panicle at heading. investigation. The physical boundaries defined by the Whether it is a result of cell death or accumulation of genetic markers are approximate due to expected differ- pigment under a relatively more stressful environment ences in the japonica versus indica genomes. Thus, without needs further investigation. Thus, we cannot rule out the additional filtering, it would be virtually impossible to possibility that lesion mimics could be formed under extract promising candidate genes within the 3.8-Mb stressed or growth-limiting conditions. Campbell and region. We reasoned that if the mutation had caused a Ronald (2005) reported three mutations ebr1, ebr2, and down expression of the mutated gene, it could be detected ebr3 with enhanced resistance to blast. The ebr1 and ebr3 by examining the expression of a subset of genes in the mutants also showed enhanced resistance to bacterial chromosomal region delimited by mapping (Fig. 3). Since blight. In addition, ebr3 showed lesion mimic phenotype the location of the mutation is known, we took a liberal when challenged by the pathogens. We tested for allelism approach to identify candidate genes in the target region to between the G978 and the ebr1 and ebr3 mutations. In our avoid making a false negative declaration (i.e., rather test, ebr1, ebr3, and their F s with G978 were all as committing a Type I than Type II error). In total, 40 down-expressed genes (mutant vs. wild type) were found susceptible to PXO99 as CO39. Based on allelism tests and the phenotypic descriptions of the ebr mutants, Bsdr1 from the genotypic contrasts. Of these, the putative U-box appears distinct from the ebr1, ebr2, and ebr3 mutations. gene showed null transcript and co-localized within the Using a genetic cross with Azucena, we mapped Bsdr1 3.8 Mb defined by QTL mapping. Co-segregation analysis onto a 30.8 cM region on chromosome 12. The LOD score is further supported the hypothesis that mutation in this U-box highly significant (P<0.0001) indicating a major phenotypic gene is causally related to the phenotype. effect of Bsdr1. Based on the current version of the Although the causal relationship between the U-box pseudomolecule, the physical distance defined by the gene and mutant phenotype remains to be verified, this flanking SSR markers is about 3.8 Mb, which is consider- candidate gene holds considerable promise based on its ably less than that predicted by the average physical distance predicted biological functions. Many U-Box genes function of 300 kb per cM (Arumuganathan and Earle 1991; in the ubiquitination protein degradation pathway that G978 G978 IR64 IR64 30.74 29.44 32.63 29.64 28.93 29.30 29.34 34.60 14.38 34.60 14.38 16.36 11.10 13.54 14.02 18.31 19.74 18.16 21.23 20.80 19.30 17.73 21.80 17.02 18.89 18.87 18.80 20.34 Rice (2009) 2:104–114 111 regulates cellular function through control of protein quality clipping method (Kauffman et al. 1973). Progeny from and turnovers (Devoto et al. 2003b; Moon et al. 2004). One putative gain-of-resistance mutants and “sister” M plants such U-Box gene in rice is encoded by the Spl11 locus (a total of six M lines per M ) were inoculated with 2 1 where a mutation gives enhanced resistance to blast and PXO87 to confirm the mutation. Selected BB gain-of- blight (Zeng et al. 2002, 2004). Spl11 encodes an E3 resistance mutants were later tested for blast resistance as ubiquitin ligase that functions as a negative regulator of described below. From screening approximately 10,000 M programmed cell death. It appears that mutation at Spl11 families, two gain-of-resistance mutants were identified but results in a deregulation of defense response leading to cell only analysis of G978 was considered in this study. death and an elevated level of resistance. Expression of resistance and the degree of lesion mimics are correlated in Characterization of resistance to bacterial blight spl11 mutant. Although we cannot conclude that Bsdr1 mutant is free of lesion mimics under all conditions, The gain-of-resistance mutant lines G978-18-8 and G978- resistance can be observed at stages with no apparent 18-16, together with susceptible wild-type sister line G978- lesion mimics. Thus, resistance conferred by Bsdr1 seems 12-16, were tested against representative strains of 10 Xoo to be temporally and spatially uncoupled from cell death. races from the Philippines in the greenhouse. The experi- The Bsdr1 mutation identified in this study provides an ment was done in three replications using a split-plot design entry point to identify critical steps in known or yet to be (four plants per line per isolate per replicate) using IR64 as discovered defense pathways. control. Plants were sown in pairs in 31×23 cm plastic In conclusion, through systematic screening, we have containers. The upper and lower leaves of three main tillers identified a rare mutant with gain-of-resistance to blast and of each plant were clip-inoculated at 45 das. Lesion lengths bacterial blight without serious pleiotropic effects. So far, were scored 14 days after inoculation. The representative enhanced resistance against multiple diseases in rice has strains of the ten races were PXO61 (race 1), PXO86 (race been reported mostly in lesion mimic mutants (Mizobuchi 2), PXO340 (race 3), PXO71 (race 4), PXO112 (race 5), et al. 2002; Wu et al. 2008), thus, determining how a PXO99 (race 6), PXO145 (race7), PXO280 (race 8), mutation with no pronounced cell death (lesion mimics) can PXO339 (race 9), and PXO341 (race 10). mediate multiple disease resistance is of practical interest. By combining mapping and gene expression profiling, we Characterization of resistance to blast narrowed down to a candidate gene responsible for the mutation. By examining transcripts rather than genomic The BB resistant G978-18-8 line was tested against blast DNA, this approach can be generally applicable to fungal isolates Ca89 and PO6-6 from the Philippines at 4 4 detecting mutations caused by deletions as well as other inoculum concentrations of 25×10 conidia/ml and 75×10 physical changes (e.g., point mutations or chromosomal conidia/ml. The susceptible sister line G978-12-16 and IR64 rearrangements) that abolish gene expression. The power of were included as controls. A randomized complete block this approach can be increased considerably through the use design with three replicates was used for each isolate and of the 44-K oligoarray (Agilent) or other chips with near- inoculum concentration. Fifteen seeds per line per replicate complete genome coverage. were sown in 35×28×11-cm plastic trays and grown in the greenhouse. Fourteen-day-old seedlings were sprayed with 100 ml of the spore suspension for each tray. Plants were Methods incubated in a chamber with 100% relative humidity for 24 h at 26°C and then transferred to an air-conditioned glasshouse Screening IR64 mutants for gain-of resistance room at 25–28°C. The seedlings were misted with water four times a day at daytime until scoring. Lesion density and Gamma-ray mutagenesis was conducted at the International percent diseased leaf area (DLA) were measured as Atomic Energy Agency, Austria, whereas the mutagenized described by Roumen (1993) and Notteghem et al. (1981) seeds were planted in the greenhouse at the International were scored 5–7 days after inoculation (dai). Rice Research Institute (IRRI), Philippines. Details of the production and maintenance of the IR64 mutants are Field evaluation under natural infection described in Wu et al. (2005). Because of the ease of using lesion length as a Two G978-18 lines, susceptible sister line G978-12-16, quantitative trait, our initial screen for gain-of-resistance IR64, and CO39 (susceptible check) were exposed to mutants was based on reaction to bacterial blight. Ten natural pathogen population in the blast nursery at IRRI, plants per M family were inoculated at 45 days after Philippines during the wet seasons of 2001 and 2003. Fifty sowing (das) with PXO87, a virulent Xoo isolate, by leaf- seeds per line were planted in half row in a 10×1.2-m 112 Rice (2009) 2:104–114 concrete bed. Five replicates and a randomized complete DNA extraction and microsatellite analysis block design were used. The plot was surrounded by spreader rows of susceptible cultivars IR50 and IR72 to DNA from individual G978-18 × Azucena F plants multiply the natural inocula. Percent DLA and lesion selected for mapping was isolated from fresh leaves using number was recorded at 14, 17, 20, 23, and 26 das. the CTAB method (Murray and Thompson, 1980). A total To evaluate the reaction of the mutant against panicle of 73 Rice Microsatellite (RM) primer pairs (Research blast, the same lines tested in the IRRI blast nursery were Genetics, Inc., Huntsville, Alabama) were used to genotype grown to maturity in a farmer’s field in Cavinti, Laguna, the selected 200 F progeny. Genomic DNA was used as a Philippines (a non-irrigated, upland rice system). A main template for PCR amplification with cycling conditions of plot of 4×21 m was established and divided into three 94°C for 5 min as hot start; 35 cycles of 94°C at 1 min for subplots to serve as three replicates. Each subplot (4×8 m ) denaturation; 55–60°C annealing for 1 min; 72°C extension was divided into four small plots. The five test materials for 2 min; and a 7 min final extension at 72°C. The PCR (including the control CO39) were planted in rows (22 hills products were run on a 40-cm long 4% denaturing per row per line with three seeds per hill) in each of the polyacrylamide gels for 1.5 h at 80 constant watts and small plots in a completely randomized design. Each of the silver stained according to the manufacturer’s instructions small plots was surrounded with spreader rows consisting (Promega, USA). of IR50 and IR72. Disease severity at reproductive stage The resulting bands were visually scored by comparing was assessed by scoring five panicles per hill and ten hills them to the bands of the resistant (G978-18) and suscep- per line per small plot per replicate using a 0–5 scale. tible (Azucena) parents. Linkage relationships were ana- Disease severity index (0–100) was computed based on a lyzed by Quantitative Trait Loci (QTL)/Simple Interval modified formula of Hildebrand et al. (2001) where n is the Mapping and Marker regression analysis with the MAP- number of panicles for ten hills in disease rating i, for i=0 Manager program (Meer et al. 2002) using the Kosambi to 5. Data were analyzed with ANOVA in SAS Program, mapping function (Kosambi 1944). VERSION 6.12 (SAS Institute, 1989). Microarray analysis to identify differentially expressed Genetic analysis and mapping genes in target region Inheritance of the mutation in G978 was inferred by Gene expression profiles of 21-day-old G978 at 24 h and observing the segregation of reaction to Xoo strain 48 h after blast inoculation using isolate Ca89 (100×10 PXO87 among M progeny of six susceptible M plants. conidia/ml) were obtained using the Agilent Technologies 4 3 To further confirm the inheritance, the mutant, G978-18 22K rice oligoarray, which represents about 50% of the was backcrossed to IR64 and the BC F (n=98) inoculated expressed genes in rice. Total RNA was isolated from 1 2 with PXO87 for segregation analysis. To test for allelism to whole plants of G978 and IR64 in each sampled time point previously reported gain-of-resistance mutations ebr1 and using Trizol (Life Technologies Inc., Gaithersburg, MD, ebr3 which also confer resistance to both BB and blast USA) protocol. For each array, blast-inoculated and water- (Campbell and Ronald 2005), a resistant G978 line was treated RNA samples from the same variety and same crossed to the two mutants. The F s, parents, and wild-type sampling time point were hybridized and compared (G978 CO39 and IR64 were inoculated with Xoo strain PXO99 at inoculated vs. water, IR64 inoculated vs. water) with three 45 das, and lesion lengths were measured at 17 dai. Allelic biological replicates per experiment. The experiments were test between the G978 mutation and ebr2, which showed conducted at the National Institute of Agrobiological enhanced resistance only to blast (Campbell and Ronald Sciences, Tsukuba, Japan. 2005), was not done. To identify differentially expressed genes in the chro- To locate the G978 mutation, we crossed G978-18 mosomal region where the mutation was mapped, we used mutant with Azucena, a traditional japonica with a high independent t tests (Pan 2002). We adopted a less stringent level of polymorphism relative to IR64. About 1,000 F test as we were primarily interested in finding candidates in plants were advanced to F .The F population was the target region and less concerned with having false 3 2 phenotyped for BB reaction using isolates PXO61 and positives. The independent t test for each experiment was PXO87. Because the resistance phenotype is quantitative, able to detect genes that were significantly expressed even F families were evaluated to confirm F phenotypes. Based with low expression ratios because unknown error variance 3 2 on the F and F phenotypes, 60 F individuals were components were estimated based on a gene-specific 2 3 2 classified as highly resistant (R), 60 as highly susceptible variance for each gene. For the independent t tests, the (S) and 80 as moderately susceptible. These 200 F plants significant differentially expressed genes (DEGs) were first were used for genotyping to map the mutation. determined for each rice entry-treatment comparison (e.g., Rice (2009) 2:104–114 113 G978 inoculated vs. water, three bioreps) using one-class Detection Systems (Amersham Biosciences) were used for t test for each gene. The mean of gene expression (log ratio) detection of hybridization signal. for three biological replicates was tested for significant Acknowledgments We thank P. Donini, F. Zapata, and R. Afza of difference against zero (no difference in gene expression International Atomic Energy Agency, Vienna, for conducting gamma- between samples compared). Significant DEGs between the ray mutagenesis from which the mutants understudied were identified. resistant and susceptible entries were then tested using two- This work is supported in part by Swiss Agency for Development and class t test, with G978 as class A and IR64 as class B, using Cooperation (SDC). three replicates. Each t test generated a list of significant and non-significant genes. At each time point, from the DEGs set, the subsets of genes that were significantly differentially References induced (DI genes) by blast inoculation in G978 were determined by the Boolean join. All t tests were done using Anderson JP, Thatcher L, Singh K. Plant defence responses: significance α=0.01 with adjusted Bonferroni correction, conservation between models and crops. Func Plant Biol. 2005;32:21–34. and p values based on 1,000 times permutation, as Arase S, Zhao C-M, Akimitsu K, Yamamoto M, Ichii M. A recessive implemented by TIGR Multiple Experiment Viewer v 3.1, lesion mimic mutant of rice with elevated resistance to fungal available at http://www.tm4.org/. pathogens. J Gen Plant Pathol. 2000;66:109–16. Arumuganathan K, Earle ED. Nuclear DNA content of some important plant species. Plant Mol Biol Rep. 1991;9:229–41. RT-PCR and Southern blot analyses Buschges R, Holricher K, Panstruga R, Simons G, Wolter M, Frijters A, et al. The barley Mlo gene: a novel control element of plant RT-PCR was used for RNA expression analysis of the three pathogen resistance. Cell. 1997;88:695–705. down-regulated genes (AK102429, AK111521, and Campbell MA, Ronald PC. Characterization of four rice mutants with alteration in the defense response pathway. Mol Plant Pathol. AK064188) from microarray result using the gain-of- 2005;6:11–21. resistance mutant and the wild-type IR64 at different time Devoto A, Piffanelli P, Nilsson I, Wallin E, Panstruga R, von Heijne G, et points after infection. A resistant BC F G978 line was 1 3 al. Topology, subcellular localization, and sequence diversity of the grown together with IR64 in the greenhouse. Plants were Mlo family in plants. J Biol Chem. 1999;274:34993–5004. Devoto A, Hartmann HA, Piffanelli P, Elliot C, Simmons C, inoculated with blast isolate Ca89 and with H O as control Taramino G, et al. Molecular phylogeny and evolution of the at 17 das. Leaves were collected at 0, 24, and 48 h after plant-specific seven transmembrane MLO family. J Mol Evol. inoculation, same time points for microarray expression 2003a;56:77–88. Devoto A, Muskett PR, Shirasu K. Role of ubiquitination in the analysis. Total RNA was isolated by TRIZOL Reagent regulation of plant defence against pathogens. Curr Opin Plant (Invitrogen). After DNase treatment by deoxyribonuclease I Biol. 2003b;6:307–11. amplification grade (Invitrogen), first-strand cDNA was Dybbs M, Ngai J, Kaplan JM. Using microarrays to facilitate synthesized with the Superscript™ II Reverse Transcriptase positional cloning:Identification of Tomosyn as an inhibitor of (Invitrogen). TAKARA Ex Taq (TAKARA Bio Inc.) was neurosecretion. PLoS Genet. 2005;1:e2. Faris J, Li WL, Liu DJ, Chen PD, Gill BS. Candidate gene analysis of used for PCR. The PCR condition used for these genes was quantitative disease resistance in wheat. Theor Appl Genet. 95°C for 30 s, 58°C for 30 s, 72°C for 60 s for 32 cycles. 1999;98:219–25. UboxMF-TCCGTTCCACTTGATCACAA, UboxMR- Feys BJ, Parker JE. Interplay of signaling pathways in plant disease GCCGAACCATTGTCACTTTT for AK102429, M10F- resistance. Trends Genet. 2000;16:449–55. Freialdenhoven A, Peterhansel C, Kurth J, Kreuzaler F, Schulze-Lefert CTCAAGCACAGGGAGAGGAG, M10R-CAAGGAGC P. Identification of genes required for the function of non-race- CTGGACAAGAAG for AK111521, and C04(144)F-CAG specific mlo resistance to powdery mildew in barley. Plant Cell. AGGCAGAGGCTTTTGAG, C04(413)R-GGACGGT 1996;8:5–14. CAAATGCAAGAAT for AK064188. Frye CA, Innes RW. An Arabidopsis mutant with enhanced resistance to powdery mildew. The Plant Cell. 1998;10:947–56. Southern blot analysis was used to determine co- Frye CA, Tang D, Innes RW. Negative regulation of defenses response segregation of BB resistance (using PXO87) and candidate in plants by a conserved MAPKK kinase. Proc Nat Acad Sci gene AK102429 in the BC F of G978 and IR64. 2 2 USA. 2001;98:784–6. AK102429 gene probe was defined by the primer pair: Hammond-Kosack KE, Jones JDG. Resistance gene-dependent plant defense responses. The Plant Cell. 1996;8:1773–91. UboxMF-TCCGTTCCACTTGATCACAA, UboxMR- Harushima Y, Yano M, Shomura A, Sato M, Shimano T, Kuboki Y, et GCCGAACCATTGTCACTTTT. DNA from selected sus- al. A high-density rice genetic linkage map with 2275 markers ceptible (n=7) and resistant (n=17) BC F was extracted as 2 2 using a single F2 population. Genetics. 1998;148:479–94. described above. Total DNA (10 µg) was digested by Hildebrand PD, McRae KB, Lu X. Factors affecting flower infection and disease severity of lowbush blueberry by Botrytis cinerea. BamHI (Amersham Biosciences), and the digested DNA Can J Plant Pathol. 2001;23:364–70. was resolved by electrophoresis using 0.9 % agarose gel. Kauffman HE, Reddy APK, Hsiek SPV, Marca SD. An improved DNA was transferred to Hybond N+ nylon membrane under technique for evaluating resistance of race varieties to Xantho- monas oryzae. Plant Dis Rep. 1973;57:537–41. alkaline condition. ECL Direct Nucleic Acid Labeling and 114 Rice (2009) 2:104–114 Keen NT. Gene-for-gene complementarity in plant–pathogen inter- site-leucine-rich repeat type R gene triggers defense responses actions. Annu Rev Genet. 1990;24:444–63. and results in enhanced disease resistance. The Plant Cell. Kosambi DD. The estimation of map distances from recombination 2002;4:3149–62. values. Ann Eugen. 1944;12:172–5. Takahashi A, Kawasaki T, Henmi K, Shii K, Kodama O, Satoh K, et Lorrain S, Vailleau F, Balague C, Roby D. Lesion mimic mutants: al. Lesion mimic mutants of rice with alterations in early keys for deciphering cell death and defense pathways in plants? signaling events of defense. Plant J. 1999;17:535–45. Trends Plant Sci. 2003;8:263–71. Thatcher L, Anderson J, Singh K. Plant defence responses: what we Liu B, Zhang S, Zhu X, Yang Q, Wu S, Mei M, et al. Candidate learnt from Arabidopsis? Functional Plant Biol. 2005;32:1–19. defense genes as predictors of quantitative blast resistance in rice. Vogel J, Somerville S. Isolation and characterization of powdery Mol Plant-Microbe Interact. 2004;17:1146–52. mildew-resistant Arabidopsis mutants. Proc Natl Acad Sci USA. Lu G, Jantasuriyarat C, Zhou B, Wang G-L. Isolation and character- 2000;97:1897–902. ization of novel defense response genes involved in compatible Wang GL, Wu C, Zeng L, He C, Baraoidan M, de Assis Goes da Silva and incompatible interactions between rice and Magnaporthe F, et al. Isolation and characterization of rice mutants compro- grisea. Theor Appl Genet. 2004;108:525–34. mised in Xa21-mediated resistance to X. oryzae pv. oryzae. Theor Martin GB. Functional analysis of resistance genes and their own Appl Genet. 2004;108:379–84. downstream effectors. Curr Opin Plant Biol. 1999;2:273–9. Wen N, Chu Z, Wang S. Three types of defense-responsive genes are Meer JM, Manly KF, Cudmore RH Jr. Map Manager QTXb14. involved in resistance to bacterial blight and fungal blast diseases Roswell Park Cancer Institute. 2002 in rice. Mol Gen Genomics. 2003;269:331–9. Mizobuchi R, Hirabayashi H, Kaji R, Nishizawa Y, Satoh H, Ogawa Wu C, Bordeos A, Madamba MRS, Baraoidan M, Ramos M, Wang T, et al. Differential expression of disease resistance in rice GL, et al. Rice lesion mimic mutants with enhanced resistance to lesion-mimic mutants. Plant Cell Rep. 2002;21:390–6. diseases. Mol Genet Genomics. 2008;279:605–19. Moon J, Parry G, Estelle M. The ubiquitin–proteoasome pathway and Wu J, Wu C, Lei C, Baraoidan M, Bordeos A, Madamba RS, et al. plant development. The Plant Cell. 2004;16:3181–95. Chemical- and irradiation-induced mutants of indica rice IR64 Notteghem JL, Chatel M, Dechanet R. In: Symposium sur la for forward and reverse genetics. Plant Mol Biol. 2005;59:85– resistance du riz a la pyriculariose. Montpellier, France. IRAT- 97. GERDAT 1981. pp 313-330. Yin Z, Chen J, Zeng L, Goh M, Leung H, Khush GS, et al. Pan W. A comparative review of statistical methods for discovering Characterizing rice lesion mimic mutants and identifying a differentially expressed genes in replicated microarray experi- mutant with broad-spectrum resistance to rice blast and bacterial ments. Bioinformatics. 2002;18(4):546–54. blight. Mol Plant-Microbe Interact. 2000;13:869–76. Peterhansel C, Freialdenhoven A, Kurth J, Kolsch R, Schulze Lefert P. Yu I, Fengler K, Clough S, Bent A. Identification of Arabidopsis Interaction analyses of genes required for resistance responses to mutants exhibiting an altered hypersensitive response in gene- powdery mildew in barley reveal distinct pathways leading to for-gene disease resistance. Mol Plant-Microbe Interact. leaf cell death. The Plant Cell. 1997;9:1397–409. 2000;13:277–86. Ramalingam J, Vera Cruz CM, Kukreja K, Chittoor JM, Wu JL, Lee Zeng L, Yin Z, Chen J, Leung H, Wang GL. Fine genetic mapping and SW, et al. Candidate resistance genes from rice, barley, and maize physical delimitation of the lesion mimic gene Spl11 to a 160-kb and their association with qualitative and quantitative resistance DNA segment of the rice genome. Mol Genet Genomics. in rice. Mol Plant-Microbe Interact. 2003;6:14–24. 2002;268:253–61. Roumen EC. Selection for partial resistance in rice to rice blast. Durability Zeng L, Qu S, Bordeos A, Yang C, Baraoidan M, Yan H, et al. Spl11, of Disease Resistance. Norwell: Kluwer; 1993. p. 195–200. a negative regulator of plant cell death and defense, encodes a U- Shirano Y, Kachroo P, Shah J, Klessig D. A gain-of-function mutation Box/ARM repeat protein endowed with E3 ubiquitin ligase in an Arabidopsis Toll Interleukin 1 receptor-nucleotide binding activity. The Plant Cell. 2004;16:2795–808.

Journal

RiceSpringer Journals

Published: Jun 26, 2009

There are no references for this article.