Rice blast caused by Magnaporthe oryzae is the most devastating disease of rice and poses a serious threat to world food security. In this study, the distribution and effectiveness of 18 R genes in 277 accessions were investigated based on pathogenicity assays and molecular markers. The results showed that most of the accessions exhibited some degree of resistance (resistance frequency, RF >50%). Accordingly, most of the accessions were observed to harbor two or more R genes, and the number of R genes harbored in accessions was significantly positively correlated with RF. Some R genes were demonstrated to be specifically distributed in the genomes of rice sub-species, such as Pigm, Pi9, Pi5 and Pi1, which were only detected in indica-type accessions, and Pik and Piz, which were just harbored in japonica-type accessions. By analyzing the relationship between R genes and RF using a multiple stepwise regression model, the R genes Pid3, Pi5, Pi9, Pi54, Pigm and Pit were found to show the main effects against M. oryzae in indica-type accessions, while Pita, Pb1, Pik, Pizt and Pia were indicated to exhibit the main effects against M. oryzae in japonica-type accessions. Principal component analysis (PCA) and cluster analysis revealed that combination patterns of major R genes were the main factors determining the resistance of rice varieties to M. oryzae, such as ‘Pi9+Pi54’, ‘Pid3+Pigm’, ‘Pi5+Pid3+Pigm’, ‘Pi5+Pi54+Pid3+Pigm’, ‘Pi5+Pid3’ and ‘Pi5+Pit+Pid3’ in indica-type accessions and ‘Pik+Pib’, ‘Pik+Pita’, ‘Pik+Pb1’, ‘Pizt+Pia’ and ‘Pizt+Pita’ in japonica-type accessions, which were able to confer effective resistance against M. oryzae. The above results provide good theoretical support for the rational utilization of combinations of major R genes in developing rice cultivars with broad-spectrum resistance.
Citation: Wu Y, Xiao N, Yu L, Pan C, Li Y, Zhang X, et al. (2015) Combination Patterns of Major R Genes Determine the Level of Resistance to the M. oryzae in Rice (Oryza sativa L.). PLoS ONE 10(6): e0126130. https://doi.org/10.1371/journal.pone.0126130
Academic Editor: Qian Qian, China National Rice Research Institute, CHINA
Received: January 21, 2015; Accepted: March 29, 2015; Published: June 1, 2015
Copyright: © 2015 Wu et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was supported by the Innovation Fund of Jiangsu Province Agricultural Science and Technology, (CX (12)1003-4), the Supporting Project of Jiangsu Province Science and Technology, (BE2013301), the Special Construction Project of a Modern Agricultural Industry Technology System, (CARS-01-45), and the National Transgenic Major Project, (2011ZX08001-001-006).
Competing interests: The authors have declared that no competing interests exist.
Rice (Oryza sativa L.) is one of the major food crops worldwide, feeding more than half of the world’s population . Rice blast caused by the fungal pathogen Magnaporthe oryzae is the most severe disease of rice worldwide. Due to its widespread distribution and ability to survive in a wide range of environmental conditions, yield losses between 10% and 30%, i.e. destroying rice that could have fed more than 60 million people . Over the last 30 years, there were three major blast epidemics occurred in China during 1982–1985, 1992–1994, and 2001–2005 . The development of resistant cultivars via the introduction of major R genes into elite rice varieties is the most economical and environmentally friendly way to protect crops against this disease . Therefore, identification and characterization of blast resistance genes in germplasm collections is critical for rice improvement.
To date, more than 100 rice blast resistance genes have been identified in rice [4,5], and 20 major R genes (Pi1, Pi2, Pi5, Pi9, Pid3, Pi25, Pi36, Pi37, Pi54, Pia/Pi-Co39, Pib, Pid2, Pikm, Pikp, Pik, Pish, Pit, Pita and Pizt) [6–10] and two partial resistance genes (pi21 and Pb1) [11,12] have been cloned and characterized. All of the cloned R genes (except Pid2 and pi21) belong to the nucleotide-binding site and leucine-rich repeat (NBS–LRR) class of R genes , and some of these R genes such as Pi5, Pia/Pi-Co39, Pikm, Pik and Pikp require two adjacent NBS-LRR class genes for full functionality [7,8,14]. However, Pid2 encodes a B-lectin receptor kinase , while pi21 encodes a proline-rich protein that containing a putative heavy metal-binding domain and putative protein-protein interaction motifs . Although characterization of these R genes has advanced our understanding of the molecular basis of blast resistance in rice, we know little about how these cloned R genes are distributed in modern cultivated rice varieties. Additionally, information both on the resistance effects of R genes in different genetic backgrounds and on which genes are more effective in breeding practice is also missing. Furthermore, there are more than 100 R genes in the rice genome, which means that a single germplasm may harbor several R genes in various combinations, but the fact that resistance reactions show significant differences between germplasms indicates that these variations could be caused by different R gene combinations. Therefore, the question arises of which R genes could be combined to provide a favorable resistance effect across multiple backgrounds. Addressing this question will be helpful in the improvement of blast resistance breeding programs.
Accurate identification of a particular R gene in diverse elite germplasm is the first step for utilization of R genes in rice breeding programs. The conventional methodology to identify allelic variation based on phenotype is limited by large workload and time-consuming, as well as the strong dependence on environmental conditions. However, for R genes that have been isolated, it is now possible to replace phenotypes with molecular markers as the basis for defining alleles . Over the last several decades, many PCR-based tightly linked markers (LMs) have been developed closely associated with a number of R genes, such as Pi1 , Pi2 , Pi5 , Pi9 . These LMs offer an efficient and rapid way to select for the presence of target R genes in gene introgression and gene pyramiding. Moreover, the identification of functional markers (FMs), which include single nucleotide polymorphisms (SNPs) and insertion/deletions (InDels), were derived from polymorphic sites within genes causally involved in phenotypic trait variation and is particularly useful in several genetic backgrounds . Some FMs specific to R genes, such as Pita , Pib , Pit  and Pid3  have been developed, which will provide convenient ways to identify target genes.
In the present study, the distribution of 18 R genes was analyzed in 277 accessions of Chinese elite rice parental lines and the donors of R genes using FMs and LMs, and the resistance frequency (RF) in the 277 accessions was evaluated by a total of 76 isolates. The R genes showing the main effects to M. oryzae in indica and japonica rice genome were identified through multiple stepwise regression analysis. The combination patterns of major R genes in the indica-type and japonica-type rice germplasm were ascertained based on principal component analysis (PCA) and cluster analysis of various gene combination factors. The results of this study are expected to guide the deployment and combination of R genes and provide useful information to improve blast resistance.
Materials and Methods
A total of 277 accessions (128 japonica and 149 indica) mainly consisting of Chinese elite rice parental lines, were analyzed in the present study (S1 Table). Most of the accessions used in this study were categorized as showing good agronomic traits, eating quality, yield traits, grain quality and blast resistance and have been widely used as parental lines to develop new varieties in recent years. To perform a broad-scale investigation of R gene-based resistance, three sets of differential lines and the R gene donors were used as standard check varieties. The first of these sets was composed of Japanese differentials [25, 26], including Fukunishiki (Piz), Toride 1 (Pizt), K59 (Pit), Kanto 51 (Pik), Tsuyuake (Pikm), K60 (Pikp), BL 1 (Pib), Yashiromochi (Pita), K3 (Pi54), Shin 2 (Piks) and Pi No.4 (Pita2). The second set consisted of Chinese differentials, including Tetep (Pi1, Pi5, Pi54), Zhenglong 13 (unknown), Sifeng 43 (Pia, Pib), Dongnong 363 (Pia, Pik), Kanto 51 (Pik), Hejiang 18 (Pia) and Lijangxituanheigu (LTH); the third set consisted of IRRI differentials, including six resistant near isogenic lines (NILs) in the genetic background of Co39 . For the Pi9, Pi40, Pigm, Pid2 and Pid3 genes, the donor lines 75-1-127 , IR65482-4-136-2-2 , Gumei 4  and Digu  were employed in this study.
Blast isolates, inoculation and disease evaluation
To evaluate blast resistance, a total of 76 isolates collected from Sichuan (SC), Anhui (AH), Jiangsu (JS), Hainan (HN), Hubei (HB), and Zhejiang (ZJ) provinces in China from 2010–2012 were examined in this study (S2 Table). The plants used for inoculation were grown in 60 cm×30 cm×4 cm plastic trays with sieved garden soil. Each tray contained 42 experimental materials and two highly susceptible cultivars, Co39 and LTH (susceptible control). A total of 7 trays were used for each blast isolate inoculation experiment, containing 277 lines each with at least 10 plants. Three-week-old rice seedlings were placed in inoculation chambers and inoculated by spraying with 40 ml of conidial suspensions (5×104 conidia/ml) with 0.02% Tween-20 as described by Huang et al. . The inoculated seedlings were kept in dark chambers at 26°C with 95–100% relative humidity for 24 h and then transferred to the greenhouse where they were grown under a 12 h light/12 h dark cycle and 90% relative humidity. The inoculated seedlings were maintained under these conditions for 7 days to allow disease development. A randomized complete block design with two replicates was applied to the entire inoculation experiment.
Disease reactions were assessed 7 days after inoculation, and lesions on the rice leaves were evaluated on a scale of 0 to 5 according to standard procedures , where 0 indicated no visible lesion; 1 indicated only small point lesions appearing at the sites of infections; 2 indicated lesion sizes smaller than 2 mm and no visible fungal mass; 3 indicated that lesions were greater than 2 mm occupying 10% or less of the leaf area; 4 indicates that lesions were greater than 3 mm occupying more than 10% and less than 50% of the leaf area; and 5 indicates that lesions were greater than 50% of the leaf area. Plants with scores of 0–2 were classified as resistant (R) and those with scores of 3–5 were classified as susceptible (S) . The RF of the accessions in relation to the isolates of M. oryzae could be calculated according to the following formula: RF = (the number of incompatible isolates / the total number of isolates used) *100%. Each RF value was assessed from two technical replicates.
DNA markers and genotyping of 277 rice germplasms
A total of 15 FMs and 9 LMs were selected from published primer sequences to detect 18 R genes, including six alleles of the Piz locus (Piz, Pizt, Pi2, Pi9, Pi40 and Pigm) [35, 36], two alleles of the Pik locus (Pik and Pi1) [10, 37] and ten independent R genes (Pia, Pib, Pit, Pish, Pi5, Pi54, Pita, Pid2, Pid3, and Pb1)  in the 277 accessions. Detailed information on the markers is provided in Table 1.
PCR amplification was conducted following a standard protocol , including pre-denaturation for 5 min at 94°C, followed by 35 cycles of 45 s at 94°C, 45 s at the annealing temperature indicated in Table 1 and 1 min at 72°C, followed by a final extension at 72°C for 10 min. For dCAPS marker analysis, the PCR products were digested following the manufacturer’s instructions. DNA bands were visualized on an 8% denaturing polyacrylamide gel or a 4% agarose gel according to their relative fragment sizes. The reactions that amplified the same bands as the target gene donors were labeled with a “+” sign, and the others were labeled with a “-” sign.
The relationship between R genes and RF was analyzed using a multiple stepwise regression model, which is an available option in Matlab (V.7.0) software, according to Xu et al. . R genes that were significantly correlated with RF (P≤0.01) were subjected to various R gene combination factors according to factor analysis (principal component analysis, PCA) in SPSS (V.21) software. Based on various R gene combination factors, the germplasms were clustered into several groups. Each group represents one class of R gene combination patterns found among the germplasms. Significant differences in the RF between population groups were analyzed by one-way analysis of variance (ANOVA) 
Genotypic assays for 18 R genes in the rice germplasms
Ten independent R genes, including Pia, Pib, Pit, Pish, Pi5, Pi54, Pita, Pid2, Pid3, and Pb1, were easily identified based on their FMs or LMs (S1 Fig). However, the alleles of complex gene loci were difficult to distinguish from each other because of their sequence similarity. To distinguish the six alleles of the Piz locus (Pi2, Pi9, Pizt, Piz, Pigm and Pi40) from each other, several markers were selected from published primer sequences. An InDel polymorphism specific for Pi9 donor 75-1-127 was obtained using the marker F9, which discriminates Pi9 from other alleles (Fig 1A). The Pi9-2 is known to detect Pi9- and Piz-containing genotypes in a previous study , indicating its suitability as a Pi9- and Piz-specific marker. Therefore, the combination of F9 and Pi9-2 can distinguish Pi9 and Piz from other alleles. The dCAPS marker R2123 revealed a polymorphism between the Pizt donor Toride 1 and the other tested genotypes , thus behaving as a Pizt-specific marker in this study. In addition, a Pi2-selective marker , AP22, detected a polymorphism between the Pi2/Pizt donors and the other tested genotypes. Thus, these two markers are able to differentiate Pi2 and Pizt from other alleles (Fig 1B). To assess Pigm in the Chinese elite rice parental lines, two co-segregating markers S29742 and ZJ58.7 were used in the present study [30, 39], but the detection results showed that they could not distinguish Pigm from Pizt; hence, the InDel marker M-Pi2 was introduced to detect Pigm, together with S29742 and ZJ58.7. The results revealed that the combined use of these three markers could distinguish Pigm from other alleles (Fig 1C). Similar to Pigm, the dCAPS marker Pi40 was not able to discriminate between Pi40 and the other alleles; therefore, the marker F9 was used to detect Pi40 together with the Pi40 marker (Fig 1D). Another multiple gene complex loci is the Pik locus on chromosome 11, which consisted of Pi1, Pik, Pikm, Pikp and Piks alleles. In this study, only Pi1 and Pik were detected for breeding practice purpose. The dCAPS marker 2953 was verified to be a specific marker for Pik  and was further confirmed to distinguish Pik from various other Pik alleles in this study (Fig 1E). The marker RM224 has been reported to be linked to Pi1  and was further demonstrated to be a Pi1-specific marker in this study (Fig 1F).
The number of R genes present in the accessions is positively correlated with the RF against M. oryzae in the Chinese elite rice parental lines
According to the disease reactions to 7 Chinese differential rice cultivars, the 76 blast isolates obtained from rice blast specimens collected from 6 provinces were allocated to 8 groups and 24 races (Fig 2A), These results indicated that the isolates were highly diverse, with resistance/susceptibility being observed among the differential rice cultivars (Fig 2B). Therefore, the isolates could be used as differential strains in this study, with the exception of 176-1-1 (ZH), which showed avirulence on all of the differential rice cultivars (S3 Table). The inoculation results revealed that the accessions exhibited significant difierences in their reaction patterns and degrees of susceptibility to the isolates. We further calculated the RF of each accession and found that the values exhibited a skewed distribution that mostly ranged from 50–100% (Fig 2C), implying that most of the accessions showed some degree of resistance to M. oryzae.
A. The distribution of isolates in 6 typical provinces. B. The disease reactions of a typical blast isolate to 7 Chinese differential rice cultivars, 1–7 represent Tetep, Zhenlong 13, Sifeng 43, Dongnong 363, Kano51, Hejiang 18 and LTH, respectively. C. The distribution of RF values among accessions tested. D. The number of R genes present in 277 rice accessions. E. The relationship between the number of R genes and RF values.
Most of the accessions harbored more than one of the 18 detected R genes, except for three indica-type accessions (S1 Table). The number of R genes found in the accessions was normally distributed, mostly ranging from 2–5 genes (Fig 2D). Compared with the results of the pathogenicity assays, we observed that the number of R genes present in the accessions was significantly positively correlated with the RF (Fig 2E; R2 = 0.122, P<0.01), which means that the greater the number of R genes found in the accessions, the higher the RF against M. oryzae.
Various R genes show the main effects against M. oryzae in the indica and japonica rice genomes
A total of 18 R genes were detected in 277 accessions using a set of 24 FMs and LMs, and the results demonstrated that most of these R genes were present in both the indica and japonica sub-species genomes. With the exception of the R genes Pi2, Pi40, Pita, Pib and Pi54 which were evenly distributed in the indica-japonica accessions, the other R genes were found to show a disequilibrium distribution between the indica and japonica sub-species genomes. For example, the R genes Pigm, Pi9, Pi5 and Pi1 were only found in indica-type accessions, and Pik and Piz were just present in japonica-type accessions. The distribution frequency of the R genes Pit, Pid2, Pid3 and Pia was higher in indica-type accessions than in japonica-type accessions, but the distribution frequency of the R genes Pizt, Pb1 and Pish was higher in japonica-type accessions than in indica-type accessions (Fig 3 and S4 Table). This distribution specificity of the R genes implies that the R genes conferring resistance to M. oryzae may differ between the indica and japonica sub-species genomes.
By analyzing the relationship between the R genes and RF using a multiple stepwise regression model, the R genes Pid3, Pi5, Pi9, Pi54, Pigm and Pit were found to show the main effects against M. oryzae in indica-type accessions, together accounting for 44.8% (R2 = 44.8%, P<0.01) of the observed phenotypic variance, with partial correlation coefficients of 0.273, 0.271, 0.194, 0.190, 0.171 and 0.166, respectively (Table 2). In contrast to the results for the indica-type accessions, the R genes Pita, Pb1, Pik, Pizt and Pia were indicated to show the main effects against M. oryzae in japonica-type accessions, which explained 36.2% (R2 = 36.2%, P<0.01) of the observed phenotypic variance overall, with partial correlation coefficients of 0.468, 0.384, 0.268, 0.243 and 0.197, respectively (Table 2). In addition, we found that many of the R genes were associated with a province-specific resistance in the various rice-growing regions of China, such as Pi5 was detected to explain 21.0% phenotypic variances observed relative to isolates originating from HN Provence in indica-type accessions (S5 Table). The above results imply that certain R genes showed the main effects of the resistance to M. oryzae, and the resistance and susceptibility of the accessions, determined by their R genes, varied between the indica and japonica sub-species backgrounds.
R gene combination patterns are the main factors determining the resistance of rice varieties to M. oryzae
The genes that showed the main effects against M. oryzae in indica-type accessions were classified into three types of R gene combination factors: ‘Pid3, Pi5, Pit, Pi54’ (R2 = 25.27%), ‘Pid3, Pigm’ (R2 = 20.54%) ‘Pi9’ (R2 = 16.82%) and, according to the principal component analysis results (PCA, KMO = 0.65, P = 0.01), and the 149 indica accessions were further categorized into 6 groups based on cluster analysis with R gene combination factors (Fig 4A). Cluster I did not include any of these R gene combination factors and presented an RF of 64.84% (Fig 4B). Clusters II, III and IV mainly contained R gene combination factors (Pid3, Pi5, Pit and Pi54) and showed RF values of 80.03%, 89.83% and 77.15% respectively. Cluster V was mainly composed of R gene combination factors (Pid3 and Pigm) and exhibited RF values of 95.16%. Cluster VI included the R gene combination factor Pi9 and showed an RF of 90.51%. The RF values of clusters II, III, IV, Vand VI were significantly higher than that of cluster I, suggesting that major R genes could significantly improve the resistance of the rice germplasm to M. oryzae (P<0.01). Furthermore, the results showed that there were significant differences in the level of resistance between clusters that contained various R gene combination factors. For example, cluster II mainly contained R gene combination factors (Pid3, Pi5, Pit and Pi54) and could be combined into 5 gene combination patterns (Pi54, Pi54+Pid3, Pi5+Pi54, Pi5+Pi54+Pid3 and Pi5+Pit+Pi54). However, the RF of cluster II was significantly lower than that of cluster VI, which contained the R gene factor ‘Pi9’ with the combination patterns ‘Pi9+Pi54’ and ‘Pi9’, indicating that the R gene combination pattern may be the main factor determining rice variety resistance to M. oryzae in indica-background accessions.
A. R gene combination factors in indica-type accessions. B. Gene clusters in indica-type accessions. C. R gene combination factors in japonica-type accessions. D. Gene clusters in japonica-type accessions.
Similarly, three R gene combination factors, ‘Pizt, Pia, Pita’(R2 = 28.87%), ‘Pik’ (R2 = 21.61%) and ‘Pizt, Pia, Pb1’ (R2 = 21.11%), were categorized in japonica-type accessions among the R genes that showed the main effects against M. oryzae (KMO = 0.73, P = 0.01), and the 128 japonica-type accessions were further classified into six major groups, clusters I, II, III, IV, V and VI, based on the observed R gene combination factors (Fig 4C). Significant differences in the reaction patterns against M. oryzae were also observed among these cluster groups (Fig 4D). Cluster I did not include any of the above three R gene combination factors and presented an RF of only 53.93%. In contrast, clusters III and IV comprised R gene combination factors (Pizt, Pia and Pita) and showed RF values of 81.37% and 84.02%, respectively. Cluster V mainly included the R gene combination factor Pik, and its RF was 86.88%. Clusters II and VI mainly included R gene combination factors (Pizt, Pia and Pb1) and displayed RF values of 73.23% and 87.57%, respectively. The RF values of clusters III, IV, V and VI were significantly increased compared with that of cluster I (P<0.01). The resistance of cluster IV, carrying the ‘Pizt+Pia’ gene combination, and cluster VI, pyramiding ‘Pizt+Pita,’ was better than that of cluster II, carrying ‘Pia+Pb1,’ implying that various major R gene combination patterns resulted in different RFs in japonica-background accessions.
A large number of R genes have been identified and mapped in a variety of rice genotypes in previous works. Unfortunately, not all R genes are effective against pathogenic isolates of M. oryzae. Therefore, how to employ a few of these R genes to achieve broad-spectrum and durable resistance must be considered during rice blast resistance breeding.
According to the detection results of 18 R genes in 277 germplasms, we found that the distribution of R genes was significantly different between the indica and japonica sub-species genomes. For example, the R genes Pit, Pid3, Pid2 and Pia were mainly detected in indica-type accessions, and Pigm, Pi9, Pi5 and Pi1 were only present in indica-type accessions. However, the R genes Pizt, Pb1 and Pish were mainly identified in japonica-type accessions, and Pik and Piz were restricted to japonica-type accessions. These results were possible because these R genes have experienced significant differentiation due to reproductive isolation and decreased interflow between indica-japonica sub-species during the evolutionary process. Thus, it may be an effective strategy to transfer these R genes within subspecies using marker-assisted selection (MAS) to create new resistant germplasms, for example, transferring Pid3, Pigm, Pi1 and Pi5, which originated in the indica genome, to the japonica rice genome.
The specific distribution of R genes also results in various R genes showing the main effects against M. oryzae between indica and japonica germplasms, such as the R genes Pid3, Pi5, Pi9, Pi54, Pigm and Pit in indica-type accessions and Pita, Pb1, Pik, Pizt and Pia in japonica-type accessions (Table 2). Additionally, we found that particular R genes were effective against isolates originating from particular regions of China, and these genes may therefore be specifically and effectively used in the target regions, such as Pi2 in ZJ and HN Provinces and Pib in JS Province (S5 Table). Moreover, we noted that some R genes were distributed evenly between the indica and japonica genomes, such as Pita and Pi54. In contrast, Pi54 was only found to show a main effect in indica backgrounds, and Pita only played a major role in japonica backgrounds (Table 2). These results indicated that some R genes like Pita and Pi54 may be exhibiting a great interaction effect with the indica-japonica genome background. Therefore, the genetic background must be considered when using these R genes for breeding cultivars with broad-spectrum resistance.
It is thought to be a useful strategy to pyramid several R genes for breeding cultivars with broad-spectrum and durable resistance. However, random pyramiding of various R genes may not show an additive effect, as indicated by the combination of the Piz5 and Pita genes in Co39 . He et al.  also reported that the resistance effect of pyramiding lines with Pi4 and Pi1 was weaker than that in lines with the single genes in the Co39 background. Therefore, the question arises regarding which combination patterns of R genes are the most effective for achieving blast resistance.
In the present study, we demonstrated that R gene combination factor Pi9 with combination patterns ‘Pi9+Pi54’ and ‘Pi9’, and the R gene combination factors ‘Pid3, Pigm’ with combination patterns ‘Pid3+Pigm’, ‘Pi5+Pid3+Pigm’, ‘Pi5+Pi54+Pid3+Pigm’, ‘Pi5+Pid3’ and ‘Pi5+Pit+Pid3’, could confer effective resistance to M. oryzae in indica-type accessions (Fig 4B). In contrast, in japonica-type accessions, R gene combination factor Pik with ‘Pik+Pib’, ‘Pik+Pita’, ‘Pik+Pb1’ combination patterns, R gene combination factors ‘Pizt, Pia, Pita’ with ‘Pizt+Pia’, ‘Pizt+Pita’ combination patterns, could confer effective resistance to M. oryzae (Fig 4D). These results indicated that combination patterns of major R genes might be the main factor determining the resistance of rice varieties to M. oryzae. Therefore, rational utilization of these gene combinations is essential for the development of elite rice cultivars with broad-spectrum resistance to the blast pathogen in target regions (S6 and S7 Tables).
Most of above-mentioned effective R gene combination patterns contained alleles of the Piz locus (Pi9, Pigm and Pizt). It has been shown that pyramiding lines with ‘Pi1+Pi2’ genes shows complete resistance to the most virulent races of the pathogen originating from southern China . Furthermore, Yu et al.  demonstrated that the single R gene, Pigm and Pi1, possessed a complementary effect with respect to the blast isolates and the RF of pyramiding lines with ‘Pigm+Pi1’ to blast isolates originated from China was more than 90%. In addition, it has been documented that at least nine R genes Pi1, Pi2, Pi5, Piz, Pi9, Pi40, Pizt, Pi33 and Pigm appear to confer broad-spectrum resistance to a number of isolates or races from one or more countries [20,34,47], and six of which are alleles of the Piz locus (Pi2, Piz, Pi9, Pi40, Pizt and Pigm). All of these results indicate that the alleles of the Piz locus play a key role in conferring resistance to M. oryzae. Therefore, it may be an effective strategy to use different alleles of Piz locus as the backbone for pyramiding with other R genes with strong complementary effect and broad-spectrum resistance, such as Pi1, Pi5, Pi33, Pi54 and Pid3, could create new resistant germplasms and further enhance of the level and broad-spectrum of blast resistance in rice.
S1 Fig. Molecular screening of molecular markers for 10 independent R genes.
S1 Table. Information of each of the 277 rice accessions, and their haplotype with respect to 18 R genes isolated.
S2 Table. Information of blast populations for pathogenicity assays.
S3 Table. Resistance reaction of 32 standard check varieties to 76 isolates of M. oryzae.
S4 Table. Distribution of 18 rice blast resistance genes in indica and japonica genome.
S5 Table. Effect of rice blast resistance genes against M. oryzae originated from six provinces, China.
S6 Table. RF values of clusters in indica-type accessions in relation to M. oryzae originating from six provinces, China.
The authors thank Prof. Shiliang Gu (Yangzhou University) for his valuable advice and discussions on the statistical data analysis.
Conceived and designed the experiments: YYW XBP AHL. Performed the experiments: YYW LY YHL. Analyzed the data: XXZ NX CHP. Contributed reagents/materials/analysis tools: GQL ZYD. Wrote the paper: YYW NX AHL.
- 1. Khush GS. What it will take to feed 5.0 billion rice consumers in 2030. Plant Mol Biol. 2005;59: 1–6. pmid:16217597
- 2. Skamnioti P, Gurr SJ. Against the grain: safeguarding rice from rice blast disease. Trends Biotechnol. 2009;27: 141–150. pmid:19187990
- 3. Khush GS, Jena K. Current status and future prospects for research on blast resistance in rice (Oryza sativa L.). In: Advances in genetics, genomics and control of Rice blast disease. Dordrencht: Springer Verlag; 2009. pp 1–10.
- 4. Ballini E, Morel JB, Droc G, Price A, Courtois B, Notteghem JL, et al. A genome-wide meta-analysis of rice blast resistance genes and quantitative trait loci provides new insights into partial and complete resistance. Mol Plant Microbe Interact. 2008;21: 859–868. pmid:18533827
- 5. Sharma TR, Rai AK, Gupta SK, Vijayan J, Devanna BN, Ray S. Rice blast management through host-plant resistance: retrospect and prospects. Agric Res. 2012;1: 37–52.
- 6. Liu J, Wang X, Mitchell T, Hu Y, Liu X, Dai L, et al. Recent progress and understanding of the molecular mechanisms of the rice-Magnaporthe oryzae interaction. Mol Plant Pathol. 2010;11: 419–427. pmid:20447289
- 7. Cesari S, Thilliez G, Ribot C, Chalvon V, Michel C, Jauneau A, et al. The rice resistance protein pair RGA4/RGA5 recognizes the Magnaporthe oryzae effectors AVR-Pia and AVR1-CO39 by direct binding. Plant Cell. 2013;25: 1463–1481. pmid:23548743
- 8. Okuyama Y, Kanzaki H, Abe A, Yoshida K, Tamiru M, Saitoh H, et al. A multifaceted genomics approach allows the isolation of the rice Pia-blast resistance gene consisting of two adjacent NBS-LRR protein genes. Plant J. 2011;66: 467–479. pmid:21251109
- 9. Rai AK, Kumar SP, Gupta SK, Gautam N, Singh NK, Sharma TR. Functional complementation of rice blast resistance gene Pi-kh (Pi54) conferring resistance to diverse strains of Magnaporthe oryzae. J Plant Biochem Biotechnol. 2011;20: 55–65.
- 10. Hua L, Wu J, Chen C, Wu W, He X, Liu J, et al. The isolation of Pi1, an allele at the Pik locus which confers broad spectrum resistance to rice blast. Theor Appl Genet. 2012;125: 1047–1055. pmid:22643901
- 11. Fukuoka S, Saka N, Koga H, Ono K, Shimizu T, Ebana K, et al. Loss of function of a proline-containing protein confers durable disease resistance in rice. Science. 2009;325: 998–1001. pmid:19696351
- 12. Hayashi N, Inoue H, Kato T, Funao T, Shirota M, Shimizu T, et al. Durable panicle blast-resistance gene Pb1 encodes an atypical CC-NBS-LRR protein and was generated by acquiring a promoter through local genome duplication. Plant J. 2010;64: 498–510. pmid:20807214
- 13. van der Biezen EA, Jones JD. The NB-ARC domain: a novel signalling motif shared by plant resistance gene products and regulators of cell death in animals. Curr Biol. 1998;8: R226–R227. pmid:9545207
- 14. Lee SK, Song MY, Seo YS, Kim HK, Ko S, Cao PJ, et al. Rice Pi5-mediated resistance to Magnaporthe oryzae requires the presence of two coiled-coil-nucleotide-binding-leucine-rich repeat genes. Genetics. 2009;181: 1627–1638. pmid:19153255
- 15. Chen X, Shang J, Chen D, Lei C, Zou Y, Zhai W, et al. A B-lectin receptor kinase gene conferring rice blast resistance. Plant J. 2006;46: 794–804. pmid:16709195
- 16. Jiang H, Feng Y, Bao L, Li X, Gao G, Zhang Q, et al. Improving blast resistance of Jin 23B and its hybrid rice by marker-assisted gene pyramiding. Mol Breeding. 2012;30: 1679–1688.
- 17. Gouda PK, Saikumar S, Varma CMK, Nagesh K, Thippeswamy S, Shenoy V, et al. Marker-assisted breeding of pi-1 and Piz-5 genes imparting resistance to rice blast in PRR78, restorer line of Pusa RH-10 basmati rice hybrid. Plant Breeding. 2013;132: 61–69.
- 18. Cho YC, Kwon SW, Choi IS, Lee SK, Jeon JS, Oh MK, et al. Identification of major blast resistance genes in Korean rice varieties (Oryza sativa L.) using molecular markers. J Crop Sci Biotech. 2007;10: 265–276.
- 19. Tacconi G, Baldassarre V, Lanzanova C, Faivre-Rampant O, Cavigiolo S, Urso S, et al. Polymorphism analysis of genomic regions associated with broad-spectrum effective blast resistance genes for marker development in rice. Mol Breeding. 2010;26: 595–617.
- 20. Ingvardsen CR, Schejbel B, Lübberstedt T. Functional markers in resistance breeding. Prog Bot. 2008: 61–87.
- 21. Jia Y, Redus M, Wang Z, Rutger JN. Development of a SNLP marker from the pi-ta blast resistance gene by tri-primer PCR. Euphytica. 2004;138: 97–105.
- 22. Fjellstrom R, Conaway-Bormans CA, McClung AM, Marchetti MA, Shank AR, Park WD. Development of DNA markers suitable for marker assisted selection of three genes conferring resistance to multiple pathotypes. Crop Sci. 2004;44: 1790–1798.
- 23. Hayashi K, Yasuda N, Fujita Y, Koizumi S, Yoshida H. Identification of the blast resistance gene pit in rice cultivars using functional markers. Theor Appl Genet. 2010;121: 1357–1367. pmid:20589366
- 24. Wang H, Chen J, Shi Y, Pan G, Shen H, Wu J. Development and validation of CAPS markers for marker-assisted selection of rice blast resistance gene Pi25. Acta Agron Sin. 2013;38: 1960–1968.
- 25. Yamada M, Kiyosawa S, Yamaguchi T, Hirano T, Kobayashi T, Kushibuchi K, et al. Proposal of a new method for differentiating races of Pyricularia oryzae Cavara in Japan. Jpn J Phytopathol. 1976;42: 216–219.
- 26. Kiyosawa S. Establishment of differential varieties for pathogenicity test of rice blast fungus. Rice Genet Newsl 1984;1: 95–97.
- 27. Inukai T, Nelson R, Zeigler R, Sarkarung S, Mackill D. Allelism of blast resistance genes in near-isogenic lines of rice. Phytopathology. 1994;84: 1278–1283.
- 28. Qu S, Liu G, Zhou B, Bellizzi M, Zeng L, Dai L, et al. The broad-spectrum blast resistance gene Pi9 encodes a nucleotide-binding site-leucine-rich repeat protein and is a member of a multigene family in rice. Genetics. 2006;172: 1901–1914. pmid:16387888
- 29. Jeung JU, Kim BR, Cho YC, Han SS, Moon HP, Lee YT, et al. A novel gene, Pi40 (t), linked to the DNA markers derived from NBS-LRR motifs confers broad spectrum of blast resistance in rice. Theor Appl Genet. 2007;115: 1163–1177. pmid:17909744
- 30. Deng Y, Zhu X, Shen Y, He Z. Genetic characterization and fine mapping of the blast resistance locus Pigm(t) tightly linked to Pi2 and Pi9 in a broad-spectrum resistant Chinese variety. Theor Appl Genet. 2006;113: 705–713. pmid:16832648
- 31. Chen XW, Li SG, Xu JC, Zhai WX, Ling ZZ, Ma B, et al. Identification of two blast resistance genes in a rice variety, Digu. J Phytopathol. 2004;152: 77–85.
- 32. Huang H, Huang L, Feng G, Wang S, Wang Y, Liu J, et al. Molecular mapping of the new blast resistance genes Pi47 and Pi48 in the durably resistant local rice cultivar Xiangzi 3150. Phytopathology. 2011;101: 620–626. pmid:21171885
- 33. Jia Y, Liu G. Mapping quantitative trait Loci for resistance to rice blast. Phytopathology. 2011;101: 176–181. pmid:20879845
- 34. Wu Y, Bao Y, Xie L, Su Y, Chu R, He W, et al. Fine mapping and identification of blast resistance gene Pi-hk1 in a broad-spectrum resistant Japonica rice landrace. Phytopathology. 2013;103: 1162–1168. pmid:23718837
- 35. Deng Y, Zhu X, Xu J, Chen H, He Z. Map-based cloning and breeding application of a broad-spectrum resistance gene Pigm to rice blast. In: Advances in genetics, genomics and control of Rice blast disease. Springer Verlag; 2009. pp. 161–171.
- 36. Zhou B, Qu S, Liu G, Dolan M, Sakai H, Lu G, et al. The eight amino-acid differences within three leucine-rich repeats between Pi2 and Piz-t resistance proteins determine the resistance specificity to Magnaporthe grisea. Mol Plant Microbe Interact. 2006;19: 1216–1228. pmid:17073304
- 37. Zhai C, Lin F, Dong Z, He X, Yuan B, Zeng X, et al. The isolation and characterization of Pik, a rice blast resistance gene which emerged after rice domestication. New Phytol. 2011;189: 321–334. pmid:21118257
- 38. Koide Y, Kawasaki A, Telebanco-Yanoria MJ, Hairmansis A, Nguyet N, Bigirimana J, et al. Development of pyramided lines with two resistance genes, Pish and Pib, for blast disease (Magnaporthe oryzae B Couch) in rice (Oryza sativa L.). Plant Breeding. 2010;129: 670–675.
- 39. Yu M, Dai Z, Pan C, Chen X, Yu L, Zhang X, et al. Resistance spectrum difference between two broad-spectrum blast resistance genes, Pigm and Pi2, and their interaction effect on Pi1. Acta Agron Sin. 2013;39: 1927–1934.
- 40. Feng H, Yang C, Wu X, Liu Y, Peng Y. Detection and analysis of rice blast resistance gene in some hybridrice parents and 32 resistant materials of Sichuan province. Southwest China. J Agric Sci. 2013;26: 987–993.
- 41. Wang J, Yang J, Zhu J, Fang F, Li W, Wang F, et al. Development of functional marker for rice blast resistance gene Pikh and natural variation at Pikh locus in Japonica rice in Jiangsu province, China. Chin J Rice Sci. 2014;28: 141–147.
- 42. Chen X, Temnykh S, Xu Y, Cho YG, McCouch SR. Development of a microsatellite framework map providing genome-wide coverage in rice (Oryza sativa L.). Theor Appl Genet. 1997;95: 553–567.
- 43. Xu C, Ladouceur M, Dastani Z, Richards JB, Ciampi A, Greenwood CM. Multiple regression methods show great potential for rare variant association tests. PLOS ONE. 2012;7: e41694. pmid:22916111
- 44. Mertler CA, Vannatta RA. Advanced and multivariate statistical methods. Los Angeles, CA: Pyrczak; 2002.
- 45. Hittalmani S, Parco A, Mew TV, Zeigler RS, Huang N. Fine mapping and DNA marker-assisted pyramiding of the three major genes for blast resistance in rice. Theor Appl Genet. 2000;100: 1121–1128.
- 46. He Q, Tang H, Hei L, Robert S, Zeigler R. Identification of CO39 near-isogenic lines for rice blast. Acta Agron Sin. 2001;27: 838–841.
- 47. Jeon JS, Chen D, Yi GH, Wang GL, Ronald PC. Genetic and physical mapping of Pi5(t), a locus associated with broad-spectrum resistance to rice blast. Mol Genet Genomics. 2003;269: 280–289. pmid:12756540