Introduction
Rice is the staple food that feeds half of the world’s population, and plays an important role in global food security. Blast disease of rice (
Oryza sativa L.) caused by the filamentous ascomycetes fungus (
Magnaporthe oryzae, formerly
Magnaporthe grisea (T.T. Hebert) M.E. Barr) (Fig. 1), and sheath blight disease caused by
Rhizoctonia solani Kühn (Fig. 2) have been the two major damaging fungal diseases of rice worldwide (
Savary et al., 2000,
2006;
Khush and Jena, 2007). Before genetic resistance was known, rice farmers knew how to use survived seeds after disease epidemics and knew how to improve cultural practices to reduce crop damage such as adjusting time of planting, rotating crops, avoiding excessive or unbalanced nitrogen application in addition to using survived landrace varieties to prevent blast disease. Current research on blast disease has allowed the identification of several dozens of major blast resistance (
R) genes; each of them confers race specific resistance. Some of them have been molecularly characterized with user friendly molecular markers for breeding. Sheath blight disease was not a major constraint for rice until the deployment of semi-dwarf varieties in intensified high yielding production system because of the relatively tall height of rice plants. Increased deployment of semi-dwarf rice varieties worldwide has made sheath blight one of the most damaging diseases that threaten crop productivity and quality. In contrast to the blast disease, major
R genes have not been identified in cultivated rice for sheath blight disease. However, genes with major and minor effects have been readily found in rice germplasm, and some of them have been tagged with closely linked molecular markers. In this article, current progress on genetic interactions of rice with
M. oryzae and
R. solani will be reviewed.
The O. sativa - M. oryzae interaction
Genetic studies revealed that interaction of
O. sativa with
M. oryzae often follows the classical gene-for-gene specificity (
Flor et al., 1971;
Silue et al., 1992). The
R gene named as
Pyricularia (
Pi) was thought to confer resistance to races of
M. oryzae containing the corresponding avirulence (
AVR) genes.
M. oryzae commences infection immediately after the attachment of a conidium to the surface of rice leaf, and subsequently, germinated conidia produce appresoria for penetration (Fig. 1a). Penetration is believed to occur within 24 h depending on strains (
Wang et al., 2007a).
M. oryzae was known to directly penetrate the cell membranes with a high turgor pressure, and growth of mycelia results in subsequent destruction of a living cell (
Howard et al., 1991). Before a cell is completely destroyed, mycelia were predicted to move to the next cell via unknown mechanisms including the use of plasmodesmata (
Kankanala et al., 2007). Resistance responses mediated by
Pi genes can be seen within 48 h after infection (
Wang et al., 2007a), implying that
Pi genes act at the frontiers of defense responses. It was then predicted that the Pi proteins are receptors for diverse effectors from the fungus. Some of these effectors are likely encoded by the fungal
AVR genes.
The
Pi-ta gene located near the centromere of chromosome 12 encode a predicted cytoplasmic protein with a NBS-LRR domain that recognizes races of
M. oryzae that contain
AVR-Pita (
Bryan et al., 2000;
Jia et al., 2000;
Orbach et al., 2000) (Fig 3).
AVR-Pita is a metalloprotease whose expression in plant is largely unclear. Analysis of natural alleles of the
Pi-ta gene in international rice collection reveals only one resistant allele, and all other alleles were predicted to be susceptible to races of
M. oryzae that contain
AVR-Pita (
Wang et al., 2008). The
Pi-ta allele was widely deployed and was identified in 89 rice germplasm worldwide (
Wang et al., 2007b; Wang and Jia, unpublished data). The
Pi-ta gene in cultivated rice
O.
sativa was predicted to be under strong selection constraint during its domestication (Fig. 4). The frequency of distribution of single nucleotide mutations was examined across the
Pi-ta region (~2 Mb, Fig. 4b) on chromosome 12 with 60 accessions of
O. sativa and 29 accessions of
O. rufipogon. In
O. rufipogon accessions, the significant negative value of Tajima’s
D (
D = - 2.4,
P < 0.01) in the genomic region near the
Pi-ta locus is consistent with recent directional selection (Lee and Jia, unpublished data). Interestingly, the
Pi-ta gene was predicted to directly recognize the products of
AVR-Pita in triggering resistance responses (
Bryan et al., 2000;
Jia et al., 2000). If direct recognition of the fungal effector by an
R gene product is a general mechanism to activate signaling cascades to prevent further spread of blast fungus, the means of cross-kingdom translocation and processing of products of
AVR need to be identified (Fig 3). The finding that chaperones in the endoplasmic reticulum (ER) for both
Pi-ta and
AVR-Pita mediated resistance and virulence activities of the fungus (
Yi et al., 2009) marks an important milestone for the elucidation of molecular mechanisms of
Pi gene-mediated signal recognition. Another interesting feature at the
Pi-ta locus is that an unusual large linkage block was identified in several elite rice cultivars due to a large introgression of the
Pi-ta region into several rice cultivars worldwide. Mechanisms for maintaining a linkage block around the
Pi-ta gene is unknown but could be due to the effects of the centromere or/and the components in
Pi-ta resistance resides on different regions of the chromosome 12 (
Jia and Martin, 2008; Jia, unpublished data). The presence of a large linkage block would benefit the use of molecular markers for marker assisted
Pi-ta introgression. Further investigation of molecular mechanisms of recombinant suppression will benefit genetic improvement and enhancement of rice.
As promising alternatives, cloning of other matched pairs of
R and
AVR genes may facilitate the investigation of
R gene-mediated signaling recognition. To this end, nine
Pi-genes have been molecularly characterized:
Pib (
Wang et al.,1999),
Pi9 (
Qu et al., 2006),
Pi2/Piz-t (
Zhou et al., 2006),
Pi-d2 (
Chen et al., 2006),
Pi36 (
Liu et al., 2007),
Pi37 (
Lin et al., 2007),
Pikm (
Ashikawa et al., 2008),
Pi5 (
Lee et al., 2009) and
Pit (
Hayashi and Yoshida, 2009) (Fig. 5) and some other
Pi genes have been tagged with closely linked markers for isolation (
Ballini et al., 2008). Cloned
Pi genes were predicted to encode proteins with the conserved NBS-LRR domains again implying that plant
R genes possess a common mechanism of signaling recognition or/and transduction. In contrast, cloning and characterization of
AVR genes in
Magnaportheoryzae has lagged behind
R gene cloning due to the relative difficulties of genetic cross. To date, 25
AVR genes in
M. oryzae were described (
Dioh et al., 2000) and six of which were recently cloned:
AVR-Pita,
AVR1-CO39,
PWL1,
PWL2,ACE1 and
AVR-Pizt (
Kang et al., 1995;
Sweigard et al., 1995;
Farman and Leong, 1998;
Orbach et al., 2000;
Bohnert et al., 2004;
Li et al., 2009) (Table 1)
.AVR-Pita encodes a protein with conserved domain indicative of metalloprotease (
Orbach et al., 2000). Transient expression of
AVR-Pita in
Pi-ta containing plants resulted in hypersensitive cell death, indicative of a gene-for-gene resistant reaction (
Bryan et al., 2000;
Jia et al., 2000).
AVR-Pita was not found to express in culture and slight induced expression was observed in susceptible rice plants (Jia, unpublished data). Recent surveys of the
AVR-Pita alleles in
M. oryzae species complex and in field isolates revealed that transposon insertion at the promoter regions, in the coding region, deletion at 5' region are likely related to mechanisms to avoid
Pi-ta recognition and the diversifying selection of
AVR-Pita allele (
Jia et al., 2006; Dai and Jia, unpublished data). Besides
AVR-Pita,
AVR-Co39,
ACE1 and
AVR-Pizt are other three
R gene-specific
AVR genes, isolated from
M. oryzae and all of which have secret signals indicating that their products were secreted out of the fungus (Table 1). Although their modes of actions in plant are still largely unclear, these
AVR genes all seem to be diversified (
Kang et al., 2001;
Zhou et al., 2007;
Khang et al., 2008).
In the meantime, it has been known that other components in plants are also involved in transducing signals for resistance (
Martin et al., 2003). In the
Pi-ta/AVR-Pita interaction, the
Pi-
ta gene was found to require
Ptr(
t) for recognition and signaling transduction.
Ptr(
t) was recently mapped at the
Pi-ta region (
Jia and Martin, 2008) (Fig. 3). Cloning
Ptr(
t) and examining its interaction with both
Pi-ta and
AVR-Pita, and other
Pi and
AVR gene pairs are being intensely pursued. For a short term benefit, “Perfect” markers can be developed from portions of cloned
Pi genes such as the markers for
Pi-ta and
Pi-b (
Jia et al., 2002;
Jia et al., 2003;
Fjellstrom et al., 2004;
Jia et al., 2004). These perfect markers are robust and easy to use for marker-assisted selection. In the southern US, rice cultivars with the
Pi-ta gene are: Katy (
Moldenhauer et al., 1990), Drew (
Moldenhauer et al., 1998), Kaybonnet (
Gravois et al., 1995), Madison (
McClung et al., 1999), Cybonnet (
Gibbons et al., 2006), Spring (
Moldenhauer et al., 2007a), Banks (
Moldenhauer et al., 2007b) and Ahrent (
Moldenhauer et al., 2007c). The perfect markers for
Pi-ta (
Jia et al., 2002;
Jia et al., 2004) were used for the development of some of these cultivars, and were used to verify the
Pi-ta gene in all
Pi-ta containing cultivars.
The O. sativa- R. solani interaction
R. solani belongs to a necrotrophic species complex. Based on anamosis grouping, at least 13 groups infecting different hosts have been identified. The group AG1-IA of
R. solani causes sheath blight in rice (
Wamishe et al., 2007). AG1-IA is one of the largest groups causing the most damages among all other AG groups. Little is known about the pathogenicity and virulence factors of
R. solani. Although the fungal extracts presumably containing the fungal toxin from
R. solani was recently shown to induce expression of distinct genetic components in rice, the nature of toxin remains to be identified (
Brooks, 2007). Thus far, no major
R genes have been found to prevent
R.
solani infection; however, host genes each contribute to different levels of resistance as quantitative
R loci have been mapped onto different chromosomal locations of the rice genome. Several recent accomplishments on the genetic resistance of rice to
R. solani are summarized below.
Standardized rapid disease evaluation
Traditionally, disease reaction to the pathogen is determined by replicated field experiments. Evaluation of disease reactions in the field is often limited by its location, cost and the minimal time needed for evaluation, and often is the bottle neck for genetic studies and germplasm screening. Recently, breeders in Bangladesh were able to evaluate disease reactions using soft drink bottles and this method has been standardized and improved in a number of labs in the US and South America (
Jia et al., 2007) (Fig. 6). In addition, adult plants at early tiller stages were successfully subjected to disease evaluation for mapping quantitative trait loci (QTL) (
Liu et al., 2009).
Candidate genes for sheath blight resistance
Host genes involved in different biochemical pathways have also been identified using DNA microarray and serial analysis of gene expression (
Venu et al., 2007). Individual genes known to contribute minor effects in controlling disease are often referred to as defense genes or candidate genes for disease resistance. Rice germplasm with different levels of resistance and rice genes involved in resistance have been recently verified (
Manosalva et al., 2009). For example, reduction of germin-like protein (OsGLP) reduced resistance to both rice blast and sheath blight diseases (
Manosalva et al., 2009). Some of these differentially and highly expressed genes including the proteins with NBS-LRR domains were identified in some of these QTLs. Using DNA microarray analysis, it was demonstrated that at least 23 rice genes were consistently induced starting from 6 to 10, to 16 h after inoculation (data not shown). Research is underway to utilize these induced genes for developing expression markers to evaluate minor phenotypic responses of rice to the infection by
R. solani.
Mapping sheath blight resistance QTLs (ShB-QTL)
A number of chromosomal locations were identified to associate with different levels of sheath blight resistance (Table 2). These QTLs are located on chromosomes 2, 3, 4, 5, 7, 8, 9, 10, 11 and 12. Some of them were confronted with QTLs conditioning plant height and heading date in rice (
Yano et al., 1997;
Yamamoto et al., 2000). Remarkably, the major ShB-QTL
qSB9-2 at the bottom of chromosome 9 was identified and verified in several laboratories using different phenotyping methods (
Li et al., 1995;
Pinson et al., 2005;
Liu et al., 2009). Identification and verification of
qSB9-2 is one of the most significant advancements in genetic resistance to
R. solani. Two NBS-LRR gene candidates along with dozens of differentially expressed genes were identified at the
qSB9-2 locus. Detecting
qSB9-2 in a mapping population using controlled greenhouse methods makes the fine mapping and cloning of
qSB9-2 more feasible. Near isogenic lines with
qSB9-2 have been developed and more molecular markers are being identified to delimit physical regions harboring
qSB9-2.
Future perspectives
Currently, blast disease has been primarily managed by the use of major
R genes, and sheath blight has been managed by the use of chemical agents with the deployment of tolerant cultivars in integrated cultural practices. In parallel, research on
Arabidopsis and other model crops such as tomato has allowed a better understanding of molecular mechanisms of disease resistance (
Martin et al., 2003), and resulting knowledge has facilitated the development of strategies to manage both rice blast and sheath blight diseases. Recent studies worldwide have led to a promising future for better genetic control of both rice blast and sheath blight diseases. However, the following questions remain unanswered: 1) the total number of
R genes against the blast fungus is unknown in rice germplasm. Rice should have plenty of
R genes or possess elaborate mechanisms to fight against the blast fungus given the fact that the fungus is highly mutable; 2)
AVR genes in
M. oryzae were predicted to play important roles in fitness and pathogenicity and also to trigger
R gene-mediated defense responses; however, cellular targets of
AVR genes of
M. oryzae are largely unknown; 3) whether or not there is (are) a master controller(s) in rice either for blast or sheath blight or for both diseases is (are) also unknown, and if so, why has (have) it (they) not been identified? 4) Underlying mechanisms of ineffectiveness of
R genes to
R. solani are unknown. However, more new sources of resistance to both
M. oryzae and
R. solani have recently been identified from wild rice relatives and they will be used to study the relations of rice with
M. oryzae, or with
R. solani (
Eizenga et al., 2006;
Prasad and Eizenga, 2008;
Eizenga et al., 2009). Eventually the above mentioned challenging questions will be addressed with continuous investigation of genetic resistance to both fungal pathogens worldwide. Resulting knowledge will lead to environmentally benign disease management strategies for ensuring global food security.
Higher Education Press and Springer-Verlag Berlin Heidelberg