Plant innate immunity in rice: a defense against pathogen infection

A large number of pathogenic microorganisms cause rice diseases that lead to enormous yield losses worldwide. Such losses are important because rice is a staple food for more than half of the world’s population. Over the past two decades, the extensive study of the molecular interactions between rice and the fungal pathogen Magnaporthe oryzae and between rice and the bacterial pathogen Xanthomonas oryzae pv . oryzae has made rice a model for investigating plant–microbe interactions of monocotyledons. Impressive progress has been recently achieved in understanding the molecular basis of rice pathogen-associated molecular pattern-immunity and effector-triggered immunity. Here, we briefly summarize these recent advances, emphasizing the diverse functions of the structurally conserved fungal effectors, the regulatory mechanisms of the immune receptor complexes, and the novel strategies for breeding disease resistance. We also discuss future research challenges.


INTRODUCTION
Many microbial pathogens attack crop plants and cause huge yield losses that threaten global food security. Although application of chemicals has significantly reduced plant diseases, planting of resistant cultivars remains the most effective and environmental-friendly strategy to control crop diseases. Rice (Oryza sativa) is an important crop that is grown in Asia, Africa, and South and Central America. Over half of the global population consumes rice as the main food source. Throughout the growing season, a variety of pathogens, including fungi, bacteria, viruses, and nematodes, infect different parts of rice plants and greatly reduce yields. In the last two decades, considerable knowledge has been obtained regarding the recognition of pathogens by rice plants and the signaling events in rice innate immunity. Here, we summarize the advances in understanding rice innate immunity and the application of that understanding to the breeding of disease-resistant varieties. We also discuss the major challenges for future research.

PLANT INNATE IMMUNITY
Over the last two decades, extensive genetic and molecular studies of plant-microbe interactions in several model systems have revealed that plants have evolved a two-branched innate immunity system that detects and wards off various pathogens, resulting in disease resistance [1]. According to the standard zigzag model to illustrate the plant two-branched immune system in response to pathogens, the first branch uses transmembrane pattern-recognition receptors (PRRs) that recognize conserved pathogen-associated molecular patterns (PAMPs), leading to an immune response called PAMP-triggered immunity (PTI). To circumvent PTI, fungal, bacterial, viral, and nematode pathogens evolve effector proteins that suppress host defenses leading to effector-triggered susceptibility (ETS). The second branch, which mostly acts within the cell, uses highly polymorphic resistance (R) proteins that respond to pathogen effectors, leading to a rapid and robust effectortriggered immunity (ETI). However, this zigzag [4] Bakanae Fusarium fujikuroi Yield reductions and mycotoxin contamination [5] Bacterial disease Bacterial blight Xanthomonas oryzae pv. oryzae 10-50% [6] Bacterial leaf streak Xanthomonas oryzae pv. oryzicola 8-32% [7] Bacterial panicle blight Burkholderia glumae Up to 85% [ [11,12]. As the result of the plant-virus coevolution, viral suppressors of RNAi (VSRs) are regarded as effectors to overcome host RNAi (regarding as ETS) [11,12]. Plant R proteins that recognize VSRs as avirulence proteins can mediate a strong defense as ETI [11].  [13][14][15] (Fig. 1). In recent years, the following re-emerging diseases have become increasingly important: rice sheath blight caused by Rhizoctonia solani, rice false smut caused by Ustilaginoidea virens (Cooke) Takah, rice bacterial panicle blight caused by Burkholderia glumae, and rice stripe disease caused by rice stripe virus (RSV) [3,[16][17][18]. Over the past two decades, the rice/M. oryzae and rice/Xoo pathosystems in particular have been the focus of intensive studies and have become molecular models for research on plantmicrobe interactions. The major advances in the understanding of rice innate immunity against bacterial and fungal pathogens were recently reviewed by Liu et al. [19]. In this review, we consider the new progress in the two model pathosystems and novel insights into rice innate immunity against nematode and viral pathogens.

RICE PRR REPERTOIRE AND PTI
In a long-term evolutionary arms race with pathogenic microorganisms, plants have evolved a repertoire of PRR genes that recognize the conserved microbial PAMPs, leading to the inhibition of pathogen infection [1]. Plant PRRs are cell-surface receptors that perceive PAMPs released from the infecting pathogens in the extracellular environment; the perception of PAMPs by PRRs results in PTI responses. Plant PRRs are represented by transmembrane receptor-like kinases (RLKs), which typically contain extracellular leucine-rich repeats and an intracellular kinase domain, and receptor-like proteins (RLPs), which lack a kinase domain [20]. Because RLPs lack a cytoplasmic kinase domain, they recruit proteins containing kinase domains for the activation of the downstream signaling pathways. More than 1131 RLK genes have been identified in the rice genome; this is nearly two times the number in Arabidopsis and probably results from duplication events in the RLK genes of rice [21]. RLPs form a second major class of cell-surface receptors in plants, and the rice genome encodes 90 RLP genes [22]. Together, these receptor classes respond to a wide variety of activating ligands (lipid, protein, nucleic acids, carbohydrate, etc.) from various exogenous sources, such as pathogens and host-derived endogenous danger signals. Studies have increasingly shown that conserved PAMPs such as bacterial flagellin, peptidoglycan, lipopolysaccharide, and fungal chitin can be sensed by rice cells and trigger innate immunity [23][24][25][26]. Several rice PRR proteins including XA21, Os-FLS2, CEBiP, OsCERK1, LYP4, and LYP6 have been well characterized ( Table 2). The rice RLK gene Xa21 was one of the first innate immune receptor genes to be isolated and confers resistance to a wide range of Xoo strains [27]. The XA21-mediated signaling network has been intensively studied through genetic and biochemical approaches [19,23]. A number of previous studies have identified several Xoo Rax (required for activation of Xa21) genes that activate the XA21mediated immune response [28]. These genes are loca ted in a single operon (raxSTAB) that includes a tyrosine sulfotransferase (RaxST) and three LRR RLK Co-receptor kinases of XA21 and regulates BR-mediated development signaling [35] components (RaxA, RaxB, and RaxC) of a predicted type 1 secretion system [28]. Based on these findings, researchers hypothesized that a tyrosinesulfated, type 1-secreted protein activates XA21mediated immunity. Consistent with this hypothesis, a sulfated, 21-amino acid (AA) synthetic peptide (RaxX21-sY) derived from RaxX protein secreted by Xoo was proved to be essential for triggering XA21-mediated resistance [36]. Interestingly, RaxX residues between 40 to 55 share remarkable similarity with Arabidopsis signaling factor PSY1 (sulfated, secreted 18-AA peptide) and four predicted rice PSY1 orthologs [36]. The high similarities suggest that when a rice plant lacks XA21, Xoo and other Xanthomonads might use sulfated RaxX to mimic PSY1-like peptides in order to suppress host defense responses and facilitate infection [36]. OsFLS2 is the rice ortholog of Arabidopsis FLS2, and heterologous expression of OsFLS2 in the fls2 mutant can restore the fls2 mutant defects in Arabidopsis [29]. Like FLS2, OsFLS2 can directly recognize flg22 and trigger an immune response in rice [30]. These results indicate that the flg22 signaling pathway is conserved between Arabidopsis and rice and that OsFLS2 may also provide PTI-mediated defense in rice. Researchers have characterized several chitin immune receptors (CEBiP, OsCERK1, LYP4, and LYP6) that directly or indirectly recognize chitin fragments and trigger defense responses in rice [24,25,31]. Intriguingly, OsCERK1, LYP4, and LYP6 are also important for triggering immune responses to bacterial PGN in rice [24]. Furthermore, recent evidence indicates that the receptor-like cytoplasmic kinases OsRLCK185 and OsRLCK176 function downstream of OsCERK1 in the chitin and PGN signaling pathways, suggesting that chitin and PGN share intracellular signal-ing components [33]. Therefore, OsCERK1 functions as an adaptor in conjunction with OsLYP4 and OsLYP6 and plays dual roles in PGN and chitin signaling in rice innate immunity. These results demonstrate that multiple PRR proteins may work together to respond to PAMPs in rice.

RICE R GENE REPERTOIRE AND ETI
It is well known that nucleotide-binding and leucinerich repeat domain (NLR) proteins function as immune receptors in both animals and plants [37]. However, plant genomes contain many more NLRs than animal genomes, indicating differences in the two immune systems. The rice genome, for example, contains about 480 NLR genes while the human genome has only about 10 [38]. Interestingly, the majority of the cloned R genes encode NLR proteins (Table 3), although several atypical R proteins containing a variety of conserved protein domains/motifs are also identified (Fig. 2). Details concerning the structure and function of the cloned R genes have been reviewed and discussed in Liu et al. [19].
In the last 2 years, five new R genes (Pi50, Pi64, Xa10, Xa23, and STV11) have been cloned. Among them, Pi50 and Pi64 encode typical NLR proteins [39,40]. NLR genes are usually located in clusters in plant genomes; of the 480 NLR genes in rice, for example, 263 reside in 44 clusters [38]. Rice R genes Pi2, Pi9, and Piz-t are located in one of these NLR gene clusters on chromosome 6, and at least eight R genes are located at this locus in both wild and cultivated rice [41]. The newly cloned Pi50 gene is located at the Pi2/9 locus and confers broad-spectrum resistance to M. oryzae [42]. Liu and Wang 299     The Pi50 cluster contains four duplicated genes (Pi50 NBS4 1/2 and Pi50 NBS4 3/4) that differ in only four AAs [39]. Complementation tests revealed that Pi50 NBS4 1/2 but not Pi50 NBS4 3/4 con-fer Pi50-mediated blast resistance in rice [39]. Pi50 shares more than 96% AA sequence identity with Pi2, Pi9, and Piz-t, suggesting that Pi50 is derived from the functional divergence of duplicated genes [39]. The allelic gene Pi64 encodes a 1288-AA protein and is localized in both the cytoplasm and nucleus [40]. Pi64 is constitutively expressed in all tissues and at all development stages, and confers a high level of resistance to both leaf and neck blast in rice [40]. Both Xa10 and Xa23 are executor R proteins that confer the transcription activator-like effector (TALE)-dependent resistance to bacterial blight in rice [79,80]. The XA10 protein localizes as hexamers in the endoplasmic reticulum (ER) and such localization coincides with the ER Ca 2+ depletion and XA10-induced cell death in plants [79]. These results suggest that XA10 is an inducible protein that triggers programmed cell death by a conserved mechanism involving disruption of the ER and of cellular Ca 2+ homeostasis. The Xa23 protein shares 50% identity with XA10, and these two executor R proteins also have a similar predicted transmembrane helices structure [80]. Xa23 transcription is specifically activated by the TALE AvrXa23, and XA23 can trigger a strong immune response in rice, tobacco, and tomato [80]. The promoters of both Xa10 and Xa23 contain a TALE-binding element that is essential for cognate TALE-induced resistance [79,80]. These results suggest that the rice genome has evolved an executor R gene family, the members of which function in disease resistance by recognizing the cognate TALEs in Xoo. Liu and Wang 301 STV11, which confers durable resistance to RSV, was recently cloned by a map-based cloning strategy [18]. The gene encodes a sulfotransferase that can catalyze the conversion of salicylic acid (SA) into sulphonated salicylic acid (SSA) in RSV-infected plants, and SSA is more effective than SA in triggering RSV resistance and in inhibiting viral replication [18]. Moreover, SSA may also serve as a signal to enhance SA biosynthesis through a positive feedback mechanism after RSV infection; SA may contribute to the inhibition of viral replication in the RSVinfected plants [18]. STV11-R is prevalent in cultivated indica rice cultivars, whereas the susceptible allele STV11-S is prevalent in japonica cultivars. The cloning of STV11 will facilitate the breeding of RSVresistant rice through molecular marker-assisted selection; such resistance will greatly improve RSV management in rice production.

REVIEW
Our understanding of rice resistance to nematodes has lagged behind the soybean-nematode pathosystem. For instances, two soybean cyst nematode (SCN) resistance genes (Rhg1 and Rhg4) have been cloned through a map-based cloning strategy [83,84]. The Rhg1 gene encodes three proteins [an AA transporter (Glyma18g02580), an a-SNAP protein (Glyma18g02590), and a WI12 (woundinducible domain protein), (Glyma18g02610)], all of which are essential for the resistance to SCN [83]. A physical structure study revealed that the rhg1 locus that encodes these three proteins is present in multiple copies (10 tandem copies) in SCN resistant lines, whereas only one copy is present in susceptible cultivars [83]. Overexpression of the individual genes is ineffective, but overexpression of the three genes together enhances SCN resistance [83]. These results suggest that variation in the copy number of multiple genes at Rhg1 mediates SCN resistance in soybean. Rhg4 encodes a ubiquitous enzyme (serine hydroxymethyltransferase) that is responsible for interconversion of serine and glycine and that is important for cellular one-carbon metabolism [84]. Two genetic polymorphisms (R130P and Y358N) were detected in the Rhg4 alleles of resistant versus susceptible cultivars, suggesting that these two AAs are important for the regulatory function of this enzyme [84]. A linkage mapping study revealed a major resistance gene (Has-1 Og ) against rice cyst nematode caused by Heterodera sacchari and it was delimited to a 8.2 cM interval between the markers RM254 and RM206 on chromosome 11 in rice [85]. However, the gene encodes Has-1 Og have not been cloned. Because another three species of cyst nematodes (H. oryzicola, H. elachista, and H. oryzae) also frequently infect rice and cause significant annual yield lost, additional identification and cloning of genes responsi-ble for resistance to the cyst nematodes that attacks rice is urgently needed.
Recently, many new resistance genes have been mapped via genome-wide association studies (GWASs) of large collections of rice germplasm. Wang et al., for example, investigated 366 diverse indica rice accessions using 0.8 million singlenucleotide polymorphisms (SNPs) and identified 30 loci that are significantly related to resistance to M. oryzae [86]. In that study, a new R gene locus was identified on chromosome 3 where no blast R gene had been previously reported [86]. Using 372 diverse rice cultivars collected from 82 countries and 700 000-SNP arrays, Kang et al. identified 97 loci associated with blast resistance (LABRs) against five diverse isolates [87]. Among these loci, 82 are new regions, and 15 are co-localized with known blast resistance loci [87]. Further functional analysis of the candidate genes in the LABR 64 region via RNAi technology identified two new R alleles at the Pi5 locus [87]. These results suggest that GWAS is an efficient strategy for rapid allele discovery and that GWAS, when coupled with RNAi technology, will help researchers dissect complex disease resistance in rice. Another recent study investigated the function of 332 NLR genes that were cloned from five blast-resistant rice cultivars [88]. Strikingly, 98 of them confer resistance to one of the tested blast isolates, demonstrating that a systemic approach can increase the efficiency of R gene cloning in rice.

PATHOGEN EFFECTORS AND THEIR HOST TARGETS
In a broad sense, effectors are pathogen proteins and small molecules that can alter host cell structure and function [89]. Avr effectors are those molecules that are recognized by the cognate host R proteins directly or indirectly in plant cells; the recognition triggers a rapid and robust hypersensitive reaction. To date, a total of 21 Avr effector genes have been cloned in rice pathogens, and these include 13 from M. oryzae, 7 from Xoo, and 1 from Xoc ( Table 3). The identification of these Avr genes has greatly facilitated the investigation of the molecular basis of the interaction between Avr effectors and R proteins. The examples of direct and indirect interactions between two types of proteins and host targets of the Avr effectors have recently been reviewed [19,90].
AvrPib and AvrPi9 were recently cloned in M. oryzae. AvrPib, the cognate Avr gene of the R gene Pib, was cloned using a map-based cloning strategy. It encodes a 75-AA protein with no homology REVIEW to any protein in the database [43]. Phenotyping and genotyping of 60 M. oryzae isolates collected from five geographically distinct areas suggested that AvrPib has undergone host-driven selection [43]. Resequencing of the AvrPib allele of 108 diverse isolates revealed that transposable element (TE) insertion (frequency 81.7%) is the prevalent mechanism that leads to the loss of its avirulence function [43]. AvrPi9, the Avr gene of the R gene Pi9, was cloned using a comparative genomic approach with virulent mutant strains derived from a sequential planting method [47]. The AvrPi9 protein is highly expressed at early stages of M. oryzae infection [47]. Moreover, the AvrPi9 protein localizes in the biotrophic interfacial complex and appears to be translocated into rice cells during infection [47]. Like AvrPib, TEs also play an important role in acquisition of virulence in the AvrPi9 alleles in M. oryzae.
Magnaporthe oryzae secretes various effectors that enter infected rice cells and then move to neighboring cells, presumably targeting host proteins to prepare for infection [91]. Several host targets of Avr effectors have been recently characterized. For instance, the AvrPiz-t effector targets the rice RING E3 ligase APIP6 and suppresses PTI [92]. Interestingly, the interaction between AvrPiz-t and APIP6 leads to their mutual degradation [92]. Transgenic rice plants expressing the APIP6 RNAi construct have reduced PTI responses and reduced basal resistance to M. oryzae [92], suggesting that APIP6 positively regulates rice innate immunity. A recent study showed that APIP6 interacts with and degrades OsELF3-2 (ortholog of Arabidopsis flowering and circadian regulator ELF3) [93]. The oself3-2 T-DNA mutant and RNAi plant exhibit enhanced resistance to M. oryzae [93], indicating that OsELF3-2 negatively regulates rice innate immunity against M. oryzae.
The exocyst is an octameric protein complex that functions in vesicle trafficking. Its subunits Exo70B2 and Exo70H1 in Arabidopsis are involved in the response to pathogens, with Exo70B2 having a more important role in cell wall apposition formation related to plant defense [94]. The Avr-Pii effector targets two rice Exo70 proteins (OsExo70-F2 and OsExo70-F3) to form a protein complex in rice cells [95]. Functional assays showed that OsExo70-F3 but not OsExo70-F2 is specifically involved in Piidependent resistance [95]. Moreover, overexpression of Avr-Pii or silencing of OsExo70-F2 and -F3 genes in rice did not affect the virulence to compatible M. oryzae strains [95]. These results suggest that the Avr-Pii targets OsExo70-F3 and the rice exocytosis pathway are important for ETI and that Os-Exo70 functions as a decoy or helper in Pii/Avr-Pii interactions.

HORMONE-MEDIATED IMMUNITY IN RICE
Rice hormones such as SA (salicylic acid), JA (jasmonate acid), and ET (ethylene) are important regulators of immune responses [96][97][98]. Two excellent reviews summarized the advances in understanding the functions of various hormones in rice immunity in 2013 [99,100]. Here, we provide the recent progress on hormone-mediated immunity in rice during the past few years.
SA, JA, and ET are three main hormones that play important roles in plant immunity. SA is usually considered to regulate immunity against biotrophic pathogens, whereas JA and ET are believed to be involved in resistance to necrotrophic and insect pests [101]. However, this dichotomy does not fully fit into the monocotyledonous plant rice [10]. Different from the dicot plant Arabidopsis, rice plants challenged by fungal and bacterial pathogens do not show SA accumulation [102]. However, rice plants indeed respond to exogenous SA treatment [102]. These results suggest that rather than the endogenous SA level, the involvement of SA in rice defense responses is more dependent on the SA signaling [99].
Accumulating evidence reveals that extensive crosstalk between different hormones exists in rice plants in response to pathogen infections. For instance, the rice DELLA protein SLR1 (slender rice1) represses the transcription of gibberellic acid (GA)-responsive genes and functions as a key regulator of GA signaling [103]. Vleesschauwer et al. recently found that SLR1 functions in resistance to hemibiotrophic but not necrotrophic pathogens [104]. Moreover, they demonstrated that SLR1 mediates resistance through integrating and amplifying both SA-and JA-dependent defense signaling pathways in rice [104]. A recent transcriptome study of root-knot nematode-infected rice plants reveals that a number of well-identified marker genes involved in the SA/JA/ET pathways show significantly differential expression patterns between susceptible and resistant interactions [105]. These results indicate that various plant hormones are involved in the ricenematode interaction and further in-depth studies are needed to decipher the underlying mechanism of hormone-mediated resistance in this pathosystem.
Plant hormone pathways are often targeted by pathogen effectors for suppression of hormonemediated immunity. For example, M. oryzae encodes an antibiotic biosynthesis monooxygenase (Abm) that converts endogenous free JA into hydroxylated JA (12OH-JA) to attenuate rice innate immunity during fungal colonization [106]. The wild-type strain of M. oryzae secretes 12OH-JA during host REVIEW Liu and Wang 303 penetration to avoid the defense response, whereas the Abm mutant of M. oryzae accumulates methyl JA (MeJA), which induces rice defense [106]. Notably, M. oryzae also secretes Abm after invasion, and the secreted Abm appears to convert plant JA into 12OH-JA to facilitate host colonization [106], indicating that Abm is an effector protein that is important for M. oryzae pathogenicity. The host target of Abm remains to be identified. In addition to inducing or manipulating host hormone biosynthesis, most plant pathogens are producing hormones as virulence factors [107]. For example, rice bakanae disease pathogen Fusarium fujikuroi produces chemically similar GA that probably functions as a suppressor of host defense responses through modulating hormonal balance in plants [107]. Many gall-forming bacteria and biotrophic fungi produce cytokinins (CKs) that are required for the establishment of diseases [107]. However, the underlying mechanism of CKs produced by plant pathogens during infection remains largely unknown. Recently, Chanclud et al. identified the gene CKS1 (cytokinin synthesis 1) that is required for CK synthesis and full virulence in M. oryzae [108]. Moreover, they showed that the CKs produced by M. oryzae are important for dampening host defense and affecting plant nutrients (sugar and AAs) distribution that facilitate for fungal growth in and around the infection site [108], indicating this fungal-secreted CKs are key effectors that are similar with the TALE from bacteria. Interestingly, Bockhaven et al. recently found that rice plants treated with 2 mM silicon (si) significantly increase resistance to the brown spot fungus Cochliobolus miyabeanus [109]. Rather than suppressing rice ET signaling, Si application increases resistance to rice brown spot probably through interfering with the production and/or action of ET in C. miyabeanus [109]. These results suggest that impairment of hormone production in pathogens is an efficient strategy to control plant diseases resistance.

STRUCTURAL INSIGHT INTO RICE/PATHOGEN SYSTEMS
Advances in X-ray crystallography promise to deepen our understanding of the recognition between plant NLRs and pathogen effectors at the molecular level. The technique has been recently used to analyze the interaction between rice NLRs and M. oryzae effectors. According to X-ray crystallography, the Avr effector AvrPiz-t adopts a six-stranded β-sandwich-fold structure, and Cys62 forms a disulphide bond with Cys75 [110]. de Guillen et al. recently used NMR spec-troscopy to determine the 3D structures of the M. oryzae effectors Avr1-CO39, Avr-Pia, and AvrPiz-t and of the Pyrenophora tritici-repentis (wheat tan spot pathogen) effector ToxB [111]. The analysis showed that these effectors have very similar six β-sandwich structures that are stabilized by a disulfide bridge between two conserved cysteins located in similar positions of the proteins. These sequence unrelated but structurally similar fungal effectors were termed MAX effectors. Most M. oryzae MAX effectors are highly expressed early during infection. Determining whether the MAX effectors have similar functions in pathogenesis and whether they can target conserved host proteins will require further investigation.
Maqbool et al. recently used biochemical, structural, and activity-based assays to study how the rice NLR protein Pik directly interacts with the M. oryzae effector Avr-Pik [112]. Coexpression of Pikp-HMA and Avr-PikD and the analysis of the 3D crystal structure of their complex revealed that Avr-PikD has high affinity binding to the so-called integrated HMA domain in Pikp [112]; this binding initiates immunity responses. Furthermore, mutated Avr-PikD compromises the interaction with the Pikp-HMA domain and therefore abolishes the Avr-PikD-Pikp-triggered defense response in rice [112].
Finally, a recent copurification and crystal structure study revealed that the Xanthomonas type III effector AvrRox1-ORF1 binds to a molecular chaperone AvrRox1-ORF2 to form a tetramer complex with a distinct fold containing a novel kinase-binding domain [113]; the AvrRox1-ORF2 chaperone is structurally different from typical effector-binding chaperones. This tetramer complex is structurally homologous to zeta toxin:epsilon antitoxin [113]. AvrRox1-ORF1 encodes a T4 polynucleotide kinase-like domain that might directly phosphorylate a host target [113].

BREEDING OF DISEASE-RESISTANT RICE
Researchers have estimated that crop yields must be increased by150% before 2030 to meet the global food demand [114]. This increase in yield will be difficult to achieve because of many limiting factors including pathogens. During the past decades, the breeding of disease-resistant rice cultivars has greatly increased yield in China and several Asian countries. For example, many R genes against M. oryzae, Xoo, and RSV have been integrated into new rice cultivars through marker-assisted selection and genetic engineering breeding strategies in China [114]. Readers are referred to a recent comprehensive review on the progress of rice molecular breeding in China [114].

REVIEW
In addition to conventional approaches, novel strategies based on host-induced gene silencing (HIGS), Xanthomonas spp. transcription activatorlike effector nucleases (TALENs), and a bacterial monomeric DNA endonuclease CRISPR-associated protein 9 (CRISPR/Cas9) have been successfully used to increase resistance against pathogens in plants. The first successful application of HIGS in disease control was the expression of papaya ringspot virus (PSRV) coat protein in transgenic papaya plants to inhibit PSRV infection [115]. Growing evidence suggests that the expression of dsRNA molecules that target important genes in nematodes, fungi, and even insects might also generate resistant plants [116]. For instance, transgenic plants expressing fungal virulence gene constructs can specifically silence host targets in the case of the pathogenic fungi Blumeria graminis, Fusarium species, and Puccinia striiformis f.sp. tritici [117][118][119]. The use of HIGS to control rice blast and sheath blight is being studied in several laboratories and may generate transgenic lines with resistance to multiple pathogens if the target pathogen DNA sequence is highly conserved.
Genome-editing technology has great potential for the engineering of plants that have a broad spectrum of resistance but are free of antibiotic markers. TALENs encode artificial bipartite enzymes that consist of a modular DNA-binding domain and the FokI nuclease domain [120]. The DNAbinding domain has been engineered to recognize a specific DNA sequence. The ability to precisely edit a specific host gene, such as the target of a bacterial virulence gene, can result in the development of transgenic crops that thwart the virulence strategy of Xanthomonas spp. For example, resistant and hygromycin-free rice plants have been generated with TALEN technology; the resistance of these plants is based on the targeting of the bacterial blight susceptibility gene Os11N3 (also called Os-SWEET14) [121]. The CRISPR/Cas9-based geneediting tool is becoming increasingly important. This technology simply uses engineered 20 base pair (bp) RNA guide sequence that binds to its DNA target site of interests to cause DNA cleavage and mismatching repairing or homologous replacement [122]. To date, CRISPR/Cas9-based gene editing has been used for many organisms, including the model crop plants rice, maize, and wheat [123]. Simultaneous editing of three mildew resistance locus o (Mlo) genes in hexaploid bread wheat led to the generation of heritable resistance to the powdery mildew fungus Blumeria graminis f. sp. tritici (Bgt) [124]. A new method to edit plant genomes without introducing foreign DNA into cells was recently reported; this may alleviate regulatory con-cerns related to genetically modified plants [125]. With this new method, transgenic plants were generated from the protoplasts of Arabidopsis thaliana, tobacco, lettuce, and rice transfected with purified Cas9 protein and guide RNA. These plants contain only small insertions or deletions that are indistinguishable from naturally occurring genetic variations. In the future, improvements in the application of CRISPR/Cas9 technology will likely lead to novel and broad-spectrum disease resistance in crops.

CONCLUSION AND PERSPECTIVES
During the last two decades, tremendous progress has been made in understanding the innate immune receptor complex in rice. More than 40 rice PRR and R genes have been identified and functionally characterized. These genes help regulate the defense responses to bacterial, fungal, and viral pathogens. Breakthroughs have included the determination of rice immune receptors and how such receptors recognize fungal and bacterial ligands, the understanding of the structure of the rice immune receptor complex, and the development of novel strategies for rice diseases management. Research is needed in the following areas: (1) the connections and interactions between the signaling components of rice PRR and NLR-mediated resistance for defense activation, (2) the function of transcriptional factors that receive signals from PRRs and NLRs and that control the downstream defense gene activation in the nucleus, (3) the role of epigenetic regulations in rice immunity, and (4) the application of our increasing understanding of rice innate immunity to achieve disease control in rice fields.