Resistance Genes and their Interactions with Bacterial Blight/Leaf Streak Pathogens (Xanthomonas oryzae) in Rice (Oryza sativa L.)—an Updated Review

Rice (Oryza sativa L.) is a staple food crop, feeding more than 50% of the world’s population. Diseases caused by bacterial, fungal, and viral pathogens constantly threaten the rice production and lead to enormous yield losses. Bacterial blight (BB) and bacterial leaf streak (BLS), caused respectively by gram-negative bacteria Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc), are two important diseases affecting rice production worldwide. Due to the economic importance, extensive genetic and genomic studies have been conducted to elucidate the molecular mechanism of rice response to Xoo and Xoc in the last two decades. A series of resistance (R) genes and their cognate avirulence and virulence effector genes have been characterized. Here, we summarize the recent advances in studies on interactions between rice and the two pathogens through these R genes or their products and effectors. Breeding strategies to develop varieties with durable and broad-spectrum resistance to Xanthomonas oryzae based on the published studies are also discussed.


Background
Plants are always attacked by diverse and widespread potential pathogens, which cause numerous diseases. These diseases lead to 16% of global crop yield losses (Oerke 2006). Plants have evolved sophisticated innate ability of each cell to fend off the attack (Spoel and Dong 2012). There are two-layered system involved in plant immune response. The first layer is governed by cell surfacelocalized pattern recognition receptors (PRRs) that detect pathogen-associated molecular patterns (PAMPs), such as bacterial flagellin or fungal chitin, which are highly conserved molecules essential for the pathogen's life cycle, and trigger a relatively weak immunity (PTI). PTI comprises a wide array of responses, including the production of reactive oxygen species (ROS), increases in intracellular calcium concentration, callose deposition in cell wall, antimicrobial compounds called phytoalexins and the activation of mitogen-activated protein kinases (MAPKs) . It is a broadspectrum resistance that wards off most invading organisms. To counter PTI, the pathogens evolved mechanisms to secret and deliver highly variable effectors into host cells to suppress PTI, which is called effectortriggered susceptibility (ETS). The second layer of plant defense acts largely inside the cell and is based on highly polymorphic resistance proteins which directly or indirectly recognize specific virulence effectors secreted within host cells by pathogens, inducing the effectortriggered immunity (ETI). ETI is a rapid and stronger resistance response, usually associated with programmed cell death at sites of infection, termed the hypersensitive response (HR). Other defense responses include the production of ROS, enhancement of cell walls, accumulation of toxic metabolites or proteins, and altered levels of hormone ).
The ancient domesticated crop, rice (Oryza sativa L.) is the most important staple food for humans and is one of the most widely cultivated crops all over the world (Ainsworth 2008). Though rice production has been almost doubled over the recent decades due to the introduction of the semi-dwarf gene sd1, hybrids, and improvements in cultivation management practices, it needs to significantly increase in order to meet the projected demand from the ever-expanding human population (Khush 2005;Skamnioti and Gurr 2009). However, the increase is challenged by farmland availability, water, soil fertility, climate change, insects and diseases. Rice is vulnerable to a number of diseases caused by bacteria, viruses, or fungi (Dai et al. 2010). Rice bacterial blight (BB) and bacterial leaf streak (BLS) are caused by gram negative bacteria Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc), respectively. BB is one of the most devastating rice diseases, which can cause severe yield loss of up to 50% depending on the rice variety, growth stage, the geographic location and environmental conditions (Liu et al. 2014). Losses due to the kresek syndrome of BB can reach as much as 75% (Ou 1985). BLS is another devastating rice disease which could spread rapidly under favourable conditions and cause tremendous damage. Yield losses due to BLS range from 8%-32% (Liu et al. 2014). It is becoming more and more important, especially in Asia and Africa. In China, quarantine regulations are now in force for BLS (Li and Wang 2013). In this updated review, we provide an overview of these two diseases and summarize the advances in studies on the Xoo/Xoc-rice interaction. We also discuss strategies for breeding broadspectrum and durable disease-resistant rice varieties.

Overview of the Pathogens and Diseases
BB is one of the oldest recorded rice diseases, which was first found by a farmer in the Fukuoka area of southern Japan in 1884 (Nino-Liu et al. 2006). Since then, it was observed in other regions of Japan and gradually spread to all the rice-growing areas of this country. In China, rice BB was observed as early as 1930s and it spread throughout ten provinces in the south of China by the end of 1950s. However, rice BB was not a severe disease until the 1970s (Zhang 2009). Damage caused by this disease was significantly increased due to the widespread cultivation of semi-dwarf and hybrid rice varieties, as well as massive input of nitrogen fertilizer. It was prevalent in other Asian countries during this period, including India, Philippines, Nepal, Indonesia and Sri Lanka. After that, its incidence was reported in Australia, America and West Africa. To date, rice BB is widely distributed in almost all the rice-growing countries in the world (Naqvi 2019).
BLS was first observed in Philippines in 1918. Since then, the occurrence of BLS in the tropical and subtropical Asia, northern Australia and West Africa was also reported. In China, it was first observed in Guangdong Province, and has recently become one of the major diseases in South China (Tang et al. 2000;Xie et al. 2014).
Though Xoc and Xoo are highly related bacterial species, they infect rice in different ways. Xoo enters leaf through the hydathodes or wounds, multiplies in the intercellular spaces of the underlying epitheme, and propagate to reach the xylem vessels. The bacteria move through the veins of leaves and spread into the plant. Water-soaked spots at the leaf tips and margins were first observed. Then, the leaves become chlorotic and necrotic along the leaf veins (Lee et al. 2011) (Fig. 1a). Xoc penetrates the leaf mainly through stomata or wounds, multiplies in the substomatal cavity and then colonizes the intercellular spaces of the parenchyma. Different from BB, small, water-soaked lesions anywhere along the leaf between the veins were observed during the early stage of BLS infection, resulting in translucent and yellow streaks (Fig. 1b). The infected leaves turn greyish white and die later on (Nino-Liu et al. 2006).
Diverse effector proteins with virulence, avirulence functions or both are secreted by Xanthomonas oryzae. Among them, transcription activator like (TAL) effector proteins are a structurally and functionally distinct class of proteins secreted into plant cells by a type III secretion (T3S) system. TAL effectors (also termed as TALEs) import in the nucleus and bind to TALE-specific DNA, which is termed as effector binding elements (EBEs). The recognition transcriptionally activates host target genes, resulting in susceptibility or resistance Bogdanove and Voytas 2011).

Disease Resistance Genes and the Interactions
Deployment of gene-conferred host plant resistance provides an economical, effective, environment friendly approach to control plant diseases and minimize the losses. Extensive genetic studies on rice resistance to BB have been conducted over the last 20 years. To date, more than 40 resistance (R) genes conferring host resistance to various strains of Xoo have been identified and 11 of them were cloned, namely Xa1, Xa3/Xa26, Xa4, xa5, Xa10, xa13, Xa21, Xa23, xa25, Xa27, and xa41 (Table 1)  . These R genes can be classified into four groups based on their encoding proteins, including receptor-like kinase (RLK) genes (Xa21, Xa3/ Xa26 and Xa4), sugar will eventually be exported transporter (SWEET) genes (xa13, xa25 and xa41), executor genes (Xa10, Xa23 and Xa27) and other types of genes (Xa1 and xa5). Some of these isolated R genes are widely employed in rice breeding programs to control BB, such as Xa3/Xa26 and Xa4, which played an important role in controlling the disease in Asia since 1970s. Nearly all  TAL effector (Antony et al. 2010;Yu et al. 2011;Streubel et al. 2013;Hutin et al. 2015) Xa10 Executor R protein AvrXa10 TAL effector (Tian et al. 2014)

Rxo1
NLR AvrRxo1 TAL effector (Zhao et al. 2004a;Zhao et al. 2004b) NLR nucleotide-binding domain and leucine-rich repeat, LRR-RLK leucine-rich repeat receptor-like kinase, TFIIA transcription factor IIA, SWEET sugar will eventually be exported transporter, TAL transcription activator like the commercial indica hybrid rice varieties in China are known to contain Xa4, and Xa3/Xa26 is widely distributed in both indica and japonica varieties in China (Deng et al. 2018;Hu et al. 2017). The cognate avirulence (Avr) genes to all the R genes except Xa4 have been reported (Table 1). In contrast to BB, no native major R gene controlling resistance to BLS has been identified in rice and only a few of quantitative resistance loci have been mapped. Interestingly, one of them, qBlsr5a, with relatively large effect, was mainly controlled by xa5 (Xie et al. 2014). A non-host R gene, Rxo1, was isolated from maize, and the transgenic rice with Rxo1 has been proved to confer high level resistance to BLS (Zhao et al. 2005).
In addition, some defense-related or susceptible genes in rice were reported to be involved in the interaction with Xoc (Shen et al. 2010;Tao et al. 2009). Here, we focus on the recent advances in identification of the R genes or their products and the cognate pathogen effectors. The underlying molecular mechanisms of the interaction between rice and Xoo or Xoc are discussed. Additionally, two genes, Xa7 and Xo1, which have not been cloned yet, are also discussed due to their potential value in rice breeding programmes and special features. To date, most of the cloned plant R genes encode nucleotide-binding and leucine-rich repeat domain (NLR) proteins .
However, only one encodes NLR protein among the 11 cloned Xa genes (Yoshimura et al. 1998). These Xa genes are classified into four groups based on the encoded protein types including RLK (receptor-like kinase), SWEET (sugar will eventually be exported transporter), executor R proteins and other proteins.

Receptor-Like Kinase (RLK) Genes
In plants, PRRs, which can recognize diverse pathogenassociated molecular patterns are a key component of the innate immune system. All the known plant PRRs are either transmembrane receptor-like kinases (RLKs) or transmembrane receptor-like proteins (RLPs) (Antolín-Llovera et al. 2012). There are over 1100 candidate RLKs/RLPs in rice genome (Shiu et al. 2004). RLKs typically contain an extracellular domain, a single-pass transmembrane domain, and an intracellular kinase domains, whereas RLPs lack the kinase domain (Monaghan and Zipfel 2012). Leucine-rich repeat receptor-like kinases (LRR-RLKs) represent the largest subfamily of plant RLKs (Afzal et al. 2008).
The LRR-RLK gene Xa21, originated from the wild rice species Oryza longistaminata, was the first cloned R gene in rice (Song et al. 1995). Xa21 have been proved to confer broad-spectrum resistance to Xoo. However, Xa21-mediated resistance progressively increases from the susceptible juvenile two-leaf stage through later stages, with full resistance only at the adult stage (Century et al. 1999;Wang et al. 1996). Overexpression of Xa21 gene can enable plants with resistance at both seedling and adult stages (Park et al. 2010a). The regulation of Xa21-mediated immunity has been extensively and comprehensively studied. Several XA21 binding proteins (XBs) with diverse functions have been characterized (Table 2, Fig. 2). The phosphorylation state of XA21 is important for its function. In the absence of infection, the ATPase XB24 physically associates with the XA21 juxtamembrane domain and promote phosphorylation of specific serine and threonine residues to maintain the inactive state of the XA21 protein. On recognition of pathogen invasion, the XA21 kinase disassociates from XB24 and triggers downstream defense responses (Chen et al. 2010b). After activation, XB15, a PP2C phosphatase, acts on XA21 and dephosphorylates the autophosphorylated XA21 (Park et al. 2008). The Xoo tyrosine-sulfated and type I-secreted protein RaxX is the ligand to induce the XA21-mediated immunity (Pruitt et al. 2015). The sulfated RaxX directly binds XA21 with high affinity (Luu et al. 2019). More details are shown in Table 2 and Fig. 2.
Xa4, encoding a cell wall-associated kinase, confers a race-specific resistance to Xoo at all stages of rice growth (Leach et al. 2001;Sun et al. 2003;Hu et al. 2017). Wallassociated kinases (WAKs) are also a subfamily of RLKs that physically link the cell wall with the plasma membrane to transmit extracellular signals to the cytoplasm (Anderson et al. 2001). Xa4 was first introgressed into commercial rice varieties in the early 1970s. It is one of the most widely employed resistance genes in breeding programs. Nearly all the indica hybrid rice cultivars in China carry Xa4 (Leach et al. 2001). XA4 prevents the invasion of Xoo through reinforcing the cell wall (Hu  . 2017). The accumulation of the two phytoalexins, sakuranetin and momilactone A, which are likely to suppress Xoo in plant, is proved to be associated with Xa4mediated resistance. In addition to conferring durable resistance to Xoo, Xa4 increases the mechanical strength of the culm and reduces the plant height slightly, and thus may enhance the lodging resistance (Hu et al. 2017). The multiple favorable agronomic traits related with Xa4 may explain why it is widely used.

Sugar Will Eventually be Exported Transporter (SWEET) Genes
Three recessive R genes, xa13, xa25 and xa41, encodes clade III SWEET proteins. SWEET, a unique family of sugar efflux transporters, play a vital role in various biological processes, including pollen nutrition, senescence, seed filling and plant-pathogen interactions (Chen et al. 2012;Guan et al. 2008;Quirino et al. 1999;Streubel et al. 2013). SWEETs are grouped into a four-clade phylogenetic tree in plants . There are 17 and 22 SWEET genes in Arabidopsis and rice genomes, respectively (Chen et al. 2010a). Over the last 10 years, several studies have suggested sugar exporting into the apoplast via clade III SWEETs is hijacked by TAL effectors of pathogen, which is essential for pathogen growth and virulence . The xa13 (also known as Os8N3 and OsS-WEET11) confers specific resistance to Xoo race 6, which was originally identified in cultivar BJ1 (Chu et al. 2006). It was isolated through different strategies by two groups (Chu et al. 2006;Yang et al. 2006). The TAL effector PthXo1 from Xoo directly targets to the EBEs, in the promoter of dominant Xa13 but not xa13 alleles to induce its expression, which is critical for susceptibility (Fig. 3) (Römer et al. 2010;Yuan et al. 2009). Further studies showed that the XA13 protein cooperates with two copper transporters, COPT1 and COPT5, to participate in copper redistribution. Copper is widely used as an important element for pesticides in agriculture. XA13, COPT1 and COPT5 are employed by TAL effectors of Xoo and remove toxic Cu from xylem vessels, where pathogen multiplies and spreads to cause disease ). Interestingly, knock-out of OsSWEET11 showed increased resistance to Rhizoctonia solani, which causes sheath blight disease. It suggests that OsSWEET11 may also be employed by the fungal pathogen Rhizoctonia solani (Gao et al. 2018). In addition, Xa13 was found to be required for pollen development. The Xa13-silenced plants had low fertility, and most pollen grains were defective in comparison with normal pollen grains (Chu et al. 2006). As well as Xa3/Xa26, xa25 (also known as OsS-WEET13) was identified from Minghui 63 (Chen et al. 2002). It confers race-specific resistance to Xoo strain PXO339 at both seedling and adult stages. Similar to xa13, the expression of dominant Xa25 but not recessive xa25 was rapidly induced by PXO339 (Liu et al. 2011). Another type of recessive xa25 alleles was Fig. 3 The SWEET gene xa13 and the executor gene Xa23 mediated immune signaling pathways triggered by Xanthomonas oryzae. The TALEs are secreted into the cytoplasm of plant cells through the type III secretion system, enter the nucleus, bind to the specific promoter elements and induce the expression of Xa13. XA13 is hijacked by TALEs to export sucrose to the apoplast, which provides nutrition to the pathogen. XA13 is also employed together with COPT1 and COPT5 by TALEs to remove toxic Cu from xylem vessels. The resistant allele xa13 with mutations in the EBEs disrupt the binding of TALEs, leading to disease resistance. Like SWEET genes, the executor R gene Xa23 is transcriptionally activated by TALEs, triggering host defense responses identified in japonica rice varieties Nipponbare and Kitaake (Zhou et al. 2015). Further studies showed that OsSWEET13 as the disease-susceptibility gene is directly targeted by PthXo2. In a very recent study, two types of PthXo2-like TALEs were found to bind with different EBE sequences in the OsSWEET13 promoter and activate its expression (Xu et al. 2019).
Xa41 (also known as Os11N3 and OsSWEET14) was found to be targeted as a susceptibility gene by different TAL effectors from numerous Xoo strains, including AvrXa7, PthXo3, TalC and Tal5 (Antony et al. 2010;Hutin et al. 2015;Streubel et al. 2013;Yu et al. 2011). A germplasm screening for polymorphisms in the OsS-WEET14 promoter uncovers a natural candidate plant disease resistance gene from African wild and cultivated rice species O. barthii and O. glaberrima (Hutin et al. 2015). An allele of OsSWEET14 was identified to carry an 18-bp deletion at 8 bp downstream of the predicted TATA box, and could prevent OsSWEET14 induction by AvrXa7 and Tal5. The xa41 confers broad-spectrum resistance to 50% of the tested strains representing genetically distant groups isolated from different countries in Asia and Africa (Hutin et al. 2015). In another study, in silico mining of OsSWEET13 and OsSWEET14 promoter polymorphisms in a diversity germplasm panel containing 3000 rice genome sequences and the Pakistani aromatic germplasm collection was conducted (Zaka et al. 2018). Novel variations in the EBEs of OsSWEET13 and OsSWEET14 promoter regions were identified (Zaka et al. 2018).

Executor Genes
Xa27, Xa10 and Xa23 are three executor genes with multiple potential transmembrane domains functioning as a promoter trap, which are transcriptionally activated by TAL effectors and trigger defense responses (Gu et al. 2005;Tian et al. 2014;Wang et al. 2015). Xa27 originated from wild rice O. minuta Acc. 101,141 and confers broad-spectrum resistance to Xoo strains from different countries (Gu et al. 2005). Xa27-mediated resistance is also affected by developmental stage like Xa21 and Xa3/26. Challenged by Xoo containing AvrXa27, Xa27 was specifically induced and secreted to the apoplast, leading to inhibition of bacterial growth. However, the allele from the susceptible variety IR24 was not induced. Increased expression of Xa27 showed thickened vascular bundle elements, even in the absence of Xoo infection. Further study showed that localization of XA27 to the apoplast depending on the N-terminal signal-anchor-like sequence is important for its resistance to Xoo (Wu et al. 2008). Rice lines with both AvrXa27 and Xa27 showed enhanced resistance when inoculated with compatible strains of Xoo and Xoc .
Xa10, which confers resistance to some Philippine races of Xoo, was first identified from rice cultivar Cas 209 (Gu et al. 2008;Lee et al. 2003). AvrXa10 specifically induces Xa10 expression through direct binding Xa10 promoter. Rice plants with constitutive but weak expression of Xa10 showed lesion mimic phenotype. Further study has revealed that XA10 forms hexamers and locate in the ER membrane of plant and HeLa cells, which mediates the disruption of the ER, cellular Ca 2+ homeostasis and triggers programmed cell death (Tian et al. 2014).
Another executor gene Xa23 isolated from a wild rice species of O. rufipogon, confers an extremely broad spectrum of resistance to Xoo strains isolated from different regions at all growth stages of rice. Similar to Xa27, Xa23 shares identical ORF with the susceptible xa23 allele, and a 7-bp polymorphism in the promoter regions leads to induction of Xa23, but not xa23, by AvrXa23. Transient expression analysis indicated that XA23 triggers HR in N. benthamiana and tomato . AvrXa23 was found to be highly conserved in all the tested Xoo isolates . It is possible that AvrXa23 contributes to the virulence of Xoo for infection or growth in host plants. The prevalence of AvrXa23 in natural Xoo strains explains why Xa23 shows the broad-spectrum resistance.

Other Genes
In rice genome, 480 nucleotide-binding domain and leucine-rich repeat (NLR) genes have been revealed, but only a single one, Xa1, conferring resistance to Xoo, was isolated (Yoshimura et al. 1998). Xa1 was isolated from japonica cultivar Kogyoku and its expression was induced by bacterial infection and wounding (Yoshimura et al. 1996). Xa1 confers resistance against Xoo by recognizing several TAL effectors including PthXo1, Tal4 and Tal9d, but truncated interfering TAL effectors (also termed as iTALEs). The iTALEs may function as decoys interfering with the recognition of intact TALEs by XA1 and block its function (Ji et al. 2016a).
The recessive gene xa5 confers broad resistance spectrum to Xoo and is most commonly found in the Aus-Boro varieties from Bangladesh. The xa5 is a natural allele of Xa5 for the transcription factor IIA gamma subunit 5 (TFIIAγ5), contains a mutation in the 39th residue, in which the valine (V) residue is replaced with glutamine (E) (V39E) . TFIIA is a basal transcription factor of eukaryotes and it is essential for polymerase II-dependent transcription (Høiby et al. 2007). TFIIAγ5 is hijacked by TAL effectors by direct physical interaction with a transcription factor binding (TFB) region of TALEs and attenuate the TALEassociated transcription of host S or R genes (Yuan et al. 2016). The induction of susceptibility genes, such as OsSWEET11 and OsSWEET14 by TALEs are almost abolished in xa5 background or TFIIAγ5-RNAi transgenic plants, which leads to the improvement of BB resistance. TFIIAγ5 is also necessary for TALE-associated transcription of R genes, including Xa27 and Xa23, to defend against disease (Yuan et al. 2016). Xa27 and Xa23-mediated BB resistance are attenuated in the xa5 background (Gu et al. 2009;Yuan et al. 2016). In the absence TFIIAγ5, the other OsTFIIAγ gene in rice, OsT-FIIAγ1 plays a compensatory role. Its expression is activated by TALE PthXo7, which increases expression of the host genes . Interestingly, TFIIAγ5 is also employed by Xoc TALEs to cause disease (Yuan et al. 2016). Mutation and suppression of TFIIAγ5 can also improve BLS resistance. In another study, xa5 was found through genetic mapping as a quantitative trait locus with a relatively large effect for resistance to Xoc (Xie et al. 2014).
The non-host resistance gene, Rxo1 encoding a NLR protein, confers high level resistance to Xoc in rice. It also controls resistance to the pathogen Burkholderia andropogonis, which causes bacterial stripe of sorghum and maize. Transgenic lines with Rxo1 also showed HR when inoculated with avrRxo1 containing Xoc strain (Zhao et al. 2004a;Zhao et al. 2004b;Zhou et al. 2010). It exhibits the characteristics consistent with those mediated by host resistance genes, activating multiple defensive pathways related to HR. A microarray analysis showed that Rxo1 functions in the early stage of rice-Xoc interaction and involved in signaling pathways leading to HR and some basal defensive pathways such as SA and ET pathways .
In addition to the cloned genes above, the dominant R gene Xa7, which has not been isolated yet, is known for its durable resistance and potential value in rice breeding programmes (Vera Cruz et al. 2000). Xa7 was originally identified in rice cultivar DV85 and fine mapped to an interval of approximately 118.5 kb on chromosome 6 . The durable resistance of Xa7 is due to a fitness penalty in Xoo associated with adaptation to Xa7 (Vera Cruz et al. 2000;Bai et al. 2000). Mutations occurred specifically at the avrXa7 gene in the adapted strains, which displayed reduced aggressiveness on susceptible rice cultivars (Vera Cruz et al. 2000;Bai et al. 2000). Additionally, Xa7 are more effective at high temperatures, whereas other R genes are less effective (Webb et al. 2010). Another yet uncharacterized gene Xo1, was identified in the American heirloom rice variety Carolina Gold Select, and confers resistance to the tested African strains of Xoc, but not Asian strains . Like Xa1, Xo1-mediated recognition of full-length TALEs can also be blocked by truncated TALEs (Read et al. 2016). Interestingly, Xa1 and Xo1 are located in the same region . Further studies are needed to determine whether Xo1 is controlled by Xa1 or another gene.

Breeding Strategies to Develop Broad-Spectrum and Durable Resistance to Xoo and Xoc
Use of host plant resistance is generally the most favorable tactic to control diseases due to economic and environmental reasons. Marker-assisted selection (MAS) and genetic transformation are the two major approaches for R gene deployment in plant breeding programs. However, controversy on food safety and constraints on regulatory in some countries have serious plagued the application of genetically modified varieties. MAS, free of political issues and social problems, is more widely used by breeders. Pyramiding R genes resistant to different races of the pathogen through marker-assisted breeding strategies, is a very effective way to achieve durable and broad-spectrum resistance, while employment of a single R gene and adaption of the pathogen often lead to resistance breakdown in a short period.
Based on the previous reports, xa5, Xa7, xa13, Xa21 and Xa23 are more frequently used by rice breeders due to the comparatively broader spectra of resistance. Xu et al. (2012) transferred Xa7 and Xa21 into Yihui 1577, an elite hybrid rice restorer line. The pyramiding lines and their derived hybrids displayed resistance to all the seven Xoo strains, while the lines containing single Xa7 or Xa21 were resistant to six of them. Two Basmati rice varieties PB1121 and PB6 were improved for resistance to BB (xa13 and Xa21) through MAS (Ellur et al. 2016). In another study, three genes, xa5, xa13 and Xa21 were transferred into Lalat, a popular indica variety in Eastern India but susceptible to bacterial blight (Dokku et al. 2013). The improved lines showed significant enhanced resistance.
Because Xa23 displays broadest resistance, it is often used alone, or along with R genes against rice blast disease or/and brown planthopper Huang et al. 2012;Ni et al. 2015;Jiang et al. 2015;Ji et al. 2016b;Xiao et al. 2016). In addition, some gene combinations are ineffective, such as, xa5 + Xa23, xa5 + Xa27 (Gu et al. 2009;Yuan et al. 2016). Therefore, deep understanding the underlying molecular mechanisms of R gene-mediated resistance is important for its effective application. It is noteworthy that no native major R gene effective against Xoc has been discovered so far in rice. The recessive xa5 confers quantitative resistance to BLS, and should be used in combination with other resistance QTL or genes. Fine mapping of previously identified resistance loci with large effect, including the dominant locus Xo1 and the recessive locus bls1, will facilitate employment of them in rice breeding programmes He et al. 2012).
In our breeding practice, we introgressed Xa7 + Xa21 into an elite restorer line R900 of hybrid rice through marker-assisted backcrossing (MABC) scheme in less than 3 years, which is much more efficient than the conventional breeding method. The improved lines recovered more than 99% genome background of the recurrent parent R900, and showed a broad-spectrum resistance to Xoo without any significant difference in main agronomic traits in both the growth chambers and paddy fields (unpublished data). In addition, the R genes can be used separately in time and space. Development of near-isogenic lines and rotation of the R genes could reduce the selection pressure on pathogens and maximize the life span of R genes. Multi-lines containing different R genes also has the potential to provide broad-spectrum and durable disease resistance.
In recent years, the emerging genome-editing technologies, including zinc-finger nucleases (ZFNs), TAL effector nucleases (TALENs) and clustered regularly interspaced short palindromic repeats (CRISPR)/Cas9 (CRISPR-associated protein-9 nuclease), have revolutionized biology by enabling targeted modifications of genomes (Christian et al. 2010;Jinek et al. 2012;Kim et al. 1996). These technologies have been successfully applied in model species Arabidopsis thaliana, Nicotiana benthamiana and multiple crops including rice, wheat, maize, barley, soyben, tomato, potato, citrus, and sorghum (Shah et al. 2018). The powerful tools have great potential in improving the plant disease resistance. Elimination of EBEs in promoters of susceptibility genes or adding EBEs to promoters of executor R genes through genome editing, could enhance the resistance to BLB. For example, the EBEs of AvrXa7 and PthXo3 in the OsSWEET14 promoter were precisely edited by TALENs, which prevents the induction by TALEs. The mutated lines showed strong resistance to both AvrXa7-and PthXo3-dependent Xoo strains . Similarly, the promoter of Xa13 (OsSWEET11) was targeted by CRISPR/Cas9-based disruption, leading to enhanced resistance without affecting rice fertility . In a very recent study, EBEs in the promoters of OsSWEET11, OsSWEET13 and OsSWEET14 were edited simultaneously by CRISPR/Cas9 technology and rice lines conferring broad-spectrum resistance to Xoo were created (Xu et al. 2019). In another study, six EBEs corresponding three TALEs from Xoo and three from Xoc, were added to the Xa27 promoter, resulting in broad-spectrum resistance to both Xoo and Xoc (Hummel et al. 2012). It suggests that engineering of EBEs upstream of rice executor R genes through genome-editing technologies is a potential strategy to generate germplasms with broad-spectrum resistance to Xoo, Xoc and other bacterial pathogens.

Conclusions
Rice-Xanthomonas oryzae patho-system is a powerful model for research toward solutions in disease control.
Although tremendous progress has been made in the past decades, there are still many queries and challenges. For example, whether there is any major R gene in rice against BLS? The xa5 confers resistance to both Xoo and Xoc, does any other identified Xa genes have the same effect? The ligand from Xoo mediating Xa4 resistance is still not determined. The partners and/or components associated with R proteins remain largely unknown in rice. It will be interesting to understand how R genes activate downstream signaling components and trigger plant defense response system. TAL effectors injected into plant cells have to be translocated into nucleus to bind to the target S or R genes for virulence or plant immunity. However, the underlying mechanism needs to be further elucidated. Altogether, a comprehensive understanding of the molecular interactions between rice and Xanthomonas oryzae is the pivotal for more efficient and durable disease control.