TALE-induced immunity against the bacterial blight pathogen Xanthomonas oryzae pv. oryzae in rice

Abstract

Transcription activator-like effectors (TALEs) are proteins produced by plant pathogenic Xanthomonas spp. TALEs exhibit a conserved structure and have the ability to directly bind to the promoter region of host target genes where they activate transcription. TALEs in Xanthomonas oryzae pv. oryzae (Xoo), the causal agent of bacterial blight (BB) in rice, play important roles in triggering resistance (ETI) and susceptibility (ETS) for rice immunity. This review briefly describes rice resistance breeding in China, TALE properties and their roles, BB resistance (R) and susceptibility (S) genes in rice, the arms-race between TALEs and TALE-targets, and strategies for breeding disease-resistant crops. A systematic overview of the complex roles of TALEs are presented along with ongoing efforts to breed crops with durable and broad-spectrum resistance to the pathogenic bacterium.

Background

Rice is one of the most fundamental staple crops worldwide, and provides sustenance and nutrition for over half the global population. Efforts to stabilize rice production is a matter of great urgency in the face of complex environmental challenges and the growing global population. However, bacterial blight (BB), caused by Xanthomonas oryzae pv. oryzae (Xoo), can easily spread in irrigated environments and result in a 10–50% reduction in crop yields (Mansfield et al. 2012). There is a general consensus that breeding rice varieties with durable and broad-spectrum resistance to Xoo is the most effective and environmentally-sustainable strategy to prevent losses caused by BB (Jiang et al. 2020). On the other hand, Xoo may evolve decoys or new virulence effectors to evade resistance gene recognition and suppress the resistance triggered by effectors (ETI), resulting in effector-induced susceptibility (ETS). These decoys or effectors are quite often related with transcription activator-like effectors (TALEs) that are injected into rice plants via the bacterial-encoded type III secretion system (T3SS) (White and Yang 2009; Xu et al. 2017; Timilsina et al. 2020; Xu et al. 2022). In this review, we focus on rice resistance breeding in China, the host-encoded resistance (R) or susceptibility (S) genes to BB, the characteristics and functions of TALE proteins, and recent approaches to breed rice varieties with broad-spectrum BB resistance. Our review concludes with a summary of recent research on TALE-induced immunity to BB and future control strategies.

BB resistance breeding in China

BB is a major disease in rice across the globe, and the first recorded observation was noted in 1884 by Japanese farmers (Ou 1985). Since the 1950s, BB has been reported in most Asian countries including India, Philippines, China, Korea, Thailand, Vietnam, and Sri Lanka where rice is cultivated (Ou 1985). BB has been regarded as one of the most devastating diseases of rice worldwide (Liu et al. 2014; Ji et al. 2018). In China, BB was initially observed in the 1930s in Jiangsu and Zhejiang provinces, and in the 1960s, BB quickly spread in irrigated fields and reduced rice production and grain quality primarily as a result of large-scale cultivation of semi-dwarf and high-yielding rice varieties (Zhang 2009). Over the past decades, intermittent epidemiology of BB on rice have been recorded in 28 provinces of China, especially in south of the Yangtze River and the Jianghuai Plain where climate conditions are more tropical than the northern areas of the country.

Breeding for BB resistance began in the late 1960s in China, and approximately 9.6% of indica rice varieties bred between 1960 and 1970 were resistant to BB, including Shuangzhuzhan that is resistant to both BB and rice blast (Zhang 2007). The R gene Xa4 conveys resistance to the majority of Xoo races, and indica rice varieties such as Zhuzhan, BG90-2, IR20, IR22, IR26, IR30, and IR36 were utilized to develop BB-resistant cultivars (Zhang 2009). The Xa3 resistance gene was primarily used for breeding japonica rice several decades ago (Cao et al. 2007). The wide use of Xa4 and Xa3 in the breeding of hybrid and conventional rice cultivars in China led to a well-controlled BB, and the disease rarely occurred in rice fields for two decades from 1980 to 2000 (Zhang 2009). However, the large-scale cultivation of rice varieties with Xa4 and Xa3 resistance on the other hand promoted the evolution of new pathogenic strains and races of Xoo and contributed to the breakdown of resistance not just in China (Mew et al. 1992; Quibod et al. 2016). Indeed, these cultivation practices have led to outbreaks of BB, and an example is that rice cultivar Xiushui 11 carrying the resistance gene Xa3 is susceptible to BB (Zhang 2007).

Continually in the late 1990s, Xa7, Xa21, xa5, and Xa23 were adopted to enrich the reservoir of resistance genes by molecular marker-assisted selection and transgenic technology (Huang et al. 2012; Luo et al. 2012; Chukwu et al. 2019; Tian et al. 2019; Chen et al. 2021; Joe et al. 2021; Luo et al. 2021; Wei et al. 2021). Although the deployment of these resistance genes initially reduced the occurrence of BB, ongoing changes in climate and continuous evolution of Xoo caused a resurgence of BB in some parts of China. For example, after a large-scale rainstorm in Hunan Province in 2017, BB broke out in paddy fields of approximately 0.667 M hm² (Li et al. 2013b; Chen et al. 2019; Chukwu et al. 2019; Jiang et al. 2020).

Rice cultivars containing single R genes have been planted for long periods of time over large areas, and the resistance conferred by some R genes has been overcome by Xoo strains that evolved during co-evolution with host plants (Wang et al. 2020; Lu et al. 2021). Our laboratory took more than two decades during 2000–2020 to collect Xoo strains from rice-growing areas in China, and the isolates were then inoculated onto near-isogenic lines (NILs) that contain single R genes, including Xa3, Xa4, xa5, Xa7, xa13, and Xa23, to see their compatibility or incompatibility with the NILs. The results clearly showed that rice lines carrying Xa3 or Xa4 were susceptible to multiple Xoo strains (Table 1), indicating that new physiological races or pathotypes of the pathogen have emerged in natural environments. Recently, R genes such as xa5, Xa7, and Xa23 were incorporated into some rice varieties to improve their resistance to BB, however, the resistance has now been overcome by few strains (Table 1), implying that new R genes must be continually incorporated into rice varieties to avoid the long-term and large-scale failure of varieties with single dominant R genes.

Table 1 Pathotypes of Chinese Xoo isolates in near-isogenic rice lines containing different R genes

TALEs determine Xoo virulence in rice

The rice genes conferring resistance and susceptibility can be activated by corresponding avirulence (avr) and virulence (vir) genes derived from the pathogen, respectively (Nino-Liu et al. 2006; Perez-Quintero and Szurek 2019). Since the defense response is genetically stimulated by interaction of avr gene products in Xoo with R gene products in rice, it has been coined gene-for-gene resistance (Flor 1971; Gabriel et al. 1986). Most avr and vir genes from Xoo are TALE genes (Tran et al. 2018; Mucke et al. 2019) with an exception of raxX. It interacts with Xa21 and triggers Xa21-mediated resistance that is regarded as PTI (PAMP-triggered immunity) (Pruitt et al. 2015; Luu et al. 2019; Joe et al. 2021). The elucidated whole-genome sequences of different Xoo races show that 80% of the encoded genes have homology with genes in other Xanthomonas spp., and Xoo-specific genes, particularly TALE genes, are associated with virulence variation (Ochiai et al. 2005).

The highly-conserved TALE proteins, which are injected into host cells by the pathogen via the T3SS (Rossier et al. 1999; Tampakaki et al. 2004; Boch and Bonas 2010), contain a type-III secretion signal at the N-terminus, two or three nuclear localization signals (NLSs), an activation domain (AD), and a transcription factor binding domain (TFB) at the C-terminus. The TFB mediates a complex of TALEs with the plant gamma subunit of basal transcription factor TFIIA, which in turn activates the transcription of TALE-targeted genes in plant host (Yang and White 2004; Yuan et al. 2016). TALEs differ from each other by the number of 33–35 amino acid repeats and the composition of highly variable residues at positions 12 and 13 in each repeat, which are known as repeat variable diresidues (RVDs). The composition of RVDs determines the specificity of DNA binding to the host plant promoter regions called effector-binding elements (EBEs), following the code that confers DNA binding specificity to TALE RVDs. For example, HD, NI, and NG repeats have a strong preference for C, A, and T nucleotides, respectively (Boch et al. 2009; Moscou and Bogdanove 2009). Subsequently, the crystal structure of TALE proteins was determined by experimental and computational predictions and ultimately confirmed the preference of RVDs for specific nucleotides (Deng et al. 2012; Mak et al. 2012). Remarkably, two types of truncated tal genes were characterized in X. oryzae strains, initially assumed as pseudogenes and subsequently confirmed as truncated-TALEs or interfering TALEs (iTALEs). Compared with typical TALEs (tTALEs), iTALEs contain 45 or 129 bp deletions in the sequence encoding the N-terminal region and lack the C-terminal AD domains (Ji et al. 2016). Moreover, tal3a, which encodes a type A iTALE, retains two NLS sequences, whereas tal3b, a type B iTALE, has only one NLS. Experimental analysis confirmed that the unique structural features of iTALEs are essential for their suppressive function on Xa1-mediated resistance triggered by tTALEs (Ji et al. 2016; Read et al. 2016). Currently it has been confirmed that PthXo1, PthXo2, PthXo3, and AvrXa7 are major TALEs of Xoo, and a single major TALE gene contributes more than 80% virulence to the pathogen in rice as measured by lesion length when compared with the full virulence caused by the wild-type strains (White and Yang 2009).

BB susceptibility genes activated by major virulence TALEs

Susceptibility (S) genes are genetically dominant in plants and their expression are induced by pathogen infection. The induction of S genes in turn is beneficial to pathogen nutrient acquisition and disease development. By contrast, the recessive resistance mediated by alleles of S genes, defined as recessive genes, differs from dominant form of resistance to BB in rice. The exocytosis function of SWEET (sugars will eventually be exported transporters) proteins is exploited by Xanthomonas to obtain nutrients required for bacterial colonization (Chen et al. 2010). For BB, TALEs activate not only S genes of the SWEET family (e.g., OsSWEET11, OsSWEET13, and OsSWEET14) but also transcription factor genes (e.g., OsTFIIAγ1, OsTFX1, and OsERF123) in rice (Table 2) (Yang et al. 2006; Sugio et al. 2007; Yu et al. 2011; Streubel et al. 2013; Tran et al. 2018). In the promoters of these S gene, there are EBEs that can be directly recognized and bound by Xoo-specific major virulence TALEs (Mak et al. 2012).

Table 2 Rice genes targeted by TALEs

OsSWEET11, also called Xa13 or Os8N3, is targeted by PthXo1, and this is the first pair to confirm that Xanthomonas induces host susceptibility in a gene-for-gene manner (Yang et al. 2006). SWEET14 (Xa41 or Os11N3) can be targeted by AvrXa7 and PthXo3 from Asian Xoo strains, and TalC and Tal5 from Africa Xoo strains (Antony et al. 2010; Yu et al. 2011; Streubel et al. 2013). The susceptibility gene SWEET13 (Xa25 or Os12N3) can be activated by PthXo2 and PthXo2-like alleles that bind to variable EBEs in the promoter of the same gene (Zhou et al. 2015; Xu et al. 2019), suggesting a co-evolutionary relationship between rice and X. oryzae. SWEET12, SWEET15, and SWEET11b also belong to the third clade of SWEET family that can be induced by designer TALEs, thus causing BB in rice (Li et al. 2013a; Streubel et al. 2013; Wu et al. 2022). This implies that there are possibly uncovered TALEs targeting these S genes naturally.

Regarding rice transcription factors, PthXo6 and PthXo7, derived from PXO99^A strain, target the rice bZIP transcription factor-encoding gene OsTFX1 and the basal transcription factor-encoding gene OsTFIIAγ1, respectively (Sugio et al. 2007), and TalB_MAI1 from Africa strain binds to the promoter of AP2/ERF transcription factor coded by OsERF#123 (Tran et al. 2018).

BB resistance genes capable of trapping Xoo avirulence TALEs

Rice utilizes multiple strategies to trigger defense responses that are intended to ward off pathogens (Ji et al. 2018). Pathogen-associated molecular pattern (PAMP)-triggered immunity (PTI) and effector-triggered immunity (ETI) occur during pathogen infection (Jones and Dangl 2006; Ngou et al. 2022). To date, 47 BB resistance genes have been identified (Jiang et al. 2020; Xing et al. 2021), which can be grouped into the following five categories: receptor-like kinase genes (Xa3/Xa26, Xa4, and Xa21) (Song et al. 1995; Xiang et al. 2006; Hu et al. 2017), EBE-mutated alleles of SWEET genes (xa13, xa25, and xa41) (Chu et al. 2006; Liu et al. 2011; Hutin et al. 2015), executor R genes (Xa7, Xa10, Xa23, and Xa27) (Gu et al. 2005; Tian et al. 2014; Wang et al. 2015; Chen et al. 2021; Luo et al. 2021), NLR genes encoding nucleotide-binding leucine-rich repeat receptors (Xa1, Xa2, Xa14, Xa45(t), Xo1, and Xa47(t)) (Yoshimura et al. 1998; Ji et al. 2020; Zhang et al. 2020; Xing et al. 2021), and basal transcription factor gene xa5 (Iyer and McCouch 2004; Jiang et al. 2006). Xa21 is the first cloned R gene, and the cognate avirulence gene product in the pathogen was ultimately identified as sulfated RaxX (Song et al. 1995; Pruitt et al. 2015). Except for RaxX that interacts with XA21 to trigger PTI, other avirulence genes identified in Xoo are all TALE genes so far as we know (Table 2). Actually the recessive R genes xa13, xa25, and xa41 are unable to cause either ETI or ETS in rice, since the mutated EBE of these genes cannot be recognized by their ligand TALEs, resulting in a failure for the pathogen to colonize host rice (Chu et al. 2006; Liu et al. 2011; Hutin et al. 2015). Later, editing the EBEs of host S genes has been developed into a new breeding strategy aiming at breeding rice varieties with broad-spectrum disease resistance (Oliva et al. 2019; Xu et al. 2019).

Host plants also utilize executor and NLR-type R genes to trigger an effective immune response. Four executor R genes have been identified in rice including Xa7, Xa10, Xa23, and Xa27 (Gu et al. 2005; Tian et al. 2014; Wang et al. 2015; Chen et al. 2021; Luo et al. 2021). Executor R genes encode small, diversiform proteins that are trapped and transcriptionally activated by TALEs (Zhang et al. 2015); these genes share no sequence homology with other known BB resistance genes. The first NLR-type R gene cloned from rice is Xa1, which encodes nucleotide-binding site (NBS) and leucine-rich repeat (LRR) domains (Yoshimura et al. 1998). Xa1-mediated resistance can be triggered by multiple TALE proteins but inhibited by iTALEs (Ji et al. 2016; Read et al. 2016). XA1 is localized to plant nuclei, and Xa1-mediated resistance is independent of the basal transcription factors OsTFIIAγ1 and OsTFIIAγ5 (Xu et al. 2021). Allelic forms of Xa1, namely Xa2, Xa14, Xa45(t), and CGS-Xo1₁₁, have been recently cloned, which are also activated by TALEs and inhibited by iTALEs (Ji et al. 2020; Zhang et al. 2020). Approximately 95% of sequenced Xoo Asian strains contain iTALEs, and consequently, Xa1-type resistance is limited (Ji et al. 2020). Although XA1 interacts in the nucleus with TALEs and a rice transcription factor OsERF101 via the XA1 BED domain to trigger BB resistance (Yoshihisa et al. 2022), how the iTALEs suppress Xa1-mediated defense in rice is still unknown.

Another type of BB resistance genes, the recessive xa5, encodes naturally-occurring V39E variant of the gamma subunit of basal transcription factor (TFIIAγ5, also called Xa5) (Iyer and McCouch 2004; Jiang et al. 2006). Xa5 directly binds to the TFB region of TALEs to form a complex, facilitating the transcription of TALE-activated genes (Yuan et al. 2016). The mutant variant xa5 cannot interact with TALEs or iTALEs, reducing the expression of TALE-driven S or/and E genes (Ma et al. 2018; Xu et al. 2021). However, TALEs can recruit OsTFIIAγ1 to compensate for the absence of OsTFIIAγ5 in rice, explaining the reason that PthXo7-containing strains overcome xa5 resistance in rice (Table 1) (Ma et al. 2018).

Transcriptional complexes of TALEs with rice TFIIAγ

Deciphering the TALE code provides the basis for identifying TALE target genes in rice via computational predictions and experimental analyses (Boch et al. 2009; Mak et al. 2012). Xoo utilizes the T3SS to secrete TALEs, which are subsequently transported into the host nucleus to form a complex by binding with the promoter region of their target genes. This complex is essential for TALE-activated gene expression (Szurek et al. 2001; Perez-Quintero and Szurek 2019). Xa5 is a key component of the eukaryotic transcription complex, which is required for expression of S genes and executor R genes in rice (Yuan et al. 2016; Ma et al. 2018). OsTFIIAγ1 encoded by chromosome 1 is another gamma subunit of the basal transcription factor (Jiang et al. 2006). Consistent with Xa5, TALEs also interact with TFIIAγ1 but show a lower binding affinity. In the absence of Xa5, susceptibility genes were still induced when TFIIAγ1 was activated by PthXo7 (Ma et al. 2018). The complexes formed by TALEs and basal transcription factors (TFIIAγ) in host induce the expression of S genes, leading to disease susceptibility. In other words, genome-editing of these two OsTFIIAγ genes could be used to impair the expression of susceptibility genes (Huang et al. 2017; Ma et al. 2018). Interestingly, TALE-triggered and iTALE-suppressed Xa1 resistance is independent of these two TFIIAγ factors, suggesting a novel mechanism for NLR Xa1-mediated resistance to BB (Xu et al. 2021).

The arms race between TALEs and TALE-targets

The co-evolution of Xoo and rice has led to scenarios where they compete in attempts to co-opt TALEs to combat against each other. Initially, pathogens utilize TALEs to bind to the promoter of S genes, and this is necessary for establishing and stabilizing the parasitic relationship between host and pathogen (Boch et al. 2009; Yu et al. 2011). Host plants may respond by evolving executor R genes and/or modifying TALE-targeted EBEs (e.g., xa13, xa25, and xa41) or eukaryotic basal transcription factors (e.g., xa5) to subvert the pathogen (Jiang et al. 2020). In an effort to evade host resistance, pathogens inject new virulence factors into host cells. New TALEs transcriptionally activate S genes and regain the ability to weaponize TALEs (Streubel et al. 2013). Consequently, TALEs represent a double-edged sword for effector-triggered susceptibility (ETS) and ETI (Fig. 1). The typical example is that two alleles of AvrXa7 and PthXo3 are not only the avirulence factor trapped by the EBE of Xa7 gene for triggering ETI (Chen et al. 2021; Luo et al. 2021) but also the major virulence factors targeting OsSWEET14 for activating ETS (Hutin et al. 2015; Zaka et al. 2018).

Fig. 1

Working model of rice immunity triggered by transcription activator-like effectors (TALEs) and Xops of Xanthomonas oryzae pv. oryzae (Xoo), the causal agent of bacterial blight (BB) in rice. Rice may use executor (E) genes to trap TALEs, or the BED domain of NLR-A type proteins (e.g., XA1 and its alleles) to recognize TALEs, or NLR-B type proteins (e.g., XA47) to recognize unknown type-III secretion effectors (T3SEs) for effector-triggered immunity (ETI) against Xoo. On the other hand, Xoo evolves iTALE to suppress XA1-mediated resistance via an unknown TIP1 in rice. These ETIs may have connection with calcium (Ca²⁺) release and the burst of reactive oxygen species (ROS) for programmed cell death. Possibly, XA47 forms a resistosome like other coiled-coil NLR (CNL) proteins. Basically, Xoo secretes major virulence TALEs via the type-III secretion system (T3SS) to target the EBE (effector-binding element) of susceptibility (S) genes for ETS (effector-triggered susceptibility), and as a result to acquire nutrients from rice. This ETS may require unknown Xops to suppress PAMP-triggered immunity (PTI) via pattern-recognition receptors (PRR) on rice cell membrane. TALEs forms a complex with rice basic transcription factor OsTFIIAγ5 (Xa5) via a TFB domain to precisely activate the expression of E and S genes for ETI and ETS, respectively. If Xa5 mutated to xa5, the ETS turns to resistance by the loss of susceptibility (RLS). In addition, the disruption in EBEs of S genes makes rice acquire RLS. Xop, Xanthomonas outer protein; iTALE, interfering TALE; NLS, nuclear localization signals; AD, activation domain; TFB, transcription factor binding domain; TIP, TALE/iTALE interacting protein; NLR, nucleotide-binding leucine-rich repeat; CC, coiled-coil domain; BED, zinc-finger BED domain; NBS, nucleotide-binding site

The co-evolution between TALEs and their targets in host plants also occurs in nature, resulting in TALE variants and multiple TALE-binding EBEs. For instance, PthXo2, PthXo2.1, and PthXo2.2 from Xoo strains JXO1, PXO339, and PXO163, respectively, bind to the same EBE sequence in OsSWEET13_IR24 (Zhou et al. 2015). Subsequently, it was reported that two other PthXo2-like TALEs, Tal5_LN18 and Tal7_PXO61, also induce the expression of SWEET13 but recognize and bind to different EBEs in the OsSWEET13 promoter, and correspondingly, the host rice has evolved ten EBE variants in the promoter of OsSWEET13 to avoid being recognized by PthXo2-like TALEs (Xu et al. 2019). Similarly, AvrXa23C of seven AvrXa23 alleles is not trapped by the EBE of Xa23, resulting in the loss of Xa23 resistance in rice fields (Xu et al. 2022).

The arms-race battle is also reflected in the activation and suppression of NLR-mediated resistance to BB. For example, typical TALEs (tTALEs) trigger Xa1-mediated resistance, which can be suppressed by iTALEs (Ji et al. 2016; Read et al. 2016). XA1, tTALEs, and iTALEs are localized to the host plant nucleus, suggesting that activation and inhibition of Xa1-mediated resistance occurs in host nuclei (Read et al. 2020; Xu et al. 2021). The contrasting functions of TALEs and their truncated variants (iTALEs) reveal the complex interactions underlying disease resistance and susceptibility in rice.

The interplay between TALEs and non-TALEs is unknown

Xoo type-III secretion effectors (T3SEs) include TALEs and non-TALEs, the latter are defined as Xanthomonas outer proteins (Xops) (Kay and Bonas 2009). Compared with TALEs, the contribution of non-TALEs to host plant susceptibility is poorly understood. In Xanthomonas spp., non-TALEs are involved in the disruption of plant defense signaling and the suppression of PTI (Buttner 2016). For example, XopK interacts and ubiquitinates somatic embryogenic receptor kinase 2 (SERK2) to inhibit PTI upstream of MAPK cascades (Qin et al. 2018). In addition, XopP interacts with the U-box domain of rice ubiquitin ligase PUB44 to inhibit ligase activity and suppress peptidoglycan- and chitin-induced immunity (Ishikawa et al. 2014; Ichimaru et al. 2022); XopY interacts with the rice receptor-like cytoplasmic kinase OsRLCK185 to inhibit rice immunity (Yamaguchi et al. 2013); XopL catalyzes the ubiquitination of NbFd and induces defense-related response in plants (Ma et al. 2020); XopZ, XopN, and XopV suppress the peptidoglycan-triggered PTI response (Long et al. 2018); and XopN, XopQ, XopX, and XopZ suppress cell wall damage-induced immune responses in a functionally redundant manner (Sinha et al. 2013). However, it is still a mystery what are the core Xops necessary for Xoo to cause BB in rice in the presence of a single major virulence TALE.

The manners for rice to perceive TALEs

When TALEs function as transcription activators, rice perceives TALEs in host cells through DNA-protein interactions. For example, TALEs are trapped by executor R and SWEET genes via their EBEs (Zhang et al. 2015; Wu et al. 2022). The crystal structure of TALEs bound to DNA sequences was used to understand how RVDs recognize and bind to different nucleotides. For example, the RVD HD binds adenosine via hydrogen bonding, whereas NI binds thymine via van der Waals forces (Deng et al. 2012). The direct interaction between host gene promoters and TALE can be experimentally validated by gel retardation, chromatin immunoprecipitation, and glucuronidase assays (Romer et al. 2009; Streubel et al. 2013).

Rice plants may also perceive TALEs via NLR proteins and subsequently mount ETI. For example, TALEs containing various RVDs triggered Xa1-mediated resistance in a transcriptional activation-independent manner, while truncated iTALEs suppressed host resistance (Read et al. 2016). The structural differences between tTALEs and iTALEs suggest that the N-terminal deletion and the absence of the AD domain in iTALEs are critical for the activation and suppression of Xa1/Xo1- mediated resistance, respectively. the unique N- and C-terminal structures of iTALEs are essential for overcoming the resistance controlled by Xa1 (Ji et al. 2016), and Xa1-mediated resistance is independent of rice transcription factor TFIIAγ (Xu et al. 2021). It is speculated that resistance mediated by Xa1 and its alleles requires the N- and C-terminal regions of TALEs, rather than the central repeat regions; furthermore, the inhibition of NLR-type resistance by iTALEs supports this speculation. The resistance spectrum of Xa1 and its alleles is greatly reduced by the presence of iTALEs in approximately 95% of Asian Xoo strains (Ji et al. 2020), and XA1 interacts with TALEs and OsERF101 via its BED domain to trigger BB resistance (Yoshihisa et al. 2022), which may be helpful to understand how iTALEs suppress Xa1-mediated resistance. Xa47 represents another type of NLR gene and is predicted to localize to the host cytoplasm (Xing et al. 2021). The structure of Xa47 is very different from that of Xa1 localized to the nucleus (Xu et al. 2021). A recent report showed that a coiled-coil NLR (CNL) in Arabidopsis designated ZAR1 formed a calcium-permeable channel at the plasma membrane in its active oligomeric states and functioned as a resistosome (Bi et al. 2021). This resulted in plant innate immune responses (Fig. 1), including Ca²⁺ ion flux, production of reactive oxygen species (ROS), and cell death (Bi et al. 2021). Recently, another CNL gene Xa47(t) was identified. As a type-B NLR in rice, Xa47(t) confers resistance to Xoo via recognizing a yet unknown effector (Xing et al. 2021).

TALE-trapping EBEs upstream of executor R genes

Among the BB-resistant rice varieties in China, cultivars harboring Xa23 exhibit the widest resistance spectrum (Table 1). Xa23 is an executor-type R gene, and its broad spectrum of resistance imparts great breeding potential (Wang et al. 2020). The promoter regions of executor R genes contain EBEs that can be recognized and bound by TALEs. In rice, Xa7, Xa10, Xa23, and Xa27 are trapped and activated by AvrXa7, AvrXa10, AvrXa23, and AvrXa27, respectively (Jiang et al. 2020; Chen et al. 2021; Luo et al. 2021). A recent study reported that the Xoo strain AH28, which produces the AvrXa23-like effector Tal7b, overcame Xa23 resistance due to the varied RVDs, indicating that a co-evolutionary process between rice and Xoo leads to evasion of host resistance by pathogens (Xu et al. 2022). AvrXa10 was previously shown to transcriptionally activate Xa10 in rice, resulting in a hypersensitive response (HR) and defense (Tian et al. 2014). AvrXa10 also elicited HR in the non-host plant Nicotiana benthamiana by targeting NbZnFP1; when NbZnFP1 was overexpressed in rice protoplasts, cell death was observed (Haq et al. 2022). The rapid and strong defense response mediated by executor R genes indicates that executor (E) genes have great value for disease resistance breeding in rice. For example, the E gene Xa23 was used to confer resistance to the BB-susceptible Nipponbare by deleting EBE_AvrXa23 using CRISPR/Cas9 technology (Wei et al. 2021).

EBE-editing rice provides durable resistance to TALE-containing Xoo

The resistance genes widely used in rice breeding to control BB, such as Xa3, Xa4, xa5, and Xa23, have failed to provide resistance in some rice-growing areas (Li et al. 2020) (Table 1). New Xoo strains have evolved in nature, imposing great challenges to BB-resistant rice cultivars bred in the past several decades. Some combined BB resistance genes, such as xa5 + Xa23 and xa5 + Xa7, were introduced into transgenic rice lines to help alleviate the breakdown of resistance (Chukwu et al. 2019). As mentioned elsewhere in this review, the PthXo7–containing strains without AvrXa23 or AvrXa7 would overcome xa5 + Xa23 or xa5 + Xa7 resistance. Thus, new rice breeding strategies to prevent resistance loss caused by co-evolutionary forces from host and pathogen are urgently needed. Among the tested Xoo strains, TALEs that function as major virulence factors have been identified, including PthXo1, PthXo2-like TALEs, and PthXo3 (AvrXa7) (Oliva et al. 2019), with OsSWEET11, OsSWEET13, and OsSWEET14 as their recognized S genes, respectively. The promoter regions of S genes have been edited in rice to engineer broad-spectrum resistance to BB without introducing R genes (Oliva et al. 2019; Xu et al. 2019). This novel strategy shortens the breeding cycle. In other words, rice lines harboring homozygous mutations in S genes without transgenic elements can be obtained in the T₁ transgenic generation (Deng et al. 2020). Rice plants with triple mutations in SWEET11, SWEET13, and SWEET14 are resistant to all the tested Xoo strains (Oliva et al. 2019; Xu et al. 2019). Thus, the process of engineering resistance in S genes, including SWEET and rice general transcription factor-encoding genes, will increase resistance without introducing R genes and will help clarify the mechanisms underlying S genes.

Another strategy is to utilize gene-editing technology to insert EBE sequences into executor R gene, by which rice cultivars with executor-mediated broad-spectrum resistance were obtained. For example, the EBE of Xa23 was inserted into the susceptible rice cultivar Nipponbare that has no the EBE to trap AvrXa23, and the resulting transgenic line showed resistance to multiple Xoo strains (Wei et al. 2021). When combined with pathogen monitoring and intelligent planning of rice cultivars, this type of disease resistance might be deployed for long periods of time. However, it is important to mention that E gene-mediated resistance could be defeated by variants of trapped TALEs unless the underlying mechanisms on how the executors take functions are deciphered.

Concluding remarks and future perspectives

Xoo and its host rice represent a model system for studying plant-pathogen interactions, by which many innovative results have been obtained. As structurally unique pathogenic effectors, TALEs play a complex role in the co-evolution of Xoo and rice (Perez-Quintero and Szurek 2019). Xoo strains utilize surface-associated virulence factors and non-TALEs secreted via the T3SS to inhibit PTI in rice (Timilsina et al. 2020). TALE proteins are then injected into host cells and bind to EBE sequences at the promoter region of S genes. The subsequent expression of S genes further increases susceptibility to Xoo stains and helps the pathogen obtain nutrients for growth (Li et al. 2004; Yang and White 2004; Boch and Bonas 2010). Rice susceptibility to BB is mainly conferred by the SWEET family genes, which provide sucrose for growth of pathogenic bacteria (Streubel et al. 2013; Eom et al. 2015). Rice plants with genome-edited mutations in the EBEs of SWEET genes are resistant to BB, and this provides a new avenue for breeding BB-resistant rice (Oliva et al. 2019; Xu et al. 2019; Ni et al. 2021).

Of approximately 47 R genes to BB (Jiang et al. 2020; Xing et al. 2021), how the four executor R gene products (XA7, XA10, XA23, and XA27) lead to HR-like programmed cell death in rice is still unclear. We still do not know the avr genes in Xoo that match receptor-like kinases Xa3/Xa26 and Xa4 to trigger rice immunity. There are two types of NLR R genes. Xa1 and alleles Xa2, Xa14, Xa45(t), and CGS-Xo1₁₁ belong to the NLR-A type. Their encoding proteins that contain a BED domain and several leucine-rich repeats (LRR) (93 amino acids in length) are localized in rice nuclei (Read et al. 2020; Xu et al. 2021) (Fig. 1). It is still being explored to know how the resistance conferred by this NLR-A type is suppressed by Xoo iTALEs. XA47 is a NLR-B type protein without the BED domain (Xing et al. 2021) (Fig. 1). If the paired Avr of this type of resistance protein in Xoo is clear, the revealed mechanism of this NLR-type resistance shall be helpful in rice breeding programs.

Researchers across the globe utilize modern technology in an attempt to protect crops from pathogen infection. Breeding strategies aimed at developing rice varieties with durable and broad-spectrum resistance to BB can be divided into two general categories, that is, introducing R genes into cultivated rice varieties and increasing rice resistance to the pathogen by the loss of susceptibility (RLS). Although much has been accomplished in understanding the rice-Xoo interaction, many questions remain unanswered. For instance, is it possible to rearrange potential RVDs of a TALE to simultaneously activate at least two or more R genes in resistant rice varieties with different executor R genes for pyramiding breeding? What is the function of non-TALEs in TALE-triggered susceptibility on rice? Is an unidentified factor (TALE-interacting protein 1, TIP1) required for TALE-triggered and iTALE-suppressed Xa1-mediated resistance? And what is the underlying mechanism of iTALE-suppressed NLR-mediated resistance to BB? The R gene product XA1 could be recognized by multiple typical TALEs with diverse RVDs, which could attenuate the effect of TALE diversity and reduce the possibility of breakthrough resistance. The design of new effector recognition specificities in R genes, especially NLRs, via molecular engineering of integrated decoy domains has been reported to confer resistance to rice blast (Cesari et al. 2022) and might be applicable to BB. In summary, editing of the EBE of three S genes and the NLR genes have great potential in breeding rice with durable and broad-spectrum resistance to TALE-containing pathogens.