Tal6b/AvrXa27A, a hidden TALE targeting the susceptibility gene OsSWEET11a and the resistance gene Xa27 in rice

Xanthomonas oryzae pv. oryzae (Xoo) secretes transcription activator-like effectors (TALEs) to activate rice susceptibility (S) genes, causing bacterial blight (BB), as well as resistance (R) genes, leading to defense against BB. This activation follows a gene-for-gene paradigm that results in an arms race between the TALE of the pathogen and effector-binding elements (EBEs) in the promoters of host genes. In this study, we characterized a novel TALE, designated Tal6b/AvrXa27A, that activates the rice S gene OsSWEET11a and the rice R gene Xa27. Tal6b/AvrXa27A is a member of the AvrXa27/TalAO class and contains 16 repeat variable diresidues (RVDs); one RVD is altered and one is deleted in Tal6b/AvrXa27A compared with AvrXa27, a known avirulence (avr) effector of Xa27. Tal6b/AvrXa27A can transcriptionally activate the expression of Xa27 and OsSWEET11a via EBEs in their corresponding promoters, leading to effector-triggered immunity and susceptibility, respectively. The 16 RVDs in Tal6b/AvrXa27A have no obvious similarity to the 24 RVDs in the effector PthXo1, but EBETal6b and EBEPthXo1 are overlapped in the OsSWEET11a promoter. Tal6b/AvrXa27A is prevalent among Asian Xoo isolates, but PthXo1 has only been reported in the Philippine strain PXO99A. Genome editing of EBETal6b in the OsSWEET11a promoter further confirmed the requirement for OsSWEET11a expression in Tal6b/AvrXa27A-dependent susceptibility to Xoo. Moreover, Tal6b/AvrXa27A resulted in higher transcription of Xa27 than of OsSWEET11a, which led to a strong, rapid resistance response that blocked disease development. These findings suggest that Tal6b/AvrXa27A has a dual function: triggering resistance by activating Xa27 gene expression as an avirulence factor and inducing transcription of the S gene OsSWEET11a, resulting in virulence. Intriguingly, Tal6b/AvrXa27A, but not AvrXa27, can bind to the promoter of OsSWEET11a. The underlying recognition mechanism for this binding remains unclear but appears to deviate from the currently accepted TALE code.


INTRODUCTION
Effectors are proteins secreted by pathogens into or near host cells that increase pathogen virulence (Koseoglou et al., 2022).Effectors can interfere with pathogen-associated molecular patterntriggered immunity, which is the first layer of the plant innate immune system; this interference can result in effector-triggered susceptibility (ETS) (Jones and Dangl, 2006).In some cases, effectors are intercepted by host resistance (R) genes that activate effector-triggered immunity (ETI) (Jones and Dangl, 2006;Zhou and Zhang, 2020).Transcription activator-like effectors (TALEs) are translocated into host plants via the type III secretion system in phytopathogenic bacteria of the genera Xanthomonas and Ralstonia (Boch and Bonas, 2010;Boch et al., 2014;Xu et al., 2017).After injection into plant cells, TALEs bind to EBEs (effector-binding elements), which are sequences in the promoters of host genes; activation of these genes can reprogram the host transcriptome (Boch et al., 2014).The specificity of EBE binding is determined by repeat-variable diresidues (RVDs) that are located at amino acid residues 12 and 13 in the central repeat regions (CRRs) of TALEs (Boch et al., 2009;Moscou and Bogdanove, 2009).Genes targeted and activated by TALEs include host susceptibility (S) and R genes (Yang and White, 2004;White and Yang, 2009;Zhang et al., 2015).
Among the TALEs identified in Xoo, AvrXa7 and PthXo3 are unique because they have dual functions as virulence and avirulence factors; both induce the S gene SWEET14 (leading to virulence and disease) and trigger resistance in Xa7-containing rice (Antony et al., 2010;R€ omer et al., 2010;Chen et al., 2021;Luo et al., 2021).The EBEs that recognize AvrXa7 and PthXo3 in the Xa7 promoter exhibit partial overlap, which is also true for corresponding EBEs in the OsSWEET14 promoter (Chen et al., 2021;Luo et al., 2021).Xa7-mediated BB resistance is considered to be broad and durable because ETI will be triggered by Xoo strains harboring either AvrXa7 or PthXo3; moreover, these two TALEs also function as major virulence determinants, and their loss penalizes the pathogen with reduced virulence and fitness (Zhao and Zhang, 2021).Consequently, Xa7 is regarded as a valuable gene for breeding BB-resistant rice.It remains unclear whether other R genes, such as Xa27, can be activated by (A) Disease symptoms on rice cultivars IR24 and 78-15 after inoculation with Xoo strains PXO99 A , PH, LN18, and LN4.The images were taken 14 days post-inoculation (dpi).(B) Boxplots of mean disease lesion lengths (cm) on cvs.IR24 and 78-15.Lesions were measured at 14 dpi; dots denote individual observations from at least 5 inoculated leaves, and whiskers display the first and third quartiles, split by the median.Values with the same lowercase letters do not differ significantly at P < 0.05 according to ANOVA.(C) Genomic digest of tal genes in Xoo strains PXO99 A , LN18, and LN4 as determined by Southern blot analysis.Genomic DNA was digested with BamHI, blotted onto nylon membranes, and hybridized with the SphI fragment from avrXa27.
cognate avirulence factors that have dual functions and also induce S genes.
In the current study, we show that Xoo strains LN18 and LN4, which lack avrXa27, remain incompatible with the rice cultivar (cv.) 78-15, which harbors Xa27.The genomic sequences of LN18 and LN4 revealed that both encode a TALE from the AvrXa27/TalAO class (designated Tal6b/AvrXa27A) that triggers Xa27-mediated resistance.Four variants of TALEs in the AvrXa27/TalAO class are present in Xoo strains and function as avirulence factors that mediate Xa27 resistance.Tal6b/ AvrXa27A also activated OsSWEET11a expression and enhanced susceptibility to Xoo in rice.EBE prediction and gel shift assays revealed that Tal6b/AvrXa27A binds directly to the Xa27 and OsSWEET11a promoters.An EBE Tal6b mutation in the OsSWEET11a promoter rendered rice susceptible to Xoo strains harboring Tal6b/AvrXa27A.Our results indicate that Tal6b/AvrXa27A recognizes the R gene Xa27 and the S gene OsSWEET11a and suggest that Xa27 may mediate broadspectrum durable resistance to BB in a manner analogous to Xa7.

RESULTS
Xa27 rice is resistant to Xoo strains LN18 and LN4, which do not contain avrXa27 The Xa27 gene is a broad-spectrum R gene that confers resistance to 27 of 35 Xoo strains collected from 11 countries (Gu et al., 2004;2005).To determine whether Xa27 confers resistance to the hypervirulent Xoo strains LN18 and LN4 (A) Repeat variable diresidues (RVDs) in AvrXa27 and Tal6b.Single letters represent amino acids at the 12th and 13th positions of individual repeats.The asterisk indicates a predicted missing 13th residue, and RVDs in red font differ between Tal6b and AvrXa27.(B) Reaction of IR24 and 78-15 rice leaves to Xoo strains PH/ev, PH/avrXa27, and PH/tal6b.Bacterial suspensions were infiltrated into rice leaves using needleless syringes, and leaves were photographed at 3 dpi.(C) Expression of Xa27 in 78-15 rice leaves inoculated with Xoo strains.The expression of Xa27 was measured in leaves infiltrated with Xoo PH/ev (control), PH/avrXa27, and PH/tal6b.RNA was extracted from leaves at 24 hpi and used for qRT-PCR with Xa27-specific primers.Actin expression was used as an internal control.Columns labeled with two asterisks are significantly different at P < 0.01 (Student's t-test).(Xu et al., 2019;2020), rice cvs.78-15 (with Xa27) and IR24 (without Xa27) were inoculated with the two strains by the tipcutting method.Lesion lengths and symptoms on 78-15 and IR24 leaves inoculated with Xoo LN18 and LN4 were similar to those elicited by Xoo PXO99 A , which contains avrXa27 (Figure 1A and 1B).Our results indicated that Xoo LN18 and LN4 are incompatible with rice cv.78-15.
To determine whether Xoo LN18 and LN4 contain avrXa27 (an avr gene that recognizes Xa27), restriction digests of tal genes in the two strains were hybridized with a probe that encompassed the CRR of avrXa27 (GenBank: AY986494).The fragment containing avrXa27 was absent from Southern blots of Xoo LN18 and LN4 (Figure 1C), suggesting that these two strains do not contain avrXa27.
The incompatibility of Xoo LN18 and LN4 with rice cv.78-15 (Xa27) (Figure 1A and 1B) and the lack of an avrXa27hybridizing band in these two strains (Figure 1C) prompted us to speculate that LN18 and LN4 may harbor a functional equivalent of avrXa27.Using the LN18 and LN4 genome sequences (Xu et al., 2019;2020), we identified two orthologs of avrXa27, designated tal6b and tal1b (Supplemental Figure 14), in LN18 and LN4, respectively (Figure 1C).Although the PXO99 A strain also contained a band similar to tal6b and tal1b in the Southern blot (Figure 1C), this band contained BamHI fragments of two tal genes, tal4 and tal5a, which are quite different from avrXa27 (Salzberg et al., 2008).Sequence analysis of tal6b from Xoo LN18 indicated that the gene comprises 3312 bp with 95.8% nucleic acid identity to avrXa27 from PXO99 A , indicating that tal6b encodes an AvrXa27-like TALE (designated Tal6b) of 1103 amino acid residues with 16 RVD repeats (Figure 2A).Differences occurred at the 12th RVD, where N* in AvrXa27 is NG in Tal6b; furthermore, the 17th RVD is missing in Tal6b (Figure 2A).The RVD array of Tal1b in LN4 is identical to that of Tal6b (Xu et al., 2020).These results indicate that Tal6b is a functional equivalent of AvrXa27; hence, we refer to Tal6b as AvrXa27A in this study.
Tal6b/AvrXa27A, similar to AvrXa27, triggers Xa27 resistance To verify that Tal6b is a variant of AvrXa27, the isolated tal6b was subcloned as plasmid pHZW-tal6b (Supplemental Table 1); this construct was transferred into Xoo strain PH, a tal-free derivative of PXO99 A (Ji et al., 2016), resulting in the Xoo strain PH/tal6b (Supplemental Table 1).The Xoo PH/avrXa27 strain was used as a control (Supplemental Table 1).Western blot analysis showed that Tal6b and AvrXa27 were detected as 120-kDa proteins in the corresponding Xoo strains (Supplemental Figure 1).Xoo strains PH/tal6b, PH/avrXa27, and PH/ev were infiltrated into seedling leaves of rice cv.78-15 using needleless syringes and also inoculated by the tip-cutting method.PH/tal6b, which contains a derivative of avrXa27, elicited a hypersensitive response (HR) in rice cv.78-15, as did AvrXa27 (Figure 2B).Lesion lengths on leaves of rice cv.78-15 inoculated with PH/tal6b did not differ significantly from those on leaves inoculated with PH/avrXa27 (Supplemental Figure 2), implying that Tal6b is a functional equivalent of AvrXa27.
Gene expression assays were used to determine whether Tal6b (AvrXa27A) could activate Xa27 expression in 78-15 rice.Expression of Xa27 was significantly higher in 78-15 rice leaves inoculated with Xoo PH/avrXa27 and PH/tal6b than in leaves infiltrated with Xoo PH/ev (control) (Figure 2C).Collectively, these results suggest that tal6b functions as a derivative of avrXa27.

Prevalence of the AvrXa27/TalAO class in Xoo strains
The AvrXa27/TalAO class has been categorized previously as a ''core TALE'' class that is present in over 80% of Asian Xoo strains (M€ ucke et al., 2019).We investigated the distribution of avrXa27 and homologs by performing BLAST searches of the complete genome sequences of 80 Xoo strains in the NCBI database.These strains originated from Asia (47 strains), Africa (31), South America (1), and Oceania (1) and represent diverse geographic areas.Our results showed that 36 Xoo strains contained TALEs of the AvrXa27/TalAO class, whereas 44 strains lacked AvrXa27/TalAO orthologs, including all African strains and 13 Asian strains (Supplemental Table 2).AvrXa27/TalAO class TALEs in these 36 Xoo strains were grouped into 4 versions based on their RVD sequences (Table 1): (1) the AvrXa27 prototype in 12 Asian and 1 South American strain; (2) the AvrXa27A type in 18 Asian strains, including LN18 and LN4, in which the 12th RVD (NG) is altered and the 17th RVD is missing; (3) the Avr-Xa27B type, as in strain HuN37, in which the 9th (NN) and the 12th (NG) RVDs are altered and the 17th RVD is missing; and (4) the AvrXa27C type in 2 Indian, 1 Thai, and 1 Australian strain, in which the 12th (NG), 16th (HG), and 17th (N*) RVDs are altered (Table 1).These RVD differences illustrate the diversity in the AvrXa27/ TalAO class of TALEs, which is likely due to evolutionary pressure between the avrXa27 locus and the cognate Xa27 R gene.Moreover, AvrXa27A appears to have a prominent role in the AvrXa27/ TalAO class because it is present in 48.65% of 37 Xoo strains.
Xoo strains PXO99 A , HuN37, and IXO221, containing AvrXa27, Tal17/AvrXa27B, and Tal6c/AvrXa27C, respectively, were chosen for further evaluation of their interactions with Xa27.Inoculation assays showed that Xoo PXO99 A , HuN37, and IXO221 were avirulent on rice cv.78-15 containing Xa27, which indicates that Tal17/ AvrXa27B and Tal6c/AvrXa27C may also trigger Xa27 resistance (Supplemental Figure 3).To evaluate this possibility, tal17 and tal6c were cloned from Xoo HuN37 and IXO221, respectively, and introduced into Xoo PH.The resulting strains, Xoo PH/tal17 and PH/tal6c, both elicited an HR on rice 78-15 (Supplemental Figure 4), which indicates that Tal17 and Tal6c are orthologs of AvrXa27 and interact with Xa27.In addition, qRT-PCR assays revealed a significant upregulation of Xa27 gene expression in rice inoculated with both PH/tal17 and PH/tal6c strains compared with the control (Supplemental Figure 5), confirming that Tal17/ AvrXa27B and Tal6c/AvrXa27C can activate the expression of the Xa27 gene.Thus, all four members of the AvrXa27/TalAO class triggered ETI via Xa27, which helps explain the broad-spectrum resistance of the Xa27 executor R gene.

Plant Communications
Tal6b/AvrXa27A is a novel virulence factor for BB Introduction of tal6b into Xoo PH resulted in a high level of virulence on rice cv.IR24 (Supplemental Figure 2), indicating that Tal6b/AvrXa27A may also be a virulence factor.To determine whether other TALEs in the AvrXa27/TalAO class contribute to virulence, PH strains expressing the four avrXa27 variants were inoculated onto Nipponbare and Kitaake rice using the tip-cutting method.Only talb6 (avrXa27A variant) rendered PH highly virulent on Nipponbare and Kitaake at 14 days post inoculation (dpi) (Figure 3A and 3B).PH/tal6b induced BB lesions exceeding 10.0 cm in length on Nipponbare and Kitaake rice; the other three forms of avrXa27 induced small lesions less than 3.0 cm in length (Figure 3B).Thus, our results indicate that Tal6b/AvrXa27A functions as a virulence factor in Xoo.
The CRRs in TALEs are known to determine their specificity (Boch et al., 2009;Moscou and Bogdanove, 2009); we therefore constructed deletion derivatives of tal6b that lacked various repeats to further validate the function of Tal6b as a virulence factor.Derivatives Tal6b-2 and Tal6b-6 lacked repeats 1-9 and 1-12, respectively (Figure 3C and 3D).When these derivatives were introduced into Xoo, strains PH/tal6b-2 and PH/tal6b-6 lost their ability to induce disease symptoms in Nipponbare and Kitaake rice (Figure 3A and 3B).Overall, our data suggested that Tal6b/AvrXa27A was a novel virulence factor targeting an unknown S gene in rice.
Tal6b/AvrXa27A targets the well-known S gene OsSWEET11a We previously demonstrated that rice lines harboring triple mutations in the promoter regions of three S genes (OsS-WEET11a, OsSWEET13, and OsSWEET14) exhibited broadspectrum resistance against all tested Xoo strains, including LN18, which contains tal6b (Xu et al., 2019).To understand how Tal6b/AvrXa27A promotes disease, we sought to identify its S gene target.We speculated that Tal6b/AvrXa27A might target one or more of the three OsSWEET genes in rice.To test this hypothesis, rice Kitaake leaves were infiltrated with PH/ tal6b and PH/ev, which carried tal6b/avrXa27A and an empty vector, respectively.The expression levels of OsSWEET11a were hundreds of times higher in rice plants infected with PH/ tal6b than PH/ev, whereas OsSWEET13 and OsSWEET14 were not induced by Xoo PH/tal6b (Figure 3E); this suggested that Tal6b/AvrXa27 could transcriptionally activate the S gene OsSWEET11a.These findings suggested that Tal6b/AvrXa27A has a dual function; in other words, it triggers resistance by interacting with Xa27 as an avirulence factor, and it induces transcription of the S gene OsSWEET11a, resulting in virulence.
Tal6b/AvrXa27A binds to the promoters of Xa27 and OsSWEET11a To analyze the Tal6b/AvrXa27A inducibility of the OsSWEET11a and Xa27 promoters, we cloned the promoter regions of OsSWEET11a and Xa27 upstream of a gusA reporter gene in pCAMBIA1381.The two promoter::GUS fusions were individually co-expressed in Nicotiana benthamiana with pHB-tal6b by Agrobacterium-mediated transformation.GUS (b-glucuronidase) activities revealed that the promoters of OsSWEET11a and Xa27 were responsive to Tal6b/AvrXa27A (Supplemental Figure 6).
Activation of target genes by TALEs is dependent on the presence of an EBE in the target gene promoter (Boch et al., 2009;Moscou and Bogdanove, 2009); we therefore searched for possible EBEs recognizing Tal6b/AvrXa27A in the promoters of Xa27 (designated EBE AvrXa27A ) and OsSWEET11a (designated EBE Tal6b ) using the programs TALE-NT and TALgetter (Figure 3G and Supplemental Figure 7; Doyle et al., 2012;Grau et al., 2013).
The putative EBE AvrXa27A in Xa27 is a 16-bp sequence located 86-71 bp upstream of the Xa27 ATG (Supplemental Figure 8).EBE AvrXa27A overlaps with the sequence of EBE AvrXa27 , and the predicted binding scores are similar (Supplemental Figure 8) (R€ omer et al., 2009).Six EBEs for Tal6b/AvrXa27A were predicted near EBE PthXo1 ; these are located À256 to À223 bp upstream of the transcriptional start site of OsSWEET11a and have similar TALE-NT scores (Figure 3G).
We next investigated whether Tal6b/AvrXa27A binds to the predicted EBEs of Xa27 and OsSWEET11a by electrophoretic

Plant Communications
Tal6b targets OsSWEET11a and Xa27 mobility shift assays (EMSAs).The His-tagged fusion proteins His-Tal6b/AvrXa27A, His-AvrXa27, and His-PthXo1 were overproduced in Escherichia coli and purified.A 34-bp double-stranded DNA fragment containing EBE AvrXa27A (Xa27p) was used to test the interaction of Xa27 with His-Tal6b/ AvrXa27A (Figure 4A).EMSA results indicated that Tal6b/ AvrXa27A binds to EBE AvrXa27A (Figure 4B).Detection of labeled Xa27p was reduced by addition of unlabeled Xa27p (Figure 4B), demonstrating that this binding was specific.Furthermore, we also identified putative EBEs for the other two types of AvrXa27/TalAO-class TALEs in Xa27 (Supplemental Figure 8) and verified them (Supplemental Figure 9).
EMSAs confirmed that other members of the AvrXa27/TalAO-class TALEs, including AvrXa27, Tal17/AvrXa27B, and Tal6c/AvrXa27C, do not bind to EBE Tal6b in the OsSWEET11a promoter (Supplemental Figure 11).Given that there are only two RVD differences between AvrXa27 and Tal6b/AvrXa27A (Figure 2A), we wondered whether these two RVDs affect the binding of AvrXa27 to EBEtal6b in the OsSWEET11a promoter.We therefore mutated 2 nt in the OsSWEET11a promoter (Supplemental Figure 12A) to match the two polymorphic RVDs in AvrXa27 according to the TALE code (Boch et al., 2009;Moscou and Bogdanove, 2009) and used the resulting probe (Xa13p-14) for EMSA.The results indicated that AvrXa27 was still unable to bind the mutated promoter fragment of OsSWEET11a (Supplemental Figure 12B).
OsSWEET11a promoter mutations confer Tal6b/ AvrXa27A-dependent resistance to Xoo We further investigated the requirement for the OsSWEET11a EBE Tal6b in disease susceptibility in the presence of Tal6b/ AvrXa27A by generating mutations in the OsSWEET11a EBE using CRISPR technology.Three homozygous rice lines (T 2 generation) were obtained in rice cv.Kitaake and named MS13K-14, MS13K-16, and MS13K-19.Rice line MS1K-14 carried a thymine insertion, whereas the other two lines carried 14-and 10-bp deletions, respectively (Figure 5A).

DISCUSSION
In this study, we showed that Tal6b/AvrXa27A, an AvrXa27-like TALE in Xoo, functions as an avirulence factor to trigger Xa27mediated resistance in rice.Tal6b/AvrXa27A was also shown to function as a virulence factor that targets the susceptibility gene OsSWEET11a.This new TALE targets R and S genes for BB by binding to the EBEs in the promoters of Xa27 and OsSWEET11a, respectively.
Rice Xa27 is an executor gene that is specifically activated by the cognate avirulence TALE AvrXa27; this TALE binds to the 16-bp EBE AvrXa27 sequence present in the Xa27 promoter (Gu et al., 2005;R€ omer et al., 2009).In this study, we analyzed the molecular diversity of avrXa27 and identified four variants of AvrXa27/TalAO-class TALEs in 37 different Xoo strains (Table 1).In addition to the parental TALE, AvrXa27, Xoo strains may produce AvrXa27A, AvrXa27B, or AvrXa27C; these also trigger Xa27-mediated ETI (Supplemental Figures 4 and 5) and contain several RVDs that differ from those in AvrXa27 (Table 1).These four AvrXa27/TalAO-class TALEs were not present in the genomes of 13 Asian Xoo strains or accessible African strains (Supplemental Table 2).The Asian strain AH28 was originally isolated from the Anhui province of China (Xu et al., 2022; Supplemental Table 2) and lacks TALEs of the AvrXa27/ TalAO class (Supplemental Table 2).Xoo AH28 was virulent on rice cv.78-15, which contains Xa27 (Supplemental Figure 3), suggesting that Xoo can evolve to overcome Xa27-mediated resistance by disposing of the corresponding avr gene.
OsSWEET11a, a member of the clade III sugar transporters in rice, is a known S gene targeted by the major TALE PthXo1 (Yang et al., 2006;Streubel et al., 2013).Our data showed that Tal6b/AvrXa27A is a novel major TALE that also activates OsSWEET11a expression (Figure 3E and 3F).Tal6b/AvrXa27A targets an EBE in OsSWEET11a that overlaps with the EBE of PthXo1 and harbors a completely different RVD array compared with that in PthXo1 (Figures 3G, 4E, and 4F).Similarly, the promoters of OsSWEET14 and OsTFX1 are targeted by several unrelated TALEs (Sugio et al., 2007;Antony et al., 2010;Yu et al., 2011;Streubel et al., 2013;Tran et al., 2018); thus, OsSWEET11a represents a new case of functional convergence in which two completely unrelated TALEs from different strains have evolved to bind overlapping EBEs.
It is widely accepted that plants mount a defense response against pathogens by tricking them into inducing resistance via E genes, which are dominant R genes (Hutin et al., 2015a;Timilsina et al., 2020).Four E genes, Xa27, Xa10, Xa23, and Xa7, have been cloned from rice (Gu et al., 2005;Tian et al., 2014;Wang et al., 2015;Chen et al., 2021;Luo et al., 2021).Xa7 is activated by the TALEs AvrXa7 and PthXo3, which also target the rice susceptibility gene OsSWEET14 (Chen et al., 2021;Luo et al., 2021).The Xa7 EBE mimics the OsSWEET14 EBE, which protects OsSWEET14 against exploitation by Xoo by triggering cell death (Luo et al., 2021).In this study, we identified another TALE, Tal6b/ AvrXa27A, that targets E and S genes in rice (Figure 4).Unlike those of AvrXa7 and PthXo3, the EBE of Xa27 for Tal6b/ AvrXa27A (EBE AvrXa27A ) bears no obvious similarities to the corresponding EBE of SWEET11a (EBE Tal6b ) (Figure 4), implying that the Xa27 EBE may not be a mimic of the OsSWEET11a EBE.Expression analysis showed that transcription of Xa27 is higher than expression of OsSWEET11a in rice cv.78-15 when induced (E) Boxplots of mean lesion lengths (cm) on Kitaake and MS1K-14 rice.Lesions were measured at 14 dpi.Values with identical lowercase letters do not differ significantly at P < 0.05 according to ANOVA.
Tal6b targets OsSWEET11a and Xa27 Plant Communications by Tal6b/AvrXa27A (Supplemental Figure 13).These results suggest that Tal6b/AvrXa27 has a greater capacity for inducing Xa27 than OsSWEET11a, resulting in a disease-resistant phenotype.Because the co-evolution of AvrXa27/TalAO class TALE-Xa27-OsSWEET11a is not clear, further research is needed to confirm the hypothesis that Xa27 evolved a promoter mimic that triggers a defense response and confounds the desired effects of the pathogen (Gu et al., 2005).
All four known types of AvrXa27/TalAO-class TALEs demonstrated binding to the respective EBEs in the Xa27 promoter, triggering Xa27-mediated ETI, regardless of differences in their RVDs (Figures 2 and 5 and Supplemental Figures 4 and 9).However, among these TALEs, only Tal6b/AvrXa27A exhibited binding to EBE Tal6b in the OsSWEET11a promoter and induced OsSWEET11a-mediated ETS (Figures 3 and 5 and Supplemental Figure 11).Tal6b/AvrXa27A shares only one to two RVD differences with other AvrXa27/TalAO-class TALEs (Table 1).The distinct functions displayed by Tal6b/AvrXa27A in activating both an R gene and an S gene indicate the critical role of these differences in determining specific target recognition and activation abilities.Interestingly, despite mutations of the nucleotides that correspond to the polymorphic RVDs between AvrXa27 and Tal6b/AvrXa27A, AvrXa27 still failed to bind to the mutated promoter fragment of OsSWEET11a (Supplemental Figure 12).This result suggests that observed differences in activity between AvrXa27 and Tal6b/AvrXa27A cannot be attributed solely to these specific nucleotide changes.Notably, the matches between the RVDs of Tal6b/AvrXa27A and EBE Tal6b identified in our study do not strictly adhere to the TALE code, because only two RVDnucleotide pairs exhibit the best match (Supplemental Figure 7).This finding implies the limitations of the current version of the TALE code.To gain a deeper understanding of these functional differences, further investigations, such as structural analyses and mutagenesis studies, are warranted to reveal the molecular basis for the unique properties of Tal6b/AvrXa27A.These studies will contribute to a comprehensive understanding of TALEmediated host-pathogen interactions and the mechanisms that govern target recognition and activation.
Combining our data with those reported previously (Gu et al., 2005;Yang et al., 2006), we propose a working model for the roles of Tal6b/AvrXa27A in Xoo and rice (Figure 6).In this model, Xoo strains such as PXO99 A utilize the major TALE PthXo1 to target the S gene OsSWEET11a, which leads to ETS and virulence in rice.Xoo strains that harbor avrXa27 (e.g., strain PXO86) induce Xa27-mediated resistance via AvrXa27 in rice cultivars that harbor Xa27.Finally, tal6b/avrXa27A-containing Xoo strains (e.g., strain LN18) secrete the Tal6b/AvrXa27A effector, which induces ETS by targeting OsSWEET11a in rice lines that lack Xa27.However, in rice lines that contain Xa27, Tal6b/AvrXa27A is recognized by Xa27 and activates a high level of Xa27 expression.This results in Xa27-mediated immunity that suppresses OsSWEET11a-mediated susceptibility (Figure 6).However, the mechanistic basis of Xa27-mediated ETI, OsSWEET11a-mediated ETS, and their crosstalk remains unclear and warrants further research.
This study and our previous findings (Xu et al., 2019) indicate that Xoo LN18 encodes three major TALEs (Tal5 LN18 , AvrXa7, and Tal6b/AvrXa27A), which target the S genes OsSWEET13, Os-SWEET14, and OsSWEET11a, respectively.To our knowledge, this is the first example of a Xoo strain encoding three major TALEs that each target a known SWEET gene and may constitute a robust adaptation to overcome host loss-of-susceptibility recessive resistance due to xa13, xa25, and xa41 (Yang et al., 2006;Hutin et al., 2015b;Zhou et al., 2015;Xu et al., 2019).Thus, Xoo LN18 may represent an emerging, highly virulent Xoo population that should be monitored by genome sequencing and TALome analysis.It is pthXo1-containing Xoo strains employ the major TALE PthXo1 to specifically target the S gene OsSWEET11a, leading to ETS in rice.avrXa27-containing Xoo strains induce Xa27-mediated resistance via AvrXa27 in rice cultivars that harbor Xa27.Conversely, tal6b/avrXa27A-containing Xoo strains secrete Tal6b/AvrXa27A, which targets OsSWEET11a and Xa27.This results in Xa27-mediated ETI that suppresses OsSWEET11a-mediated ETS.

Plant Communications
Tal6b targets OsSWEET11a and Xa27 important to mention that rice lines with three edited OsSWEET EBEs showed resistance to Xoo LN18 (Xu et al., 2019), demonstrating that engineering resistance by exploiting EBEs in OsSWEET genes may be a promising approach for dealing with this rapidly evolving pathogen.

METHODS
Plant materials, bacterial strains, and growth conditions Rice cultivars IR24, Nipponbare, Kitaake, and 78-15 were grown in field plots and greenhouses at Shanghai Jiao Tong University (Shanghai, China).The bacterial strains used in this study are listed in Supplemental Table 1.E. coli strains were grown in Luria-Bertani medium at 37 C. Xoo strains were grown in nutrient broth (NB) or NB supplemented with 1.5% agar at 28 C. Antibiotics were used at the following concentrations when required: ampicillin (100 mg/ml), kanamycin (25 mg/ml), and spectinomycin (50 mg/ml).

Pathogen inoculation assays
Xoo strains were cultured in NB supplemented with appropriate antibiotics at 28 C for 20 h.Bacterial suspensions (optical density 600 [OD 600 ] = 0.8) were used to inoculate 2-month-old rice plants by the tipcutting method.Disease symptoms were recorded at 14 dpi, and lesion lengths (cm) were measured.For observation of water soaking and HR, strains (OD 600 = 0.6) were infiltrated into 2-week-old rice seedlings with needleless syringes, and symptoms were recorded 3 days after infiltration.Five leaves were inoculated with each Xoo strain, and experiments were repeated three times.

Southern blotting
For tal gene detection, Xoo genomic DNA was extracted, digested with BamHI, separated on agarose gels, and transferred to membranes for blotting as reported previously (Xu et al., 2019).The probe was made from a digoxigenin-labeled DNA fragment that contained the repetitive region of avrXa27.Bacterial genomic DNA was isolated with the Axygen Bacterial Genomic DNA Miniprep Kit.Restriction endonucleases and DNA molecular weight markers were provided by TaKaRa Bio (Japan).Digoxigenin-labeled Southern blotting kits were purchased from Roche (Switzerland), and Immobilon-Ny + membranes were supplied by Millipore (USA).
tal gene cloning and plasmid construction tal17 of Xoo HuN37 and tal6c of Xoo IXO221 were isolated as described previously (Xu et al., 2019).In brief, genomic DNA (2-3 mg) of HuN37 and IXO221 was digested with BamHI (New England Biolabs, USA) and separated on 1.3% agarose gels.DNA fragments were excised, ligated into BamHI-digested pBluescript II KS(À), and transferred into E. coli DH5a.The resulting plasmid library was screened for TALE-containing clones by in situ colony blot hybridization using the 3.2-kb SphI fragment of avrXa27 as a probe.Hybridizing colonies were further evaluated by PCR with the tal-specific primers TALN18-F and TALN18-R (Supplemental Table 3).Putative tal-containing clones were confirmed by restriction enzyme digestion and Sanger sequencing.tal17 and tal6c were cloned into the pZW vector in frame with C-terminal FLAG tag epitopes, resulting in pZW-tal17 and pZW-tal6c (Supplemental Table 1).These plasmids were ligated into the broad-host-range vector pHM1 at the HindIII site, generating pHZW-tal17 and pHZW-tal6c (Supplemental Table 1), which were then transferred into a tal-free PH strain (derived from PXO99 A ) (Ji et al., 2016) to obtain PH/tal17 and PH/ tal6c (Supplemental Table 1).

qRT-PCR
Total RNA was isolated from inoculated plants 24 h post-inoculation using RNAiso Plus reagent (TaKaRa Bio).Trace amounts of genomic DNA were removed with RNase-free DNase I (TransGen, China) prior to synthesis of cDNA.First-strand cDNA was diluted to a final volume of 20 mL, and SYBR Green-labeled PCR fragments were amplified using the Xa27 genespecific primers Xa27-qF and Xa27-qR (Supplemental Table 3) and the 7500 Real-Time PCR System (Applied Biosystems, USA).The rice Actin gene was used as an internal control (primers Actin-F and Actin-R; Supplemental Table 3).The comparative threshold (2 ÀDDCt ) method was used to calculate relative mRNA levels.qRT-PCR experiments were performed in triplicate.

Western blots
TALE proteins were detected in Xoo strains by immunoblotting as described previously (Xu et al., 2019).In brief, Xoo strains were cultured in NB to the logarithmic phase and harvested by centrifugation.Bacterial cells were washed twice, and the concentration was adjusted to OD 600 = 2.0 with sterile distilled water.SDS loading buffer (53) was added to the bacterial suspensions; these were then boiled in a water bath for 10 min, separated on SDS-PAGE gels, and transferred to polyvinylidene difluoride membranes for immunoblotting with anti-FLAG (TransGen) as the primary antibody.Goat anti-rabbit immunoglobulin G (TransGen) was used for detection of primary antibodies with the EasySee Western Kit supplied by TransGen.

GUS assay
GUS measurements were carried out as described previously (Haq et al., 2022).Promoter fragments of OsSWEET11a and Xa27 were cloned into the binary GUS reporter construct pCAMBIA1381 using primers listed in Supplemental Table 3.The binary vector pHB was used to clone and express TALEs PthXo1 and Tal6b in N. benthamiana.The effector and reporter constructs (Supplemental Table 1) were transformed into Agrobacterium strain EHA105 by the freeze-thaw method and then co-expressed (OD 600 = 1.0 for each strain) in 5-to 7-week-old N. benthamiana leaves via Agrobacterium-mediated transformation.Three leaf discs (1 cm diameter) were collected 48 h post inoculation, and GUS activity was measured using 4-methylumbelliferyl-b-glucuronide.Proteins were quantified using a Bradford Protein Quantification Kit (Yeasen, Shanghai).
Three independent experiments were performed with similar results.

Plant Communications
Generation of transgenic constructs and transgenic plants Kitaake rice was genetically modified with CRISPR-Cas9 technology as described previously (Xu et al., 2019).In brief, target sequences were selected within the promoter regions of OsSWEET11a, and single-guide RNAs (sgRNAs) were designed with the CRISPR MultiTargeter program (http://www.multicrispr.net/index.html) and then synthesized.BLAST searches (http://blast.ncbi.nlm.nih.gov/Blast.cgi) of the target sequences were performed against rice genome sequences to confirm target specificity.The sgRNA and Cas9 constructs were transferred into calli of rice cv.Kitaake by Agrobacterium-mediated transformation (Biorun, Wuhan, China).Genomic DNA was isolated from leaves of transgenic rice using the cetyltrimethyl ammonium bromide (CTAB) method and used for PCR amplification of the target regions with specific primers (Supplemental Table 3).The resulting amplicons were subjected to Sanger sequencing.Plant Communications 5, 100721, February 12 2024 ª 2023 Shanghai Jiao Tong University.

Plant Communications
Tal6b targets OsSWEET11a and Xa27 Plant Communications

( B )
Boxplots of mean disease lesion lengths (cm) on Nipponbare and Kitaake.Lesions were measured at 14 dpi; dots denote individual observations from at least 5 inoculated leaves, and whiskers display the first and third quartiles, split by the median.(C) Individual RVDs of Tal6b and its derivatives Tal6b-2 and Tal6b-6.Single letters denote amino acids at the 12th and 13th positions of individual repeats.An asterisk indicates a predicted missing 13th residue.(D) Western blot analysis of TALE proteins in Xoo PH/ev, PH/tal6b, PH/tal6b-2, and PH/tal6b-6.(E) Expression of OsSWEET genes in Kitaake rice leaves inoculated with Xoo strains.The expression levels of OsSWEET11a, OsSWEET13, and OsSWEET14 were measured in rice leaves infiltrated with Xoo PH/ev and PH/tal6b.RNA was extracted from leaves 24 hpi and used for qRT-PCR with OsSWEET11a-, OsSWEET13-, and OsSWEET14-specific primers.The expression level of Actin was used as an internal control.(F) Expression of OsSWEET11a in Kitaake rice leaves inoculated with Xoo PH strains containing the empty vector (PH/ev) and AvrXa27/TalAO-class TALEs (pthXo1, tal6b/avrXa27A, tal17/avrXa27B, tal6c/avrXa27C, and avrXa27).(G) Individual RVDs in the CCR region of PthXo1 and Tal6b/AvrXa27A recognize individual nucleotides in the predicted EBE regions of the OsSWEET11a promoter.Single letters denote amino acids at the 12th and 13th positions of individual repeats, and the asterisk denotes a missing amino acid at the 13th position of a particular repeat.The scores show matches between DNA sequences in the OsSWEET11a EBE and amino acid residues in the RVDs of Tal6b/AvrXa27A and PthXo1 (positive control); scores were predicted with the TALE-NT program.Lower scores indicate higher binding affinity between the RVDs and the target sequence.EBE PthXo1 is shown in red font, and EBE Tal6b is underlined.(+) refers to the forward strand, and (À) refers to the reverse strand.Values with the same lowercase letters do not differ significantly at P < 0.05 according to ANOVA.Plant Communications 5, 100721, February 12 2024 ª 2023 Shanghai Jiao Tong University.

Figure 4 .
Figure 4. Tal6b/AvrXa27A binds to the EBEs of Xa27 and OsSWEET11a.(A) Nucleotide sequence of the Xa27 promoter fragment (probe); the underscored nucleotides show the EBE recognized by Tal6b/AvrXa27A.(B) His-tagged Tal6b/AvrXa27A fusion protein binds to the Xa27p probe derived from the Xa27 promoter.Positions of the bound and free probe are indicated on the left.(C) Nucleotide sequence of the OsSWEET11a promoter fragment (probe); EBE PthXo1 is underscored.(D) His-tagged PthXo1 binds to the Xa13p probe of the OsSWEET11a promoter in gel-shift assays.Positions of the bound and free probe are indicated at the left.(E) Alignment of Xa13p, Xa13p-10, Xa13p-11, Xa13p-12, and Xa13p-13 sequences.Tal6b is underlined.Mutations in the OsSWEET11a promoter sequence are indicated in red font.(F) Binding of His-tagged Tal6b/AvrXa27A to EBE Tal6b in the OsSWEET11a promoter.

Figure 6 .
Figure 6.A working model of the role of Tal6b/AvrXa27A in the interaction between Xoo and rice.

Table 1 .
Repeat variable diresidue (RVD) sequences in TALEs of the AvrXa27/TalAO class in selected Xoo strains.
*The RVDs in italic fontdiffer from those in AvrXa27.bIXO221 is an Indian strain containing AvrXa27C; the whole genome sequence is not available.4PlantCommunications 5, 100721, February 12 2024 ª 2023 Shanghai Jiao Tong University.