TALome and phenotypic analysis of Pakistani Xanthomonas oryzae pv. oryzae population revealed novel virulent TALEs contributing to bacterial blight of rice

Abstract

Bacterial blight (BB) of rice caused by Xanthomonas oryzae pv. oryzae (Xoo), is an important disease in rice-growing countries, including Pakistan, where it was first reported in the mid-1970s. Transcription activator-like effectors (TALEs) play vital roles in many plant diseases caused by Xanthomonas spp.; however, Pakistani Xoo TALome diversity and their contribution to pathogenicity is largely unknown. In this study, 101 Xoo strains were screened using specific PCR primers. The genomic DNA from these strains underwent BamHI digestion and hybridized with the internal SphI fragment of PthXo1. Southern blot analysis revealed 16 to 20 putative tale fragments among the tested strains. These strains were further classified into 11 genotypes based on the number and size of the hybridizing bands. Genotypes 1, 2, 3, and 4 represented 24, 2, 51, and 17 strains, respectively. Pathogenicity assays on near-isogenic lines (NILs) containing different resistance (R) genes exhibited that CBB23 was incompatible with all tested Pakistani-Xoo genotypes, whereas IRBB5 and IRBB4 showed resistance against specific genotypes. In contrast, paddy trails on NILs containing single, double, and triple mutants of OsSWEET11a, OsSWEET13, and OsSWEET14 in the effector binding elements (EBEs) of cv. Kitaake revealed that KP-22 and LD-5 harbor novel virulent TAL effector/s. Interestingly, the expression analysis of six clade-III OsSWEET genes suggests that novel TALE/s targeting unidentified susceptibility gene/s. Altogether, this study highlights gene-for-gene relationships between tested rice lines and Pakistani-Xoo strains. This is the first report providing the diversity of TALEs and their relationship to R and S (susceptibility) genes. Further identification of novel virulent TALE/s and their cognate target/s is warranted to precisely elucidate their role in BB.

Background

Rice (Oryza sativa L.) is one of the vital staple foods for over half of the world's population, which meets both calories and food requirements (Ainsworth 2008). Addressing the intricate environmental challenges and escalating global population, there is a pressing need for concerted efforts to stabilize rice production, although rice production has increased over the past decade (FAO). However, this agricultural intensification brings with it the potential threat of emerging pathogens, posing a significant risk unless locally adapted control solutions are promptly implemented (Gregory et al. 2009).

Among foliar diseases, bacterial blight (BB), caused by the pathogen Xanthomonas oryzae pv. oryzae (Xoo), is an important disease of rice worldwide. It limits rice production each year owing to its high epidemic potential in tropical regions, especially in Southern Asia and parts of West Africa (Nino-Liu et al. 2006). Xoo enters the leaf through hydathodes or wounds, causing systematic infection of the vascular system. This results in green, small water-soaked spots at the margins and tips of leaves, which expand along the veins, merge, and become chlorotic and then necrotic lesions (Ou 1985; Nino-Liu et al. 2006). Reported yield losses due to BB range from 20 to 30%, with reductions as high as 50–90% in some areas (Ou 1973; Nino-Liu et al. 2006; Liu et al. 2014). Moreover, rice BB is reported as an emerging disease worldwide, seriously affecting the quality and quantity in almost all of the rice growing countries (Naqvi 2019).

Similar to other gram-negative bacteria, Xoo employs a type-III secretion system (T3SS) to translocate a cocktail of type-III effectors (T3Es) into the host cell cytoplasm. These T3Es can be categorized into transcription activator-like effectors (TALEs) and non-TALEs or Xanthomonas outer proteins (Xops). Unlike the diverse molecular activities found in Xops, members of the TALEs family resemble eukaryotic transcription factors and exhibit sequence-specific binding to the promoters of target genes within the host cells. The specific sequences targeted by TALEs are referred to as the effector binding elements (EBEs). Structurally, TALEs consist of an N-terminal domain containing a type III secretion signal followed by a central repeat region (CRR; 33–35 amino acids [aa], in which 12th and 13th aa are variable and called repeat variable di-residue; RVD), and a C-terminus domain harboring nuclear localization signal (NLS) and transcriptional activation domain (AD) that are important for effector localization and gene activation, respectively (Boch and Bonas 2010; Mak et al. 2013; Richter et al. 2014; Perez-Quintero and Szurek 2019).

In general, following injection, TALE proteins localize into the host cell nucleus, where they recognize and bind to their EBEs in the promoter region of host resistance (R) or susceptibility (S) genes, inducing their expression (Bogdanove et al. 2010). Based on a handful of characterized examples, TALEs exploit host S genes, which play important roles in host recognition, subvert plant immunity, and transport nutrients to the pathogen. Most of the known TALE-targeted S genes encode transporters and transcriptional factors. For example, three specific sugar transporter genes, belonging to clade-III of SWEET genes, are targeted by about ten major known TALEs, including OsSWEET11a/Xa13/Os8N3 induced by PthXo1 and Tal6b/AvrXa27A at overlapping EBEs (Yang et al. 2006; Xu et al. 2023), OsSWEET13/Xa25/Os12N3 targeted by multiple PthXo2-like TALEs (i.e., PthXo2, PthXo2C/Tal5_LN18, PthXo2B/Tal7_PXO61, and Tal7_K74) (Zhou et al. 2015; Xu et al. 2019), and OsSWEET14/Xa41(t)/Os11N3 activated by PthXo3, Tal5/TalF, TalC, and AvrXa7 at overlapping/different EBEs (Antony et al. 2010; Yu et al. 2011; Streubel et al. 2013; Tran et al. 2018). Recently, a new rice S gene (OsERF#123) was shown to be targeted by TalB in the African strains of Xoo (Tran et al. 2018). In addition, X. oryzae pv. oryzicola Tal2g targets OsSULTR3;6, a sulfate transporter gene, which is a major S gene for bacterial leaf streak (Cernadas et al. 2014). Another class of TALE-dependent S genes comprises host transcription factors; for instance, PthXo6 and PthXo7 of Xoo induce the expression of rice transcription factors OsTFX1 and TFIIAγ1, respectively (Sugio et al. 2007).

In response to the action of TALEs, plants have evolved effective strategies to trigger resistance, such as loss of S genes or use of R genes (Kourelis and Van Der Hoorn 2018). Amongst 47 R genes, 17 have been cloned that individually confer rice resistance against BB (Jiang et al. 2020). Eight of these genes are TALE-dependent, including four recessive and four dominant R genes (xa5, xa13, xa25, xa41, Xa7, Xa10, Xa23, and Xa27), reflecting the crucial role of TALEs in the interaction (Hutin et al. 2015; Ji et al. 2022). The recessive R genes involve mutations within the EBEs of S genes to prevent TALE-DNA interactions. For example, xa13 (OsSWEET11a), xa41 (t) (OsSWEET14), xa25 (OsSWEET13), and xa5 (TFIIAγ5) cannot induce susceptibility due to mutations in the promoter EBEs or amino acid sequences (Chu et al. 2006; Gu et al. 2009; Liu et al. 2011). TALEs mediate executor (E) genes of so-called dominant resistance, that restrict the growth of the pathogen via rapid cell death. For instance, rice executor R-genes Xa7, Xa10, Xa23, and Xa27 are induced by their matching TALEs, AvrXa7, AvrXa10, AvrXa23, and AvrXa27, respectively (Gu et al. 2005; Tian et al. 2014; Wang et al. 2015; Chen et al. 2021). These TALEs are exclusive to Asian Xoo strains, and none of the E genes have been induced by African strains. The E genes encode small proteins with transmembrane domains, and the molecular mechanisms governing their function remain largely unclear (Zhang et al. 2015; Chen et al. 2021). Moreover, receptor-like kinase (RLK) genes, such as Xa3/Xa26, Xa4, and Xa21, also confer dominant resistance (Song et al. 1995; Sun et al. 2004; Hu et al. 2017).

Direct protein–protein (TALE and NBS-LRR proteins) interaction is responsible for another type of disease resistance, independent of direct host gene activation, for example, rice Xa1 and Xo1 (Yoshimura et al. 1998; Triplett et al. 2016; Xue et al. 2020). To escape host immunity, pathogenic bacteria have developed and deployed several different strategies such as truncTALEs (truncated TALEs) or iTALEs (interfering TALEs) to restrict TALE-dependent plant defense responses; for example, iTALE/trucTALE Tal2h suppresses Xa1/Xo1 mediated resistance (Ji et al. 2016; Read et al. 2020; Xu et al. 2021). After decades of this effort, the mode of action of TALE proteins and plant resistance has taught us a valuable lesson about the never-ending co-evolutionary arms race during plant–microbe interactions.

In Xoo, there is an uneven distribution of TALE repertoires, ranging from 8 to 21 (African Xoo 8–9 while Asian Xoo 18–21), but very few of them are characterized (Tran et al. 2018; Oliva et al. 2019). Therefore, there is a need to explore the reasons behind the abundance of TALEs in Xoo and understand their role in promoting virulence. The TALE repertoires are structured into gene clusters along the chromosome, and it is proposed that the evolution of tal genes occurs through the mutation in the repeat sequences, rearrangement, and/or the deletion of single repeats (Erkes et al. 2017). Accordingly, certain repeat arrays have been identified among unrelated TALEs within the same strain, indicating the possibility of intergenic recombination as a mechanism for generating novel variants (Tran et al. 2018). Due to their involvement in pathogenicity and interaction with corresponding R genes in rice varieties, the evolution of TALEs is subjected to continuous selective pressure, leading to rapid evolutionary changes (Schandry et al. 2018). Knowing the TALomes diversity and understanding the evolutionary patterns is essential for strategically deploying varieties equipped with locally adapted resistance genes and anticipating the risk of the emergence of new, highly aggressive strains.

Resistance gene deployment requires knowledge of the bacterial population structure and their spatiotemporal variability. Therefore, the main purpose of this study was to evaluate the TALome diversity of Xoo strains collected from two provinces in Pakistan and investigate gene-for-gene interactions between Xoo and rice to improve their control. Here, we present the genetic and pathogenic characterization of 101 Pakistani Xoo strains collected in 2017, 2018, 2019, and 2021. TALE repertoires profiling revealed the presence of 16–20 putative TALE fragments ranging from about 2.2 kb to 4.6 kb. Pathotyping on NILs containing different R genes indicated that CBB23 is effective against all the tested strains. On the other hand, NILs containing EBE mutants of OsSWEET11a, OsSWEET13, and OsSWEET14, revealed that KP-22 and LD-5 harbor a novel virulent TAL effector/s. Furthermore, expression analysis of six clade-III OsSWEET genes suggested the presence of novel TALE/s targeting previously unidentified susceptibility gene/s. This report represents the first comprehensive analysis of TALE diversity and its association with R and S genes. Taken together, our approach highlighted that Pakistani-Xoo strains contain novel virulent TALEs targeting novel gene/s. Further identification of novel virulent TALEs is imperative for a precise understanding of their role in BB.

Results

Collection of Pakistani-Xoo strains

Rice growing areas of Punjab (P) and Khyber Pakhtunkhwa (KP) were surveyed in 2017, 2018, 2019, and 2021 (Additional file 1: Table S1). Five districts each in Punjab and Khyber Pakhtunkhwa were sampled where bacterial blight was present in varying intensity. In Punjab, the disease incidence was low in 2017 and 2018 compared to 2019 and 2021. The highest disease incidence (80%) was recorded in Sialkot during 2021, followed by Narowal (50%) and Gujranwala (36%). The disease was also observed in other Basmati growing areas such as Hafizabad and Sheikupura. In KP, the disease incidence was comparatively lower than in Punjab. The highest disease incidence (40%) was recorded in Swat in 2018, followed by Lower Dir, Battagram, Mansehra, and Bannu (Additional file 2: Figure S1).

To assess the diversity of TALE repertoires and pathogen aggressiveness, a collection of 101 Xoo strains were isolated, among which 39 were from Punjab and 62 from KP. In total, 40 strains were isolated from leaves collected in 2017, 22 in 2018, 24 in 2019, and 15 in 2021 (Additional file 2: Figure S1). The strains grow as pale yellow to brownish colonies on NA plates and cause disease on rice upon inoculation. Moreover, the pathovar-specific PCR primers Xoo80F/Xoo80R amplify a 162 bp DNA fragment, confirming that all the strains are Xoo (Additional file 1: Table S2). Additionally, the product size and intensity were also similar in gel compared with PXO99^A (Additional file 2: Figure S2).

TALes diversity in Pakistani-Xoo strains

To investigate the TALome diversity of 101 Xoo strains isolated from the rice growing fields in Pakistan (Additional file 2: Figure S1), we conducted Southern blot (SB) hybridization. The well-characterized Philippines and Chinese strains, PXO99^A and LN18, were included for comparative purposes. The SB results revealed that all the strains exhibited multiple fragments of putative TALEs homologous to the SphI fragment of PthXo1 (Additional file 2: Figure S3). The detected putative TALEs ranged in size from about 2.2 to 4.6 kb, and the number of fragments per strain varied from 16 to 20, considering the intensely hybridizing bands as multiple fragments (Table 1). Based on the size and number of the detected TALE fragments, the 101 strains were classified into 11 genotypes. The hybridized tale genes were indicated as A to T in descending order (Fig. 1). The TALE profiles of 11 genotypes (hereafter refer as G1, G2, and so on) of Pakistani-Xoo strains revealed geographic distinct pattern except for G3 and G4. Notably, the putative TALE fragments designated as D, E, F, L, O, and Q were consistently present in all Xoo strains. The intensely hybridizing fragments were considered as two and four bands, based on the intensity, thickness, and comparison (in-silico simulation on gel) with the available Pakistani Xoo strain PkXoo1 (TALEs = 18), which revealed multiple fragments on the same position. TALE fragments D = D1, D2; E = E1, E2; L = L1, L2, L3, L4 (in all genotypes); O = O1, O2 (in G1 to G6, G8, G10, and G11); Q = Q1, Q2 (in G7 to G11), and R = R1, R2 (in G8, G10, and G11) are intensely hybridizing and were considered multiple fragments. TALEs designated as H, I, J, K, M, N, P, and R were detected in more than two genotypes. Furthermore, the TALE fragments designated as A, B, C, G, S, and T were uniquely present in genotypes G2, G7, G11, G5, G6, and G8, respectively (Fig. 1, Table 1).

Table 1 Transcriptional activator-like effector (TALE) genotypes of Xanthomonas oryzae pv. oryzae strains collected from different rice-growing regions of Pakistan

Fig. 1

Southern blot patterns depicting the 11 genotypes of Pakistani Xoo isolates. Genomic DNAs from the mentioned strains underwent BamHI digestion and were subsequently hybridized with an internal SphI fragment of PthXo1. The 101 strains were categorized into 11 distinct genotypes: G1 (24 strains), G2 (2 strains), G3 (51 strains), G4 (17 strains), and G5 to G11, each containing a single strain, respectively (Table 1). Genotypes are displayed at the top of the image beneath the strain names. The hybridizing bands of varying sizes are identified by different letters (A to T) on the right side of the figure, while bands marked with an asterisk (*) were unique to a single strain (pointed out with the red arrows in the figure). TALE fragments D = 2, E = 2, L = 4 (in all genotypes); O = 2 (G1 to G6, G8, G10, and G11), Q = 2 (G7 to G11), and R = 2 (G8, G10, and G11) are intensely hybridizing and were considered multiple (Table 1). Strain names were assigned using a letter representing their province of origin, followed by a number, except for LD (Lower Dir, KP), BG (Battagram, KP), and BN (Bannu, KP) strains. The province abbreviations used are as follows: P for Punjab and KP for Khyber Pakhtunkhwa. The left lanes of the image feature the λEcoT14 marker (base pairs; bp) and strain LN18 for reference. The right lanes represent the PH strain (PXO99^A tal-free) containing PthXo1, PthXo2, and AvrXa7 in trans that were used to highlight the major virulence tal genes

TALE diversity analysis by clustering the presence and absence of BamHI fragment using NTSYS 2.02e tool classified the 11 genotypes into 5 clades (Fig. 2). Interestingly, the clade pattern was almost in congruence with the geographic distribution of the strains, i.e., genotypes from Punjab G1, G2, and G4; G6 grouped into clade-1 and clade-3, respectively. The other genotypes belong to KP, i.e., G3 and G5; G8, G11, and G10 were grouped into clade-2 and clade-5, correspondingly (Fig. 2). Genotypes G7 and G9, each consisting of a single strain from Punjab and KP, were grouped into clade-4. Regarding the distribution of strains across different genotypes, G3 is the most common group containing 51 strains (49 strains from KP and 2 strains from Punjab). Other genotypes, G1 and G4 (9 strains from Punjab and 8 strains from KP), also contained many strains, i.e., 24 and 17 strains, respectively. On the other hand, genotypes G5, G6, G7, G8, G9, G10, and G11 each contained an individual strain, followed by G2, which contained two strains. Moreover, the genotypes G3 and G4 have strains from both Punjab and KP as described above (Table 1). Additionally, our results indicated that strains collected from Sialkot and Swat are more diverse than those collected from other regions (Table 2).

Fig. 2

Dendrogram illustrating the relatedness of Pakistani Xoo genotypes. The dendrogram demonstrates the relationship of Xoo genotypes based on Southern blot analysis, considering the presence or absence of hybridizing BamHI fragments (Fig. 1). All genotypes were classified into 5 clades (1–5, in red fonts) as block shaded. The tree was generated by creating the similarity matrix in Simqual and clustering in SAHN using the NTSYSpc-2.02e. Genotypes are presented with their representative reference strain in parentheses

Table 2 Genotypes of Xanthomonas oryzae pv. oryzae strains collected from different localities of Pakistan

To exploit the major virulence tal genes in SB, we used PH strain (PXO99^A tal-free) containing PthXo1, PthXo2, and AvrXa7 in trans. Southern blot analysis revealed that all the genotypes contained similar fragments of PthXo1 and seven genotypes (G1, G2, G3, G4, G9, G10, and G11) harbor potential fragments of PthXo2, while the AvrXa7 fragment did not match with any of the genotypes (Fig. 1). Interestingly, the majority of the strains contained both PthXo1 and PthXo2, which implies that these strains possess the capability to simultaneously activate OsSWEET11a and OsSWEET13. Notably, SB results did not completely align with the inoculation phenotypes, possibly due to the presence of similar-sized TALEs that remained unidentified.

Germplasm screening for TALes-dependent resistance

To assess the prevalence and diversity of TALes counter resistance, 11 genotypes (a collection of 101 strains) were screened against the indica/japonica near isogenic lines (NILs) containing different R genes (IR24, IRBB3, IRBB4, IRBB5, CBB23, and Kit-Xa1) normally targeted by TAL effectors. The Pakistani-Xoo genotypes, along with reference strains (PXO99^A, PXO86, and LN18), were inoculated by leaf clipping method into the field plants and the lesion length was measured at 14 dpi (Table 3; Additional file 2: Figure S4a). Plants with lesion lengths of 0–3 and 3–5 cm were considered resistant and moderately resistant, while 5–8 and > 8cm were considered as moderately susceptible and susceptible, respectively. Interestingly, all the strains were found incompatible with rice CBB23, especially KP-3, KP-24, LD-13, and BG-3, which were compatible against NILs, similar to PXO99^A (Table 3). In contrast, P-14, P-27, and P-35 strains were incompatible with IR24 and its tested NILs. P-18, P-10, and LD-5 strains were incompatible against IRBB5, similar to PXO86. KP-22 was the only strain found incompatible with IRBB4 and IRBB5. Rice NILs IR24, IRBB3, and Kit-Xa1 were found susceptible to all tested Pakistani Xoo strains (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4a). Overall, virulence assays revealed that all Pakistani Xoo strains harbor AvrXa23, and most of them also have iTALE, which triggers Xa23 and Xa1-mediated resistance and susceptibility, respectively.

Table 3 Pathogenicity assay of Pakistani Xoo genotypes in susceptibility genes edited and different R-genes near-isogenic rice lines

Germplasm screening for TALes-dependent susceptibility

An assessment of the major virulence TALEs was performed by virulence assays on rice lines featuring polymorphisms in the OsSWEET promoter. The japonica rice variety Kitaake and its derivative lines, carrying mutations in the three OsSWEET genes normally targeted by TALes, were challenged with 11 Pakistani Xoo genotypes representing a collection of 101 strains, including well-known strains PXO99^A, PXO86, and LN18 for the comparative purposes (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4b). Kitaake NILs, accommodating OsSWEET11a, OsSWEET13, and OsSWEET14 mutated EBEs either singly (MS1K, MS3K, and MS4K) or in double and triple combinations (MS13K, MS14K, MS34K, and MS134K), were generated by CRISPR Cas9 (Additional file 1: Table S4). The 14 dpi lesion length measured between 0–3 and 3–5 cm was considered resistant and moderately resistant, while lengths between 5–8 and > 8cm were considered moderately susceptible and susceptible, respectively.

Three genotypes, namely P-14 (G2), P-35 (G4), and P-27 (G6), representing the diversity of the collection, were not virulent on the japonica reference line Kitaake. Likewise, these genotypes exhibited similar phenotypes on indica parental line IR24 and its NILs containing R genes (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4). These genotypes, containing 20 of 101 strains, may be solely dependent on a novel variant of PthXo2-like effector (PthXo2**), i.e., PthXo2, which was incompatible with Kitaake but compatible with IR24, while PthXo2B was compatible with Kitaake and incompatible with IR24 (Oliva et al. 2019). Alternatively, cv. Kitaake and IR24 may have a resistance gene that is effective against these 20 strains.

Of the remaining 81 Xoo strains, one strain P-18 (G7) was not virulent on edited rice lines, which are defective in the EBE targeted by PthXo3/AvrXa7 in OsSWEET14 (MS4K, MS14K, MS34K, and MS134K), similar to the reference strain PXO86. Genotypes P-10 (G1), KP-3 (G3), KP-24 (G9), LD-13 (G5), and BG-3 (G10), representing 78 strains, were avirulent on rice lines which are defective in the EBE of PthXo1 within OsSWEET11a (MS1K, MS13K, MS14K, and MS134K), similar to the reference strain PXO99^A (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4b). Phenotypic assays revealed that none of the strains harbored PthXo1 and/ or PthXo2 and/ or PthXo3/AvrXa7, simultaneously. Moreover, the remaining two strains, KP-22 (G8) and LD-5 (G11), elicited susceptible phenotype on all edited rice lines, which harbor mutant alleles of OsSWEET11a, OsSWEET13, and OsSWEET14, whereas the lesion length was reduced in OsSWEET11a defective EBE, unlike Kitaake plants (Fig. 3a). Altogether, these findings indicate that 78 strains from the screen rely on major TALEs that target known EBE of OsSWEET11a and one strain on OsSWEET14, whereas two deviant strains, KP-22 and LD-5, have different virulence abilities, presumably possessing novel virulent TALE/s. These strains are originated from KP, and the virulence assay distinguished that these deviant strains, individually, carry both PthXo1-like plus another novel major virulent TALE (MVT) gene/s (Table 3; Fig. 3a; Additional file 1: Table S3; Additional file 2: Figure S4b).

Fig. 3

Virulence and OsSWEETs expression induced by KP-22 and LD-5 in Kitaake and MS134K rice lines. a Lesion length induced by KP-22 and LD-5 in cv. Kitaake and major TALEs EBE edited rice lines. Bacterial strains were inoculated by tip-cutting and images were captured 14 days after inoculation. The figures display representative disease lesions, and the mean lesion lengths with standard deviation (n = 8) are detailed in Additional file 1: Table S3. b mRNA levels of OsSWEET11a, OsSWEET11b, OsSWEET12, OsSWEET13, OsSWEET14, and OsSWEET15 (2^−ΔΔCt) determined by qRT-PCR in cv. Kitaake and MS134K leaves treated with double-distilled water (ddw, mock inoculation), LN18, LD-5, and KP-22 (compatible strains to MS134K) at 24 hpi. The strain LN18 was used for reference, ddw as external control and rice Actin-6 gene as an internal control. Data represent replicated (n = 3) qRT-PCR

Characterization of deviant strains

To further test whether the novel TALE/s possibly target/s the new EBE for disease susceptibility, we conducted expression analysis of OsSWEET11a, OsSWEET11b, OsSWEET12, OsSWEET13, OsSWEET14, and OsSWEET15 by inoculating ddw (double-distilled water), KP-22, LD-5, and LN18 into Kitaake and MS134K plants (Fig. 3b), with ddw and Actin served as external and internal controls, respectively. LN18 was the sole representative strain harboring PthXo2 like (Tal5) and AvrXa7 effectors targeting both OsSWEET13 and OsSWEET14. OsSWEET13 and OsSWEET14 were significantly expressed in Kitaake but not in MS134K plants infected with LN18. This finding was consistent with previous studies where Tal5 and AvrXa7 were shown to be the major virulent TALEs in LN18 (Xu et al. 2019). Similarly, only OsSWEET11a was significantly expressed in Kitaake rice infected with LD-5 and KP-22 but not in MS134K, and the OsSWEET11a defective EBE plants were moderately susceptible compared with Kitaake (Fig. 3). On the other hand, OsSWEETs edited plants were moderately resistant or resistant to all other Pakistani Xoo genotypes. These results suggest that LD-5 and KP-22 contain two major TALEs, i.e., PthXo1-like plus another novel MVTs that contribute to the susceptibility of defective EBE plants. Moreover, LD-5 and KP-22 could not induce the expression of OsSWEET11b, OsSWEET12, and OsSWEET15, indicating the presence of unidentified susceptibility gene/s other than the tested OsSWEET genes.

Deployment of rice varieties based on major TALEs prediction in Pakistani-Xoo strains

To identify the major TALEs in Pakistani-Xoo strains, we used cv. IR24 and Kitaake NILs containing genes targeted by TALEs. The major TALEs were inferred by analyzing the phenotypic data obtained from field inoculations. Notably, rice lines in the background of IR24/Kit, i.e., IRBB5, CBB23, and Kit-Xa1 carry xa5, Xa23, and Xa1, which are targeted by PthXo7, AvrXa23, and iTALE, respectively. Briefly, Xoo strains containing PthXo7 target the recessive R gene xa5, leading to a susceptible phenotype, while strains that harbor iTALE suppress the Xa1-dependent resistance. Moreover, Xa23 is the executor R gene and confers broad-spectrum resistance to Xoo strains harboring AvrXa23 (Hutin et al. 2015; Ji et al. 2022). Additionally, to estimate the major TALEs targeting susceptibility genes, we used CRISPR-Cas9 edited rice lines containing mutations in all three EBEs recognized by the major TALEs, i.e., PthXo1/PthXo1* (Tal6b), PthXo2*, and PthXo3/AvrXa7 (Liu et al. 2024). These findings provided valuable guidance for the strategic deployment of disease-resistant rice varieties. For instance, to identify suitable rice varieties resistant to bacterial blight across various regions of Pakistan, we analyzed the geographic distribution of 101 Xoo strains and their probable R and S genes. Based on the phenotypes, all Xoo-strains harbor AvrXa23, while 81, 54, and 20 strains contain iTALE, PthXo7, and PthXo2**, respectively (Fig. 4). For susceptibility genes, 78 Xoo strains possessed major TALEs that could potentially activate OsSWEET11a. There was one Xoo strain that could possibly activate OsSWEET14. Two KP (Swat and Lowe dir) Xoo strains activated OsSWEET11a along with unidentified susceptibility gene/s (Figs. 3, 4). Most strains originating from Punjab activated OsSWEET11a and OsSWEET14, while strains from KP activate OsSWEET11a but not OsSWEET13 or OsSWEET14. Our results suggest that EBE editing of OsSWEET11a and OsSWEET14 could be employed to mitigate BB effectively in Punjab.

Fig. 4

Major TALEs prediction in Pakistani-Xoo strains. The assignment of the major TALEs was based on the virulence assessment of Xoo strains upon inoculation into Near-Isogenic Lines (NILs) harboring resistance genes or edited susceptibility genes in cv. IR24 and Kitaake, defining the relative contributions of multiple avirulence/virulence effectors. At the top, parsimony tree was generated using the NTSYSpc-2.02e, and all the Xoo genotypes including reference strains were clustered into 6 groups (1–6 in red fonts). The estimated avirulence/virulence TAL effectors carried by the genotypes/strains are shown in the table. The representative reference strains are orange highlighted and genotypes are shown at the bottom

To illustrate the relatedness of Pakistani-Xoo strains, a dendrogram was generated considering the presumed major TALEs based on the virulence assessment of NILs containing R/S genes using NTSYSpc-2.02e. The Xoo genotypes plus reference strains were clustered into 6-groups (Fig. 4). The estimated virulence/avirulence effectors carried by the genotype and their cognate genes in NILs are shown in the table (Fig. 4). Interestingly, these results are in congruence with the single available genome of Pakistani Xoo strain PkXoo1 (Additional file 1: Table S5). Collectively, our findings suggest that CBB23 is the most effective rice line that could be used to combat Xoo in Pakistan, especially in Punjab and Khyber Pakhtunkhwa.

Discussion

Transcription activator-like effectors (TALEs), recognized as virulence markers of Xanthomonas spp., are highly evolvable and subject to selective pressures, playing crucial roles in various plant diseases. In this study, we aimed to describe the population structure of the Pakistani Xoo collection through the analysis of TALEs diversity and aggressiveness of the strains on Kitaake NILs with defective EBEs/IR24 NILs containing R genes. In total, 101 Xoo strains were collected from the two rice-growing provinces in Pakistan (Additional file 2: Figure S1) and screened via PCR using pathovar-specific primers (Additional file 2: Figure S2). TALE-based genotyping clustered the 101 isolates into 11 genotypes (Fig. 1; Additional file 2: Figure S3) containing 16 to 20 potential TALE fragments (Table 1). Amongst, 51, 24, and 17 strains were assigned to genotypes 3, 1, and 4, respectively (Table 1). TALE diversity analysis further categorized the 11 genotypes into 5 clades. Notably, the clade pattern aligned with the geographic distribution of the strains except for G3 and G4 encompassing strains from both provinces and clade 4 of G7 and G9 (Fig. 2). Virulence assays on NILs containing different R genes revealed that CBB23 is the most effective rice line against all Pakistani Xoo genotypes (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4a). Moreover, pathogenicity assays on NILs containing defective EBEs for the three major virulent TALEs (PthXo1, PthXo2, and PthXo3/AvrXa7) revealed two deviant strains, KP-22 and LD-5, harboring PthXo1-like plus novel MVTs contributing to susceptibility (Table 3; Additional file 1: Table S3; Additional file 2: Figure S4b). Intriguingly, expression analysis of six clade-III OsSWEET genes suggests the presence of novel TALE/s targeting previously unidentified susceptibility gene/s (Fig. 3). These findings provide valuable insights into the diversity and pathogenicity of the Pakistani Xoo population and highlight the complex interactions between TALEs and host plants (Fig. 4).

Bacterial blight (BB) was first reported in Pakistan in the mid-1970s, and recent surveys conducted in major rice-growing areas of three provinces—Punjab, Sindh, and Khyber Pakhtunkhwa —have shown a continuous increase in disease incidence. The prevalence of BB is particularly notable in Punjab and KP, as reported in a study by Ahsan et al. (2021). In response to this concerning trend, we sought to provide a comprehensive understanding of the TALome diversity and aggressiveness of Xoo isolates collected from these two provinces (Additional file 2: Figure. S1). To date, only one complete genome sequence of a Pakistani Xoo strain, PkXoo1, has been sequenced (Ejaz et al. 2023). Through TALE prediction and analysis, we discovered that PkXoo1 harbors 18 tal genes, a finding consistent with our Southern blot results, which indicated the presence of 16, 17, 18, and 20 TALE fragments (Table 1). Similarly, genome analysis of 33 Asian Xoo strains, whether derived from databases or newly sequenced, demonstrated the presence of 18–21 TALE loci per genome (Oliva et al. 2019). Notably, Pakistani Xoo strains exhibit a remarkably high genetic diversity in terms of TALEs, surpassing the extent that has been previously elucidated. This heightened genetic diversity in the TALome of Pakistani Xoo strains may have implications for the increased incidence and prevalence of bacterial blight in the regions, particularly in Punjab and KP.

The impact of TAL effectors on pathogen virulence or avirulence can vary significantly, ranging from major to moderate or even undetectable. In this study, our focus was on identifying TAL effectors within Pakistani Xoo strains that might function as avirulence (avr) or virulence factors. Utilizing varieties carrying resistance genes is considered a highly effective method for controlling BB in rice. Dominant resistance is often triggered by TALE-mediated induction of executor genes, whose expression leads to rapid plant cell death, thereby impeding disease development (Boch et al. 2014). Here, toward identifying sources of resistance effective against Pakistani Xoo strains, we assessed the putative avr activity of 11 genotypes (representing the collection of 101 strains) on rice accessions possessing Xa1, Xa3, Xa4, xa5, and Xa23 genes. The rice line containing Xa23 was most effective against Pakistani Xoo strains, followed by xa5 and Xa4 (Table 3; Fig. 4; Additional file 1: Table S3; Additional file 2: Figure S4a). TALEs comparative analysis revealed that PkXoo1 harbors PthXo1, PthXo6, PthXo7, PthXo8, AvrXa23, and AvrXa27 (with 0–5 RVDs differences) (Additional file 1: Table S5). This observation was consistent with the phenotypes, i.e., Xa23 comes up as one of the most promising R genes in terms of resistance spectrum for the Pakistani Xoo population. In conclusion, Xa23 emerges as an excellent material for breeding to develop improved rice varieties with enhanced resistance to the conditions in Pakistan.

Numerous reports indicate that tal genes play a significant role in virulence across various Xanthomonas species, targeting the EBEs in the promoter region of susceptibility genes (Yang et al. 2006; Antony et al. 2010; Yu et al. 2011; Streubel et al. 2013; Zhou et al. 2015; Tran et al. 2018; Xu et al. 2019; Xu et al. 2023). Mutation of these EBEs is being employed as a pivotal strategy to develop resistant crops against TALE-dependent pathogens (Eom et al. 2019; Oliva et al. 2019; Xu et al. 2019). However, a potential obstacle to the successful implementation of this strategy arises from TALE adaptations through the rearrangement of their repeat regions (Teper and Wang 2021). The current investigation involved the assessment of various Xoo genotypes from Pakistan on different rice lines, with single, double, and triple edits of major TALEs EBE sequences in the promoters of three OsSWEET genes in the rice cv. Kitaake. This unique set of rice materials proved effective in predicting major virulent TALEs in Xoo strains through plant inoculation and phenotypic monitoring, without relying on genomic sequencing. The effectiveness of this set of materials was verified through genomic prediction and expression of susceptibility genes (Liu et al. 2024). The assortment of Xoo strains from diverse geographical regions, their inoculation onto edited rice lines, and the utilization of the TALE prediction table offer valuable insights for more informed breeding strategies to enhance bacterial blight resistance in rice (Liu et al. 2024). In screening 11 Pakistani Xoo genotypes through inoculation tests, three genotypes, representing 20 strains, failed to induce disease on Kitaake and its edited rice lines. This failure was attributed to the presence of only PthXo2-like TAL effectors in these strains, which are incompatible with the OsSWEET13 EBE of Kitaake. Amongst, five genotypes representing 78 strains unable to elicit susceptibility on OsSWEET11a EBE defective rice lines (MS1K, MS13K, MS14K, and MS134K), suggesting the presence of PthXo1 in these strains. OsSWEET14 EBE edited rice lines (MS4K, MS14K, and MS134K) displayed resistance to only one genotype, indicating the presence of PthXo3/AvrXa7 TAL effector. Notably, none of the genotypes contained two or three major TAL effectors simultaneously. Moreover, two genotypes (KP-22 and LD-5) were capable of inducing a susceptible phenotype on cv. Kitaake and all seven edited lines (reduced lesion length on OsSWEET11a EBE edited rice lines), implying the emergence of a new virulence TALE/s that can induce unidentified susceptibility along with PthXo1/like. These strains harbored the highest number of tal genes, specifically 20. Additionally, an expression analysis of OsSWEET genes was conducted to determine whether the novel TALE/s targeted a novel EBE among the major susceptibility genes (Figs. 3, 4). Altogether, these results highlight an important tal gene diversity of the Xoo population from Pakistan.

Bacterial blight, caused by Xoo, exhibits the potential for multiple Xoo races to incite the disease within a given geographical area. Regional convergence is evident among Xoo strains, showcasing notable diversity in races across different geographical regions. For instance, in the central-eastern (Punjab) and northwestern (Khyber Pakhtunkhwa) rice cultivation areas of Pakistan, nearly all Xoo strains carry PthXo1 which can activate OsSWEET11a. Notably, a single strain from the central-eastern region possesses PthXo3/AvrXa7, while two strains from the northwestern region can activate both OsSWEET11a and unidentified susceptibility gene/s simultaneously, as observed in Xoo KP-22 and LD-5. A parallel study of northeastern China reveals that most Xoo strains activate OsSWEET13 and OsSWEET14; some strains can concurrently activate OsSWEET11a, OsSWEET13, and OsSWEET14, exemplified by Xoo LN4, LN2, and LN18 (Liu et al. 2024). The regional differentiation is more prominent in the vast rice cultivation region in southern China, where the majority of strains selectively target OsSWEET14 (Liu et al. 2024). In environments with low selection pressure, favorable conditions facilitate the development of an initial inoculum load, leading to plant infections. Additionally, genetic flow plays a crucial role in the dissemination of pathogens among diverse field populations. These findings inspire innovative approaches in rice cultivation, such as developing cultivars with multiple mutations in EBEs, potentially conferring broad-spectrum resistance. Moreover, considering the expression patterns of OsSWEET11a, OsSWEET13, and OsSWEET14 in response to Xoo, coupled with geographical origin, offers insights into identifying effective editing variants for optimizing rice cultivation practices (Eom et al. 2019).

Although it is tempting to associate the presence of a specific band with virulence, such an association remains highly speculative. Therefore, a more valuable approach involves isolating or engineering an Xoo strain devoid of TALEs. This strain could then be utilized to reintroduce TALE genes systematically, allowing for the thorough dissection of the individual contributions of each TALE gene to virulence (Shah et al. 2019; Haq et al. 2020). Molecular identification of TALE-induced target genes in host plants becomes imperative for a comprehensive understanding of the molecular mechanisms underlying TALE-promoted diseases (Boch et al. 2014). The substantial progress in TALE biology has significantly enhanced our comprehension of the interplay between TALEs and their host targets. This progress has paved the way for the development of innovative methods for identifying TALE-targeted EBEs, including TALE-code-based EBE prediction, β-glucuronidase assay, electrophoretic mobility shift assay, and designer TALEs (Antony et al. 2010; Li et al. 2013; Haq et al. 2021; Shah et al. 2023).

In this study, we provided insights into the TALome diversity and aggressiveness of the Pakistani Xoo population. Our findings aim to bridge existing gaps in understanding whether TALE diversity is primarily linked to geographic regions. Notably, we identified that the rice line containing Xa23 is the most effective variety for breeding in Pakistan. Furthermore, genotypes G8 and G11, represented by KP-22 and LD-5, respectively, harbor an unknown virulence factor/s distinct from other tested Xoo genotypes/strains. These genotypes can overcome the resistance conferred by the MS134K rice variety, known for its broad-spectrum resistance to BB (Eom et al. 2019; Oliva et al. 2019; Xu et al. 2019; Liu et al. 2024). This suggests that KP-22 and LD-5 may represent emerging minority strains that have naturally co-evolved effectors to challenge the resistance pressure in rice fields. In general, strains with a higher number of putative TAL effector genes exhibited increased virulence compared to strains with fewer TAL genes. Additionally, expression analysis revealed co-evolved alleles targeting unidentified susceptibility gene/s that contribute to disease development. Overall, our study highlights the existence of gene-for-gene relationships between the tested rice lines and Pakistani Xoo strains. This marks the initial report presenting the diversity of TALEs and their association with R and S genes. Further efforts to identify novel virulent TALE/s and their target/s in cv. Kitaake is imperative to precisely elucidate their role in BB.

Conclusions

This study provides the first comprehensive analysis of the TALome diversity in the Pakistani Xoo population and its association with rice R and S genes. Pakistani Xoo strains exhibited extensive TALE diversity, with 16–20 putative TALE fragments and 11 distinct genotypes classified into five clades, reflecting geographic distribution with some exceptions. Notably, genotypes G8 (strain KP-22) and G11 (strain LD-5) were identified as harboring novel virulent TALEs that target previously unidentified susceptibility genes, overcoming broad-spectrum resistance in MS134K rice. Virulence assays revealed the effectiveness of Xa23-containing rice lines against all tested strains, emphasizing its potential for breeding in Pakistan. However, the emergence of strains capable of circumventing widely deployed resistance genes highlights the need for continued surveillance and functional characterization of novel TALEs. These findings underscore the gene-for-gene relationships between Pakistani Xoo strains and rice lines, emphasizing the necessity of identifying virulent TALEs and their targets to refine resistance breeding strategies and ensure sustainable management of bacterial blight.

Methods

Bacterial strains, and growth conditions

Bacterial strains used in this research are listed in Additional file 1: Table S1. Xoo strains were cultured in nutrient broth (NB; 5 g polypeptone, 10 g sucrose, 1 g yeast extract, and 3 g beef extract in one liter of distilled water, pH = 7.0–7.2) or NB with 1.5% agar (NA) at 28°C. Antibiotics were used at the following concentrations (μg/mL) when required: ampicillin (Ap) 100 μg/mL and spectinomycin (Sp) 40 μg/mL.

Bacterial strains collection

Surveys to collect the bacterial blight disease specimens were conducted between September and November of 2017, 2018, 2019, and 2021 in the key rice production regions of two provinces of Pakistan. In Punjab (P) province; Hafizabad, Sheikhupura, Sialkot, Narowal, and Gujranwala districts were surveyed (Additional file 2: Figure S1). In Khyber Pakhtunkhwa (KP) province; surveys were conducted in Mansehra, Battagram, Bannu, Swat, and Lower Dir districts (Additional file 2: Figure S1). Disease leaf samples were collected based on the presence of typical BB symptoms from cultivated rice varieties.

For the isolation of Xoo, lesion segments were subjected to disinfection using 70% ethanol and were subsequently rinsed three times with sterilized distilled water. The samples were then cut into small pieces using sterilized scissors and immersed in double distilled water for one hour at room temperature to facilitate the release of bacteria. The resulting suspension was then streaked onto fresh plates of NA followed by incubation at 28°C for a period of 5–7 days. Single pure colonies were screened molecularly through PCR using pathovar-specific primers (Xoo80F/Xoo80R) (Lang et al. 2010). Validated Xoo strains were preserved in 50% glycerol at − 80°C. A total of 101 Pakistani Xoo strains were used in this study, along with three reference strains, two from Philippines (PXO99^A and PXO86), and one from China (LN18).

Southern blot hybridization

The genomic DNA was extracted using the HiPure Bacterial-DNA extraction Kit (Magen, Guangdong, Guangzhou, China) following the manufacturer's protocols. The quality and quantity of the isolated DNA were checked with NanoDrop One spectrophotometer (Thermo Fisher Scientific, Waltham, USA). An aliquot of 10 µg of DNA was digested with BamHI (3 µg/µL) restriction endonuclease (Takara, Japan) and incubated at 37°C for 5 h. The digested DNA was separated on 1.3% agarose gel in 1 × TAE buffer (50 × = Tris 242 g, Na₂EDTA·2H₂O 37.2 g, and 57.1 mL CH₃COOH in one liter distilled water, pH = 8.5) at 80 V, 4°C for 20 h. Subsequently, the gel was kept in denaturation buffer (0.5 M NaOH and 1.5 M NaCl) and neutralization buffer (0.5 M Tris–HCl and 1.5 M NaCl, pH = 7.5), shaking for 45 and 15 (twice) minutes, respectively. The DNA fragments were then blotted from the gel on a presoaked Hybond N⁺ nylon membrane (Millipore, Billerica, MA, USA) using standard saline citrate buffer (tri-sodium citrate di-hydrate 88.2 g and NaCl 175.3 g, pH = 7.0) as capillary transfer solvent. The Xoo strain PXO99^A PthXo1, i.e., pZY-PthXo1 (Additional file 1: Table S1) internal SphI fragment (2892 bp; containing 24 RVDs) was labeled with digoxigenin (DIG) and employed as a hybridization probe for the detection of TALE fragments. Hybridization was done by adding the labeled probe into the DIG Easy hybridization buffer and incubating it in a rotatory hybridization machine at 68°C overnight. Probe labeling, hybridization, blocking, and detection were conducted following the manufacturer's instructions (Roche Diagnostics GmbH Mannheim, Germany). The Xoo PXO99^A TALE-free strain PH harboring pHZY-PthXo1, pHXY-PthXo2, and pHZY-AvrXa7 in trans were used to point out the major virulence tal genes (Additional file 1: Table S1).

Genetic similarity based on TALE repertoires was computed using the NTSYSpc-2.02e. The number and size of putative TALEs were converted into binary form, where a value of 1 denotes the presence of a specific band, and 0 indicates the absence of the corresponding band. The results were subjected to generating the dendrogram using the UPGMA method within the SimQual similarity and Sequential Agglomerative Hierarchical Nested (SAHN) clustering module of NTSYS 2.02e as described by Jamshidi and Jamshidi (2011).

Pathogenicity analysis

Rice seeds were soaked in water and incubated at 37°C for 48–72 h. The seedlings were then transferred into the greenhouse and, after 2–3 weeks, planted in the field. Oryza sativa subsp. indica/japonica containing different R genes (IR24, IRBB3, IRBB4, IRBB5, CBB23, and Kit-Xa1) and Oryza sativa subsp. japonica containing EBE mutants of OsSWEET11a, OsSWEET13, and OsSWEET14 [cv. Kitaake, MS1K (OsSWEET11a mutant), MS3K (OsSWEET13 mutant), MS4K (OsSWEET14 mutant), MS13K (OsSWEET11a and OsSWEET13 mutant), MS14K (OsSWEET11a and OsSWEET14 mutant), MS34K (OsSWEET13 and OsSWEET14 mutant), and MS134K (OsSWEET11a, OsSWEET13, and OsSWEET14 mutant)] were grown in the field conditions at Shanghai Jiao Tong University, Shanghai, China.

To investigate whether there are rice lines resistant to Pakistani Xoo strains, 14 NILs containing different R genes and S genes-edited rice lines were used in the virulence assessment of the collected Pakistani strains. Xoo strains grown overnight were employed to infect two-month-old rice plants (at the booting stage) using the tip-cutting method. Disease symptoms were assessed 14 days post-inoculation (dpi), and the lesion length (cm) was recorded. More than five flag leaves were inoculated with each Xoo strain, and experiments were repeated twice. Standard deviation analyses were performed on all measurements. Strain aggressiveness and disease severity were evaluated based on lesion length index as follows: measurements 0–3 cm were scored as resistant (R), 3–5 cm as moderately resistant (MR), 5–8 cm as moderately susceptible (MS), and > 8 cm as susceptible (S).

Confirmation of Kitaake mutant rice lines

The three major susceptibility genes’ EBE mutant plants (single, double, and triple mutants) in the background of cv. Kitaake were generated by Liu et al. (2024) for tracing the major virulence TAL effectors in Xoo strains. To confirm the EBE-edited plants, we isolated the genomic DNA from all the mutants using the CTAB (cetyltrimethylammonium bromide) method. The OsSWEET11a, OsSWEET13, and OsSWEET14 promoter regions harboring EBEs were PCR amplified using primer sets SW11p-F/SW11p-R, SW13p-F/SW13p-R, and SW14p-F/SW14p-R, respectively (Additional file 1: Table S2). The PCR amplicons were sequenced by Sanger sequencing and comparatively analyzed (Additional file 1: Table S4).

Expression analysis of six OsSWEET genes

The indicated bacterial strains (washed twice and re-suspended in double distilled water, OD₆₀₀ = 0.6) were inoculated into four-week-old rice seedlings (grown in the greenhouse with a photoperiod of 14 h light and 10 h of dark at 25°C) using a needleless syringe, and samples were collected at 24 h after inoculation. Total RNA was extracted using RNAiso Plus (TAKARA BIO INC., Japan) reagent according to the manufacturer's protocol. The quality and quantity were assessed using NanoDrop One spectrophotometer (Thermo Fisher Scientific, Waltham, USA). An aliquot of 1 µg RNA from each sample was reverse transcribed into cDNA using TransScript One-Step gDNA Removal and cDNA Synthesis SuperMix Kit (TransGen Biotech Beijing, China). The qRT-PCR was performed on Applied Biosystems 7500 Real-Time PCR System using TransStart Tip Green qPCR Supermix (+ DyeII) Kit (TransGen Biotech Co., Ltd. Beijing, China). The PCR conditions were 30 s at 95°C, followed by 40 cycles at 95°C, 60°C, and 72°C for 10 s, 34 s, and 15 s, respectively. The relative expression levels of OsSWEET11a, OsSWEET11b, OsSWEET12, OsSWEET13, OsSWEET14, and OsSWEET15 were calculated with the 2^−ΔΔCT method and normalized to the expression of rice Actin gene. The primers used in qRT-PCR are listed in Additional file 1: Table S2.

Statistical analysis

The genomic sequence of the single available Pakistani Xoo strain, PkXoo1 (Genbank accession no. CP101721.2), was retrieved from the National Center for Biotechnology Information. TALEs were annotated and analyzed using AnnoTALE v1.2 (Grau et al. 2016). The BamHI fragments of tale sequences were simulated in gel using SnapGene 6.0.2 and compared with the Southern blot results to understand the intensity of bands and possible RVDs of TAL effectors (data not shown). Strains clustering based on the number and size of TALEs and major virulent TALEs were conducted using the UPGMA method in NTSYSpc-2.02e, as described earlier.