A PthXo2B ortholog inXanthomonas oryzaepv oryzae strain IX-221 acts as a major virulence factor onindicarice without activating a Clade IIISWEETgene

46 47 To infect rice, Xanthomonas oryzae pv. oryzae (Xoo) deploys transcription activator-like 48 effectors (TALEs) that specifically bind and upregulate host “susceptibility” (S) genes. 34-49 amino acid (aa) repeats in TALEs interact one- to-one with DNA bases. V ariation at 50 positions 12 and 13 in each repeat, the repeat-variable diresidue (RVD), determine 51 specificity. Some repeat variants shorter or longer than 34 aa can disengage to 52 accommodate a single base deletion in the target sequence. OsSWEET11, 13, and 14 53 are key S genes, targeted by different TALEs from diverse Xoo strains. xa13 is a 54 SWEET11 allele lacking the TALE binding site and thus conferring resistance. xa13 is 55 overcome by TALEs that activate SWEET13 or SWEET14. We report here that an xa13-56 compatible Xoo strain, IX-221, from India, harbours an ortholog of the SWEET14-57 targeting TALE PthXo3 and two orthologs of the SWEET13-cognate PthXo2, each with 58 one or two 36-aa repeats capable of disengaging. One of the PthXo2 orthologs , 59 PthXo2BIX-221, has a repeat region identical to the previously characterized PthXo2BPXO61, 60 except for a two amino acid difference near the end of the 19th repeat. Like PthXo2BPXO61, 61 PthXo2BIX-221 upregulates SWEET13 in japonica rice and no SWEET in indica rice, but 62 unlike PthXo2B PXO61 it nonetheless renders indica rice susceptible, pointing to an 63 alternative S gene . Further, a designer TALE (dTALE) constructed using a standard, 64 consensus sequence for each repeat and RVDs identical to those of PthXo2B IX-221 failed 65 to render indica rice susceptible. Alignment of the PthXo2B IX-221 repeats shows a 66 departure from the consensus in each of two repeats carrying the RVD ‘NN’: the sequence 67 ‘MAIAN’ in place of ‘VAIAS’ beginning at position 7. Together, the PthXo2B IX-221 results 68 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint thus suggest that non-RVD sequence variation affects TALE targeting profiles . More 69 broadly, the presence of the three aberrant repeat-harbouring TALEs in IX-221 suggests 70 that widespread deployment of xa13 in India resulted in strains super-equipped to 71 overcome it, capable of activating multiple SWEET genes and alleles as well as an 72 apparent alternate S gene. 73 74 Key Words: Xanthomonas oryzae, transcription activator-like effector (TALE), aberrant 75 repeat, xa13, bacterial blight of rice, susceptibility gene, SWEET sugar transporter 76 77 78

79 80 Xanthomonas oryzae pv. oryzae (Xoo) causes bacterial blight disease of rice (BB). It is 81 important economically and as a model to study host-pathogen interaction and 82 coevolution (Hutin et al., 2015, Nino-Liu et al., 2006). Xoo injects DNA-binding proteins 83 called transcription activator-like effectors (TALEs) into rice cells, where they localize to 84 the nucleus and increase the transcription of host genes by binding to effector -specific 85 promoter sequences called effector binding elements (EBE) (Bogdanove et al., 2010). 86 Those genes that contribute to disease development when so upregulated are considered 87 susceptibility (S) gene s. Binding of an EBE by a TALE is governed by a central repeat 88 region (CRR) of the protein composed of nearly identical direct repeats of 33 -35 amino 89 acids (aa), differing at the 12 th and 13 th positions, together called the repeat variable 90 diresidue (RVD). Following a partially degenerate code, each RVD directly interacts with 91 a single nucleotide, such that the number and composition of RVDs predicts the sequence 92 of the EBE (Moscou & Bogdanove, 2009, Boch et al., 2009). Besides the standard 33-35 93 aa repeats, variants with repeat lengths of 28, 30, 36, 39, 40 and 42 aa have been 94 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint reported (Richter et al., 2014b, Wilkins et al., 2015, Oliva et al., 2019b). Some of these 95 so-called aberrant repeats can function as a standard repeat or disengage in an as yet 96 structurally uncharacterized way to accommodate a single base pair deletion at the 97 corresponding location in the target (Richter et al., 2014b, Becker et al., 2022). 98 A major class of S gene targeted by TALEs in BB consists of clade III members of 99 the ‘SWEET’ sucrose transporter gene family (Streubel et al., 2013). The cognate TALEs 100 are referred to as major TALEs, owing to their important contribution to virulence as 101 activators of these major S genes. The first such TALE identified was PthXo1, which 102 induces SWEET11, also called Os8N3 or Xa13 (Chu et al., 2006, Yang et al., 2006). 103 Recessive alleles of SWEET11, collectively known as xa13, can confer resistance (rather, 104 loss of susceptibility) to strains that depend on PthXo1 , by virtue of a ny of several 105 promoter mutations that disrupt the PthXo1 binding site . These xa13 SWEET11 alleles 106 are ineffective against Xoo strains with TALEs that activate other Clade III SWEET genes, 107 such as PthXo2 , which activates SWEET13 (Zhou et al., 2015, Oliva et al., 2019b) , or 108 AvrXa7, PthXo3, TalC, and TALE5, which activate SWEET14 (Antony et al., 2010, Yu et 109 al., 2011, Streubel et al., 2013) . While PthXo2 activates SWEET13 and acts as a major 110 TALE only on indica rice, a variant named PthXo2B with aberrant 9th and 12th repeats of 111 36 aa, found in a few Xoo strains, activates the japonica, and not the indica allele of 112 SWEET13 (Oliva et al., 2019b) and confers virulence only toward japonica rice. The key 113 roles of SWEET11, 13, and 14 in bacterial blight of rice, inspired the development of lines 114 of select indica and japonica mega rice varieties edited at EBEs in all three to provide 115 broad spectrum bacterial blight disease resistance (Oliva et al., 2019b, Xu et al., 2019). 116 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint The naturally occurring SWEET11 allele xa13 has been widely deployed in India, 117 but the Indian Xoo population is quite diverse, and strains that overcome xa13 have been 118 reported from different parts of the country (Lore et al. , 2011, Mishra et al. , 2013, 119 Yugander et al., 2017). (Midha et al., 2017, Mondal et al., 2014). In individual Xoo strains, 120 TALEs are typically numerous (15 or more), and within and across strains they are 121 diverse. Their abundance and the repetitive nature of their coding sequences likely 122 contribute to the diversity, facilitating recombination and rapid adaptation under selection 123 pressure (Booher et al., 2015, Denancé et al., 2018). Detailed molecular characterization 124 of resistance-breaking strains is essential for insight into pathogen-host coevolution, and 125 knowledge of the ways in which deployed resistance genes are overcome can guide 126 future resistance development and deployment strategies. 127 For this reason, we aimed to characterize an Indian Xoo strain, IX-221, associated 128 with an outbreak on xa13-containing rice in an experimental field in the state of Haryana 129 (Yugander et al., 2017) , by fully sequencing the genome to evaluate TALE content and 130 to functionally characterize its major TALEs. We report here that Tal7/PthXo2BIX-221 is a 131 major virulence factor in both japonica and indica rice, yet activates SWEET1 3 only in 132 japonica rice. It activates no clade III SWEET gene in indica. Further, a designer TALE 133 with the same RVD sequence as PthXo2BIX-221 acted as a virulence factor only in japonica 134 rice, pointing toward an influence of repeat backbone sequence variation on DNA 135 targeting capacity. 136 137

138 139 Genome sequencing, assembly and annotation 140 141 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint For sequencing, genomic DNA was extracted and a 20 kb library prepared as previously 142 described (Booher et al., 2015, Carpenter et al., 2020). Two single molecule real time 143 (SMRT) cells were used to sequence the library on a PacBio RSII machine (Pacific 144 Biosciences, Menlo Park, CA USA). The sequence reads were de novo assembled using 145 HGAP 3.0 and SMRTAnalysis 2.3, and the assembly was verified using local tal gene 146 assembly with PBX (Booher et al., 2015) and the variant finder PBHoney from PB Suite 147 14.7.14 (English et al., 2014). The verified whole genome sequence was annotated using 148 the National Centre for Biotechnology Information (NCBI) with Prokaryotic Genome 149 Annotation Pipeline (PGAP) (Tatusova et al., 2016). 150 151 Strains, primers, plant material and inoculations 152 The bacterial strains used in this study were E coli DH5), Xanthomonas oryzae pv oryzae 153 (Xoo) strains IX-221, PXO99A (ME2) and Agrobacterium tumefaciens (At) GV3101. E coli 154 and At cells were grown in Luria-Bertani (LB) medium at 37C and 28C respectively, Xoo 155 strains at 28C in GYE (20g/l Glucose, 10g/l yeast extract). Plasmids were introduced into 156 E coli by heat shock and by electroporation into Xoo and At. Antibiotics were used at the 157 following concentrations: Ampicillin, 100 g/ml; spectinomycin, 50 g/ml; kanamycin, 158 25g/ml; tetracycline, 10 g/ml for E coli and 2 g/ml for Xoo. Primers used in the study 159 are provided in Table S4. 160 Rice plants were grown in a growth chamber maintained at 28 C and 85% relative 161 humidity (RH) with a photoperiod of 12h. Oryza sativa ssp. japonica cv Nipponbare and 162 the near-isogenic O sativa ssp. indica line IR24 were used for disease assays and gene 163 expression assays as described (Carpenter et al., 2020) . Nicotiana benthamiana (Nb) 164 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint plants were grown under 16 h of light, 50% RH, at 25:20°C, day : night in the growth 165 chamber. Fully expanded eaves of 5 weeks old plants were inoculated with At strains with 166 constructs using a needleless syringe. Significant differences were determined using the 167 paired Student’s t-test. Experiments were repeated thrice. 168 169 TALE analysis, target prediction, cloning and generation of TALE and dTALE 170 constructs 171 All tal gene sequences were extracted and their orthology to previously sequenced TALEs 172 were determined using the AnnoTALE (Richter et al. , 2014a). The tal gene repertoires 173 were verified by Southern blots of genomic DNA digested with SphI, and probed with the 174 tal gene specific probe pZWavrXa7. 175 Targets of Tal2b, Tal6b and Tal7 were predicted using the TALE-NT 2.0 Target Finder 176 tool (Doyle et al., 2012). Predictions were made for both forward and reverse strands of 177 promoter sequences, defined 1000 bp upstream of the translational start site for TALE-178 NT 2.0, and using MSU Rice Genome Annotation Project Release 7 179 (http://rice.plantbiology.msu.edu/). Default settings were used for Target Finder 180 (upstream base of binding site = T, score cutoff = 3.0, Doyle et al. scoring matrix). 181 For cloning tal genes , a subgenomic library of IX-221 digested with BamHI was generated 182 in pBS II (KS-) and screened for tal positive clones by PCR and Sanger sequencing. The 183 Tal2b, Tal6b and Tal7 were further subcloned as SphI fragment into an entry vector 184 pCS466 on the SphI site flanked by N- and C-termini of Xo pv oryzicola ( Xoc) strain 185 BLS256 tal1c gene (Verdier et al 2012). The dTALE derivatives of Tal6b and Tal7 were 186 assembled using Golden Gate kit (Cermak et al., 2011), into another entry vector pTAL1, 187 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint also encoding the Xoc tal1c without its central repeat region. The complete tal and dtal 188 genes thus reconstituted were then transferred to broad host range destination vector 189 pKEB31(Cermak et al., 2011) by Gateway LR clonase reaction (Invitrogen/Thermo Fisher 190 Scientific) for expression in Xoo strain ME2 and in pGWB5 for expression in At strain 191 GV3101. 192 GUS assay 193 GUS reporter constructs were generated by cloning the SWEET13 EBEs of IR24 and 194 Nipponbare on the unique AscI site of binary vector pCS752, having the pepper Bs3 195 promoter driving uidA reporter gene expression with the AscI site upstream of the native 196 AvrBs3 binding element. At GV3101 transformed wit TALE and dTALE constructs and 197 GUS reporter constructs were resuspended to an OD600 of 0.8 in 10 mM MgCl2 with 150 198 µM of acetosyringone and mixed 1:1 for inoculation into fully expanded leaves of 5week 199 old Nicotiana benthamiana plants. Leaf discs were samples at 48hpi for qualitative and 200 quantitative assays as previously described (Römer et al., 2009, Carter et al., 2020). 201

202 IX-221 harbors three major TALEs, each with one or two aberrant repeats 203 We first confirmed the xa13 compatibility of IX-221 by inoculating to rice line IRBB13, 204 homozygous for xa13, with the near isogenic parent IR24 used as a susceptible control, 205 and found IX -221 indeed to be compatible with xa13 (data not shown). We then 206 sequenced the whole genome of IX-221 and assembled the data de-novo. The assembly 207 yielded a genome consisting of a single, 4.9 Mb circular chromosomal contig with 63.7% 208 G+C content with 172X average sequence coverage. The genome, li ke those of other 209 Xoo strains, contains hundreds of IS elements, which contribute to genomic plasticity 210 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint (Table S1). The number and sizes of TALE genes in the IX-221 genome was confirmed 211 by Southern blot (Figure S1). There are 20 encoded TALEs, of which 18 fall into existing 212 TALE classes as defined by AnnoTALE version. 1.5 (Grau et al. , 2016) (Table S2), 213 including a TruncTALE (also called iTALE), a class of TALEs with truncated N- and C- 214 termini that suppress resistance mediated by Xo1 or Xa1 (Ji et al. , 2016, Read et al. , 215 2016). Of the two IX-221 TALEs classified as new by AnnoTALE, one contains only five 216 RVDs. The other is orthologous with the major TALE PthXo3, based on analysis using 217 FuncTALE, which groups TALEs by predicted DNA target sequences (Pérez-Quintero et 218 al., 2015). Strikingly, in addition to the PthXo3 ortholog, IX-221 also harbors two orthologs 219 of another major TALE, PthXo2, and each of these three TALEs has one or two aberrant 220 repeats (Table S2). None of the IX- 221 TALEs is a PthXo1 ortholog, consistent with the 221 compatibility of the strain with xa13. 222 The PthXo3 ortholog, Tal2b IX-221 (AnnoTALE class IU), like PthXo3, has a 39- aa 223 repeat that in PthXo3 is important for frameshift binding to the target (Richter et al., 224 2014a). Because it is a variant of PthXo3, distinct from the three observed to date [Oliva, 225 2019], we hereafter refer to it as PthXo3DIX-221. With the full RVD sequence of PthXo3DIX-226 221 as input, using TALE-NT 2.0 (Doyle et al., 2012) and the rice cv. Nipponbare reference 227 genome sequence, no binding site in the SWEET14 promoter was predicted, but with the 228 RVD of the 39-aa repeat excluded, PthXo3DIX-221 was predicted to bind the PthXo3 EBE 229 (Figure S2). As expected, IX -221, and the pthXo1 mutant derivative ME2 of Xoo strain 230 PXO99A (Yang & White, 2004) carrying PthXo3DIX-221 on a plasmid, each induced 231 SWEET14 in Nipponbare leaves (Figure S3). Aligned to the Nipponbare EBE, PthXo3 232 has a better predicted binding score ratio (Doyle et al., 2012) than PthXo3IX-221 (Figure 233 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint S2). However, PthXo3D IX-221 includes more RVDs with relaxed or semi-relaxed base 234 specificity. Namely, it has three NS, an RVD that accommodates A, C, or G; (Yang et al., 235 2014), and one NN, which recognizes G or A; that flexibility may allow activation of yet 236 uncharacterized SWEET14 alleles with variations at the corresponding positions in the 237 EBE. The flexibility may in fact have been selected for, by the presence of such alleles. 238 The PthXo2 orthologs are Tal6bIX-221 and Tal7IX-221. PthXo2 variants reported to 239 date include PthXo2B and PthXo2C (Oliva et al., 2019a). While PthXo2 has standard, 240 34 aa repeats throughout, both PthXo2B and 2C have 36 aa in their 9th and 12th repeats 241 and differ in a few RVDs from PthXo2. Tal6bIX-221, with 36 aa only in its 12th repeat is a 242 novel variant. Alignment of the RVD sequences positions Tal6b between PthXo2, and 243 PthXo2B and 2C together (Figure S4). We hereafter refer to Tal6bIX-221 as PthXo2DIX-244 221. Tal7IX-221 has 36-aa in its 9th and 12th repeats. In fact, it is identical to PthXo2B from 245 strain PXO61 (PthXo2BPXO61) except that, relative to Tal7IX-221 , PthXo2BPXO61 has a 1 246 aa insertion (glycine) 32 aa from the N-terminus (at position 33), in the region 247 associated with type III secretion (Szurek et al., 2002), and a 2 aa substitution at 248 positions 31 and 32 of repeat 19. The 2 aa substitution replaces glutamine and aspartic 249 acid with arginine and alanine; while aspartic acid and alanine are each common at 250 position 32 in TALE repeats, the arginine at position 31 is unusual. We hereafter refer to 251 Tal7IX-221 as PthXo2BIX-221. In all three of the 36-aa repeats in the IX-221 TALEs, the 252 canonical proline and valine residues at positions 29 and 30 are repeated (as a pair). 253 254 PthXo2DIX-221 activates SWEET13 and confers virulence on indica rice only 255 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint Given the seeming redundance of PthXo2D IX-221 and PthXo2B IX-221 with the SWEET14 256 activator PthXo3IX-221 in IX-221, we questioned whether these PthXo2 orthologs are in 257 fact activators of SWEET13. Beginning with PthXo2DIX-221, we first determined the binding 258 score ratio for its RVD sequence on the SWEET13 allele present in the japonica variety 259 Nipponbare and the allele in the indica variety IR24, using the target finder tool of TALE-260 NT 2.0 (Doyle et al., 2012) . Because the 3 6 aa repeat is an aberrant type capable of 261 disengaging (Becker et al., 2022), we determined also the score ratio using the sequence 262 with the RVD of that repeat excluded. Using a score ratio of 3 or less as a cutoff for 263 predicted binding, PthXo2DIX-221 is expected to bind well to the IR24 allele with its 36 aa 264 repeat engaged and marginally to the Nipponbare allele with it excluded ( PthXo2D and 265 PthXo2D.1, respectively, Figure 1a). 266 Next, to test the predictions, we cloned the native pthXo2DIX-221 central repeat 267 region (as an SphI fragment), between the flanking N - and C-terminal regions of Tal1c 268 from X oryzae pv. oryzicola strain BLS256 in the expression vector pKEB31 (Cermak et 269 al., 2011), and assembled a designer TALE (dTALE) construct encoding PthXo2D IX-221 270 with its aberrant repeat replaced by a standard repeat, also using the Tal1c context. We 271 designated these constructs n T2D (nT for native TALE) and d T2D_34 (dT for dTALE), 272 respectively. To control for any effect of differences in the dTALE repeat backbone 273 sequences from the native ones, we also generated a dTALE equivalent of PthXo2DIX-221 274 i.e., with its 12th repeat 36 aa in length, dT2D_36. We tested each of these constructs in 275 ME2 inoculated to IR24 and Nipponbare plants. ME2 carrying nT2D and its equivalent 276 dTALE: dT2D_36 induced SWEET13 and caused long lesions when inoculated to IR24 277 (Figure 1b, c and d). In Nipponbare, neither detectably induced SWEET13 or caused long 278 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint lesions typical of a fully virulent strain. However, ME2 with nT 2D did elicit lesions longer 279 than the negative control, ME2 transformed with pAC99, a plasmid encoding a TALE with 280 the CRR removed (Cernadas et al., 2014) (Figure 1b and c). d2D_34, with the standard 281 repeat, behaved the same as PthXo2, inducing SWEET13 and restoring virulence to ME2 282 only in IR24, with no virulence increase relative to the control in Nipponbare. We therefore 283 infer that PthXo2DIX-221 is functionally distinct from PthXo2, which has no aberrant repeats 284 and restores virulence to ME2 only in indica rice (Zhou et al., 2015). 285 As a complementary approach and to confirm binding, we used Agrobacterium-286 mediated transient transformation in Nicotiana benthamiana leaves, as described (Römer 287 et al., 2009), to test whether nT2D, dT2D_34, and dT2D_36 could activate GUS reporter 288 constructs driven by a minimal promoter from the pepper Bs3 gene (Römer et al., 2009) 289 amended either with the PthXo2 EBE from the Nipponbare allele of SWEET13 or with the 290 EBE from the IR24 allele. The TALE AvrBs3, which activates the minimal Bs3 promoter, 291 was used as a positive control, and PthXo1, which has no EBE in either reporter construct, 292 was used as a negative control. nT2D, d T2D_34, and d T2D_36 strongly induced the 293 reporter harboring the IR24 EBE, and not the Nipponbare EBE (Figure 1e), validating the 294 observations made in rice leaves, and also indicating PthXo2DIX-221 binds the SWEET13 295 allele in IR24 without its 36-aa repeat disengaging. 296 Why n2D partial ly rescued ME2 in Nipponbare while d2D_36 did not is unclear. 297 Perhaps n2D, and not d2D_36, activates SWEET13 marginally enough so as not to be 298 statistically significant but sufficiently to partially restore virulence . Such a difference in 299 activity between the two proteins might derive from minor differences in target affinity due 300 to differences between the native and the designer repeat backbone sequences. 301 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint 302 FIGURE 1. PthXo2DIX-221 acts as a major TALE in indica rice variety IR24 without 303 relying on its 36-aa aberrant repeat to disengage. (a) TALE-NT 2.0-based prediction of 304 binding of PthXo2D IX-221, and a variant with the aberrant repeat removed, to SWEET13 305 alleles found in rice cv. Nipponbare (NP) and IR24. RVDs mismatching the aligned base 306 are highlighted in grey and that of the 36-aa aberrant repeat is lowercase. The position from 307 which the RVD of the aberrant repeat was omitted for the prediction is indicated by a vertical 308 line. A score ratio (ratio of observed EBE score to the best possible score for the TALE) of 309 3 was used as cut off for p redicted binding. A ‘ -’ indicates a score above cut -off, i.e., not 310 predicted to bind. (b) R epresentative images and (c) lesion lengths on leaves of 6 -week-311 old rice (cv. NP and IR24) plants 14 days after clip inoculation with ME2 expressing the 312 indicated TALE or a negative control TALE lacking a central repeat region (pAC99). (d) 313 Expression of SWEET13, measured by RT -quantitative real time PCR, in NP and IR24 314 leaves 24 -27 hr following syringe infiltration of ME2 expressing the indicated TALE or 315 control, relative to mock-inoculum. (e) Reporter assay of EBE binding by nT2D and dTALE 316 variants as indicated. N. benthamiana leaves were co -infiltrated with Agrobacterium 317 tumefaciens strains delivering the indicated TALE or dTALE construct and a GUS reporter 318 construct driven by a minimal Bs3 promoter containing the indicated EBE, and GUS activity 319 assayed 48 hr later. Shown are means for three independent infiltrations. Error bars 320 represent standard deviation. A representative leaf disc is shown for each. 321 322 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint PthXo2BIX-221 renders both indica and japonica rice susceptible, and does not rely 323 on any Clade III SWEET gene in indica 324 Following the same approach for PthXo2BIX-221, we determined using TALE-NT 2.0 that it 325 is likely to bind only the japonica allele, and only when the RVD of one or the other of its 326 36 aa repeats is excluded (PthXo2B.1 and PthXo2B.2, respectively, Figure 2a). Similar 327 to PthXo2D IX-221, the expression constructs assembled for PthXo2B IX-221 included the 328 native CRR, a designer equivalent with 3 6 aa 9 th and 12th repeats, and CRRs with one 329 the other or both aberrant repeats converted to standard, 34 aa repeats, n T2B, 330 dT2B_36_36, dT2B_34_36, dT2B_36_34, and dT2B_34_34, respectively. 331 nT2B in ME2, as predicted, induced SWEET13 and restored virulence in 332 Nipponbare, and did not upregulate SWEET13 in IR24 (Figure 2b, c and d). Results with 333 the dTALEs in ME2 indicated that activation of SWEET13 in Nipponbare depends on 334 either one or the other of the 36 aa repeats disengaging: dT2B_36_36, dT2B_34_36, and 335 dT2B36_34 each activated SWEET13 and restored virulence to ME2 in Nipponbare, 336 while the dTALE with both repeats replaced, dT2B_34_34, did not. Also, like nT2B, none 337 of the dTALEs activated SWEET13 in IR24 (Figure 2d). The EBE binding assay results 338 aligned with these observations, with all but dT B_34_34 strongly induc ing the GUS 339 reporter with the Nipponbare EBE only, thus confirming that PthXo2BIX-221 binds only the 340 Nipponbare allele and that it relies on either one of its two aberrant repeats disengaging 341 to do so (Figure 2e). This conclusion is consistent with that of Becker and colleagues 342 (2022) based on GUS reporter assays using synthetic TalBK2 (Anno TALE class for 343 PthXo2B) and variants missing either repeat. 344 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint Surprisingly, despite nT2B not activating SWEET13 in IR24, it nonetheless fully 345 restored virulence of ME2 in that variety (Figure 2b and c), and it did so without activating 346 any other clade III SWEET gene (Figure 2f). This contrasts with what Xu et al (2019) and 347 Oliva and colleagues (2019a) reported with PthXo2BPXO61, which has the same sequence 348 of RVDs as PthXo2BIX-221: that TALE did not restore virulence to ME2 in IR24. As noted 349 above, there is a 1 aa insertion in the N-terminal region and a 2 aa substitution in repeat 350 19 in PthXo2B PXO61 relative to PthXo2B IX-221. In addition, there are a few minor 351 substitutions in the N-terminal region of PthXo2BPXO61 relative to the fragment of Tal1c 352 used for the PthXo2B IX-221 expression construct (and the dTALEs) . Those substitutions 353 reside in a region that, while essential for type III secretion (Szurek et al., 2002) , is 354 dispensable for DNA binding (Miller et al., 2011). How PthXo2BIX-221 restores virulence to 355 ME2 without activating SWEET13 is unclear. It does not have a predicted binding site in 356 any other clade III SWEET promoter, in either orientation, suggesting involvement of a 357 non-SWEET susceptibility target in indica varieties. Further exploration, beyond the scope 358 of this study , will be necessary to test that hypothesis . In this context however, we 359 tentatively conclude that the non-canonical substitution at repeat 19 changes the 360 specificity or affinity contribution of that repeat, altering the targeting profile such that 361 PthXo2BPXO61 does not activate any alternative S gene. 362 More surprising still, the PthXo2B IX-221-equivalent dTALE dT2B_36_36 failed to 363 restore virulence to ME2 in IR24, just as dT2D_36, the dTALE equivalent of PthXo2D IX-364 221, failed to restore any virulence in Nipponbare, while nT2D, with the native CRR, 365 partially did so. Alignment of the CRRs (Figure S5) revealed some differences. The 366 standard repeat consensus s ubsequence VAIAS present in the dTALEs is replaced by 367 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint MAIAN in the native CRRs, in repeats harboring the RVD ‘NN’ (repeats 4 and 6 of 368 PthXo2BIX-221 and 4, 5 and 7 of PthXo2D IX-221). And, in the 6 th and 7 th repeats of 369 PthXo2BIX-221 and PthXo2D IX-221, respe ctively, the canonical D or A at position 4 is 370 replaced by T. To explore the prevalence of these substitutions, we scanned TALE repeat 371 sequences from diverse randomly picked Xoo genomes as well as genomes of the closely 372 related X. oryzae pv. oryzicola (Xoc), which causes bacterial leaf streak of rice. We found 373 that the NN repeats of all PthXo2 orthologs, from diverse Xoo strains, have the MAIAN 374 subsequence substitution for VAIAS, and that no other repeats in those orthologs or any 375 repeats in other TALEs do (Table S3). Similarly, we found T at position 4 in all PthXo2 376 orthologs, and in no other TALEs. Further, we found each of the two subsequences in 377 TALEs of Xoc (Table S3), and always in an NN repeat, but never together in the same 378 repeat. We hypothesize that the ‘MAIAN’ and ‘T’ substitutions relative to the dTALEs, in 379 PthXo2BIX-221 are important for its ability to act as major TALE in IR24 and in PthXo2D IX-380 221 its ability of to confer some virulence in Nipponbare, ostensibly by altering the targeting 381 profiles to include one or more S genes other than a clade III SWEET. Whether this is 382 indeed the case and how, remains to be explored. 383 384 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint 385 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint FIGURE 2. PthXo2B IX-221 acts as a major TALE in both indica rice variety IR24 and 386 japonica variety Nipponbare (NP) relying on either of its 36-aa repeats to disengage 387 in NP and without activating any Clade III SWEET in IR24. TALE-NT 2.0 -based 388 prediction of binding of PthXo2B IX-221 and variants with one or both aberrant repeats 389 removed, to SWEET13 alleles of rice cv. Nipponbare (NP) and IR24, as in Figure 1. (b-e) 390 as in Figure 1 with nT2B and dTALE variants as indicated. (f) Fold induction, as in (b), of 391 the other clade III SWEET genes by nT2B and selected positive control strains that induce 392 cognate SWEET genes. 393 394 395

396 The Indian Xoo strain IX-221 is armed with three, distinct TALEs that each could allow the 397 strain to cause disease in a host genotype with resistance governed by xa13. Importantly, 398 each of the three TALEs has one or more functional aberrant repeats, which could allow 399 them to accommodate SWEET allelic variation or to target alternative S genes, or both. The 400 first, PthXo3IX-221, is a variant of and functions the same as PthXo3, activating SWEET14 401 in both japonica and indica, with its sole aberrant repeat disengaged. It harbours more 402 RVDs with relaxed specificity than PthXo3 however, suggesting that it may be adapted to 403 as yet unidentified SWEET14 allelic variation, perhaps in landraces not yet characterized. 404 The second is a PthXo2 ortholog, PthXo2B IX-221. We have shown that TALE PthXo2B IX-221 405 is capable of activating SWEET13 in japonica, and ostensibly an alternative S gene in 406 indica, to render genotypes in both subspecies susceptible. Strains harbouring this TALE 407 are thereby expected to circumvent the broad spectrum resistance of recently developed 408 rice varieties with edits at known EBEs in the promoters of SWEET11, -13 and -14 (Xu et 409 al., 2019, Oliva et al., 2019a). Existence of an additional target that functions as an 410 alternative S gene has also been postulated for the SWEET14 inducer T alC from the 411 African Xoo strain BAI3 (Blanvillain-Baufume et al. , 2017); TalC confers virulence even in 412 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint a line in which its EBE in SWEET14 had been disrupted. Unfortunately, because the dTALE 413 equivalent of PthXo2BIX-221 did not render ME2 virulent in the indica cultivar IR24, we were 414 unable to test with our standard repeat derivatives of that construct whether the aberrant 415 repeats are important for the ability of PthXo2BIX-221 to do so. The third TALE in IX-221 that 416 confers an ability to overcome xa13 is PthXo2D IX-221. It functions as predicted, activating 417 the SWEET13 allele and conferring virulence in the indica cultivar IR24. Although the 36-418 aa repeat of PthXo2DIX-221 apparently does not disengage for recognition of the Nipponbare 419 SWEET13 allele, its presence hints at alleles in landraces, yet unidentified, for which the 420 capacity of this repeat to disengage is important. 421 The ability of PthXo2BIX-221 to function as a major TALE in an indica rice genotype, 422 strikingly differentiates it from the previously characterized PthXo2B PXO61. We speculate 423 that PthXo2BPXO61 does not bind and activate the presumed PthXo2B IX-221 target in indica 424 sufficiently to render the plant susceptible. The amino acid alignment of the two, points to 425 the Q to R substitution in PthXo2B PXO61 as the cause for this. As noted, it may change the 426 affinity contribution of the 19 th repeat, reducing affinity overall, or possibly the base 427 specificity of the repeat and thus the targeting profile of the TALE. An alternative, not 428 mutually exclusive possibility is that the substitution makes the 19th repeat like an aberrant 429 one but disengaging in an obligate rather than facul tative fashion. Finally, it is formally 430 possible that the difference is due to a lower amount of PthXo2BPXO61 in the plant cell since 431 that study used the low copy plasmid pHM1 for PthXo2BPXO61 while we used the moderate 432 copy number plasmid pKEB31 . However, based on the fact that the repeat backbone 433 differences between the native TALE constructs and their dTALE equivalents in our study 434 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint also impacted function, we posit that repeat backbone polymorphisms indeed can affect 435 the function of individual repeats and thus the overall targeting profile of a TALE effector. 436 Altogether, the presence of three major TALEs in IX-221, each with one or more repeat 437 types that can facultatively disengage, and one with more lax DNA targeting specificity 438 than its previously characterized ortholog, provide compelling evidence of intense 439 selection pressure on the Xoo population, particularly in India, to acquire or evolve TALEs 440 that equip a strain to overcome xa13-mediated resistance against bacterial blight disease. 441 Of note, while PthXo2B from each of four additional Philippine strains (Oliva et al, 2019) 442 has the same Q to R substitution in repeat 19 that PthXo2B PXO61 does; we found a 443 PthXo2B allele in the Taiwanese strain XM9 that matches PthXo2BIX-221, suggesting that 444 the ability to activate the putative alternative S gene is not restricted to Indian strains. 445 446 447 448 449

450 451 This work was supported by Department of Biotechnology award BT/CEIB/12/1/01 and the 452 Indian Council of Agricultural Research-National Project on Functional Genomics and 453 Genetic modification in crops to RR. RR thanks Rohini Sreevathsa, M Rathinam and 454 Narsimha Dokka, ICAR-NIPB, New Delhi, India for help and support. 455 456 457 DATA AVAILABILITY STATEMENT 458 459 Genome data for IX-221 generated in this study are available through Genbank Accession 460 number CP019228. 461 462 463 464 465 466 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint

467 468 Antony, G., Zhou, J., Huang, S., Li, T., Liu, B., White, F., et al. (2010) Rice xa13 recessive resistance to 469 bacterial blight is defeated by induction of the disease susceptibility gene Os-11N3. The Plant 470 cell, tpc. 110.078964. 471 Becker, S., Mücke, S., Grau, J. and Boch, J. (2022) Flexible TALEs for an expanded use in gene activation, 472 virulence and scaffold engineering. Nucleic Acids Res., 50, 2387-2400. 473 Blanvillain-Baufume, S., Reschke, M., Sole, M., Auguy, F., Doucoure, H., Szurek, B., et al. (2017) Targeted 474 promoter editing for rice resistance to Xanthomonas oryzae pv. oryzae reveals differential 475 activities for SWEET14-inducing TAL effectors. Plant Biotechnol J, 15, 306-317. 476 Boch, J., Scholze, H., Schornack, S., Landgraf, A., Hahn, S., Kay, S., et al. (2009) Breaking the code of DNA 477 binding specificity of TAL-type III effectors. Science, 326, 1509-1512. 478 Bogdanove, A. J., Schornack, S. and Lahaye, T. (2010) TAL effectors: finding plant genes for disease and 479 defense. Current opinion in plant biology, 13, 394-401. 480 Booher, N. J., Carpenter, S. C., Sebra, R. P., Wang, L., Salzberg, S. L., Leach, J. E., et al. (2015) Single 481 molecule real-time sequencing of Xanthomonas oryzae genomes reveals a dynamic structure 482 and complex TAL (transcription activator-like) effector gene relationships. Microbial genomics, 483 1. 484 Carpenter, S. C., Mishra, P., Ghoshal, C., Dash, P. K., Wang, L., Midha, S., et al. (2020) An xa5 resistance 485 gene-breaking Indian strain of the rice bacterial blight pathogen Xanthomonas oryzae pv. oryzae 486 is nearly identical to a Thai strain. Frontiers in microbiology, 11, 579504. 487 Carter, M. E., Carpenter, S. C., Dubrow, Z. E., Sabol, M. R., Rinaldi, F. C., Lastovetsky, O. A., et al. (2020) A 488 TAL effector-like protein of an endofungal bacterium increases the stress tolerance and alters 489 the transcriptome of the host. Proceedings of the National Academy of Sciences, 117, 17122-490 17129. 491 Cermak, T., Doyle, E. L., Christian, M., Wang, L., Zhang, Y., Schmidt, C., et al. (2011) Efficient design and 492 assembly of custom TALEN and other TAL effector-based constructs for DNA targeting. Nucleic 493 Acids Res, 39, e82. 494 Cernadas, R. A., Doyle, E. L., Nino-Liu, D. O., Wilkins, K. E., Bancroft, T., Wang, L., et al. (2014) Code-495 assisted discovery of TAL effector targets in bacterial leaf streak of rice reveals contrast with 496 bacterial blight and a novel susceptibility gene. PLoS pathogens, 10, e1003972. 497 Chu, Z., Fu, B., Yang, H., Xu, C., Li, Z., Sanchez, A., et al. (2006) Targeting xa13, a recessive gene for 498 bacterial blight resistance in rice. TAG. Theoretical and applied genetics. Theoretische und 499 angewandte Genetik, 112, 455-461. 500 Denancé, N., Szurek, B., Doyle, E. L., Lauber, E., Fontaine‐Bodin, L., Carrère, S., et al. (2018) Two 501 ancestral genes shaped the Xanthomonas campestris TAL effector gene repertoire. New Phytol., 502 219, 391-407. 503 Doyle, E. L., Booher, N. J., Standage, D. S., Voytas, D. F., Brendel, V. P., VanDyk, J. K., et al. (2012) TAL 504 Effector-Nucleotide Targeter (TALE-NT) 2.0: tools for TAL effector design and target prediction. 505 Nucleic acids research, 40, W117-W122. 506 English, A. C., Salerno, W. J. and Reid, J. G. (2014) PBHoney: identifying genomic variants via long-read 507 discordance and interrupted mapping. BMC bioinformatics, 15, 180. 508 Grau, J., Reschke, M., Erkes, A., Streubel, J., Morgan, R. D., Wilson, G. G., et al. (2016) AnnoTALE: 509 bioinformatics tools for identification, annotation, and nomenclature of TALEs from 510 Xanthomonas genomic sequences. Sci Rep, 6, 21077. 511 Hutin, M., Pérez-Quintero, A. L., Lopez, C. and Szurek, B. (2015) MorTAL Kombat: the story of defense 512 against TAL effectors through loss-of-susceptibility. Frontiers in plant science, 6, 535. 513 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint Ji, Z., Ji, C., Liu, B., Zou, L., Chen, G. and Yang, B. (2016) Interfering TAL effectors of Xanthomonas oryzae 514 neutralize R-gene-mediated plant disease resistance. Nat Commun, 7, 13435. 515 Lore, J. S., Vikal, Y., Hunjan, M. S., Goel, R. K., Bharaj, T. S. and Raina, G. L. (2011) Genotypic and 516 Pathotypic Diversity of Xanthomonas oryzae pv. oryzae, the Cause of Bacterial Blight of Rice in 517 Punjab State of India. Journal of Phytopathology, 159, 479-487. 518 Midha, S., Bansal, K., Kumar, S., Girija, A. M., Mishra, D., Brahma, K., et al. (2017) Population genomic 519 insights into variation and evolution of Xanthomonas oryzae pv. oryzae. Sci Rep, 7, 40694. 520 Miller, J. C., Tan, S., Qiao, G., Barlow, K. A., Wang, J., Xia, D. F., et al. (2011) A TALE nuclease architecture 521 for efficient genome editing. Nature biotechnology, 29, 143-148. 522 Mishra, D., Vishnupriya, M. R., Anil, M. G., Konda, K., Raj, Y. and Sonti, R. V. (2013) Pathotype and 523 genetic diversity amongst Indian isolates of Xanthomonas oryzae pv. oryzae. PLoS One, 8, 524 e81996. 525 Mondal, K. K., Meena, B. R., Junaid, A., Verma, G., Mani, C., Majumder, D., et al. (2014) Pathotyping and 526 genetic screening of type III effectors in Indian strains of Xanthomonas oryzae pv. oryzae causing 527 bacterial leaf blight of rice. Physiol. Mol. Plant Pathol., 86, 98-106. 528 Moscou, M. J. and Bogdanove, A. J. (2009) A simple cipher governs DNA recognition by TAL effectors. 529 Science, 326, 1501-1501. 530 Nino-Liu, D. O., Ronald, P. C. and Bogdanove, A. J. (2006) Xanthomonas oryzae pathovars: model 531 pathogens of a model crop. Molecular plant pathology, 7, 303-324. 532 Oliva, R., Ji, C., Atienza-Grande, G., Huguet-Tapia, J. C., Perez-Quintero, A., Li, T., et al. (2019a) Broad-533 spectrum resistance to bacterial blight in rice using genome editing. Nature biotechnology, 37, 534 1344-1350. 535 Oliva, R., Ji, C., Atienza-Grande, G., Huguet-Tapia, J. C., Perez-Quintero, A., Li, T., et al. (2019b) Broad-536 spectrum resistance to bacterial blight in rice using genome editing. Nat. Biotechnol., 37, 1344-537 1350. 538 Pérez-Quintero, A. L., Lamy, L., Gordon, J. L., Escalon, A., Cunnac, S., Szurek, B., et al. (2015) QueTAL: a 539 suite of tools to classify and compare TAL effectors functionally and phylogenetically. Front. 540 Plant Sci., 6, 545. 541 Read, A. C., Rinaldi, F. C., Hutin, M., He, Y. Q., Triplett, L. R. and Bogdanove, A. J. (2016) Suppression of 542 Xo1-Mediated Disease Resistance in Rice by a Truncated, Non-DNA-Binding TAL Effector of 543 Xanthomonas oryzae. Front Plant Sci, 7, 1516. 544 Richter, A., Streubel, J., Blucher, C., Szurek, B., Reschke, M., Grau, J., et al. (2014a) A TAL effector repeat 545 architecture for frameshift binding. Nat Commun, 5, 3447. 546 Richter, A., Streubel, J., Blücher, C., Szurek, B., Reschke, M., Grau, J., et al. (2014b) A TAL effector repeat 547 architecture for frameshift binding. Nature communications, 5, 1-10. 548 Römer, P., Recht, S. and Lahaye, T. (2009) A single plant resistance gene promoter engineered to 549 recognize multiple TAL effectors from disparate pathogens. Proceedings of the National 550 Academy of Sciences, 106, 20526-20531. 551 Streubel, J., Pesce, C., Hutin, M., Koebnik, R., Boch, J. and Szurek, B. (2013) Five phylogenetically close 552 rice SWEET genes confer TAL effector-mediated susceptibility to Xanthomonas oryzae pv. 553 oryzae. New Phytol, 200, 808-819. 554 Szurek, B., Rossier, O., Hause, G. and Bonas, U. (2002) Type III‐dependent translocation of the 555 Xanthomonas AvrBs3 protein into the plant cell. Molecular microbiology, 46, 13-23. 556 Tatusova, T., DiCuccio, M., Badretdin, A., Chetvernin, V., Nawrocki, E. P., Zaslavsky, L., et al. (2016) NCBI 557 prokaryotic genome annotation pipeline. Nucleic acids research, 44, 6614-6624. 558 Wilkins, K. E., Booher, N. J., Wang, L. and Bogdanove, A. J. (2015) TAL effectors and activation of 559 predicted host targets distinguish Asian from African strains of the rice pathogen Xanthomonas 560 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint oryzae pv. oryzicola while strict conservation suggests universal importance of five TAL 561 effectors. Front. Plant Sci., 6, 536. 562 Xu, Z., Xu, X., Gong, Q., Li, Z., Li, Y., Wang, S., et al. (2019) Engineering broad-spectrum bacterial blight 563 resistance by simultaneously disrupting variable TALE-binding elements of multiple susceptibility 564 genes in rice. Molecular plant, 12, 1434-1446. 565 Yang, B., Sugio, A. and White, F. F. (2006) Os8N3 is a host disease-susceptibility gene for bacterial blight 566 of rice. Proceedings of the National Academy of Sciences of the United States of America, 103, 567 10503-10508. 568 Yang, B. and White, F. F. (2004) Diverse members of the AvrBs3/PthA family of type III effectors are 569 major virulence determinants in bacterial blight disease of rice. Molecular plant-microbe 570 interactions, 17, 1192-1200. 571 Yang, J., Zhang, Y., Yuan, P., Zhou, Y., Cai, C., Ren, Q., et al. (2014) Complete decoding of TAL effectors 572 for DNA recognition. Cell Res., 24, 628-631. 573 Yu, Y., Streubel, J., Balzergue, S., Champion, A., Boch, J., Koebnik, R., et al. (2011) Colonization of rice leaf 574 blades by an African strain of Xanthomonas oryzae pv. oryzae depends on a new TAL effector 575 that induces the rice nodulin-3 Os11N3 gene. Molecular Plant-Microbe Interactions, 24, 1102-576 1113. 577 Yugander, A., Sundaram, R. M., Ladhalakshmi, D., Hajira, S. K., Prakasam, V., Prasad, M. S., et al. (2017) 578 Virulence profiling of Xanthomonas oryzae pv. oryzae isolates, causing bacterial blight of rice in 579 India. European Journal of Plant Pathology, 149, 171-191. 580 Zhou, J., Peng, Z., Long, J., Sosso, D., Liu, B., Eom, J. S., et al. (2015) Gene targeting by the TAL effector 581 PthXo2 reveals cryptic resistance gene for bacterial blight of rice. The Plant journal : for cell and 582 molecular biology, 82, 632-643. 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 .CC-BY-NC-ND 4.0 International licenseavailable under a (which was not certified by peer review) is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made The copyright holder for this preprintthis version posted August 16, 2024. ; https://doi.org/10.1101/2024.08.08.607113doi: bioRxiv preprint