Zhengyin Xu

Abstract Xanthomonas oryzae pv. oryzicola ( Xoc ) and X . oryzae pv. oryzae ( Xoo ) cause bacterial leaf streak (BLS) and bacterial leaf blight (BLB) in rice, respectively. Unlike Xoo , endogenous avirulence-resistance ( avr - R ) gene interactions have not been identified in the Xoc- rice pathosystem; however, both pathogens possess transcription activator-like effectors (TALEs) that are known to modulate R or S genes in rice. The transfer of individual tal genes from Xoc RS105 (hypervirulent) into Xoc YNB0-17 (hypovirulent) led to the identification of tal7 , which suppressed avrXa7-Xa7 mediated defense in rice containing an Xa7 R gene. Mobility shift and microscale thermophoresis assays showed that Tal7 bound two EBE sites in the promoters of two rice genes, Os09g29100 and Os12g42970 , which encode predicted Cyclin-D4-1 and GATA zinc finger family protein, respectively. Assays using designer TALEs and a TALE-free strain of Xoo revealed that Os09g29100 was the biologically relevant target of Tal7. Tal7 activates the expression of rice gene Os09g29100 that suppresses avrXa7-Xa7 mediated defense in Rice. TALEN editing of the Tal7-binding site in the Os09g29100 gene promoter further enhanced resistance to the pathogen Xoc RS105. The suppression of effector-trigger immunity (ETI) is a phenomenon that may contribute to the scarcity of BLS resistant cultivars.

Rice (Oryza sativa L.) is an important cereal crop consumed by almost half the world population and is vital for global food security. Bacterial leaf streak (BLS), which is caused by Xanthomonas oryzae pv. oryzicola (Xoc), is a devastating rice disease in Asia, Africa and Australia (Ji et al., 2014; Nino-Liu et al., 2006). Like other Xanthomonads, Xoc utilizes the type III secretion system (T3SS) to translocate effector proteins directly into host cells to suppress plant immunity (Nino-Liu et al., 2006; Yuan et al., 2021). Transcription activator-like effectors (TALEs) act as virulence or avirulence factors and function as eukaryotic transcription factors in plant cell nuclei, where they bind to effector-binding elements (EBEs) of targeted plant gene promoters via a TALE-encoded central repeat region (CRR). The CRR consists of highly conserved tandem repeats of 34-amino acids; the hypervariable 12th and 13th residues in each repeat determine nucleotide binding specificity and are referred to as repeat-variable diresidues (RVDs) (Boch et al., 2009; Moscou and Bogdanove, 2009). Identification of TALE-targeted genes in plants is used to facilitate plant disease resistance breeding programmes for X. oryzae pv. oryzae (Xoo), which is closely related to Xoc (Eom et al., 2019; Ji et al., 2016; Oliva et al., 2019; Xu et al., 2019). Relatively few TALE-targeted genes have been identified in the interaction of rice with Xoc. One example is OsSULTR3;6, which is up-regulated by Tal2g in Xoc strain BLS256. OsSULTR3;6 encodes a predicted sulphate transporter in rice and is considered a susceptibility (S) gene for BLS (Cernadas et al., 2014). In the Xoo-rice pathosystem, the disruption of TALE-binding elements in three S genes confers broad-spectrum resistance to Xoo (Oliva et al., 2019; Xu et al., 2019); therefore, we reasoned that a possible strategy for increasing rice resistance to Xoc might be the loss of susceptibility (RLS) genes. We speculated that rice would gain RLS if the EBE sequence of OsSULTR3;6, recognized by Tal2g, is edited by CRISPR/Cas9 technology (Zhou et al., 2014). In the present study, CRISPR/Cas9 was to disrupt the Tal2g-recognized EBE of OsSULTR3;6 in rice cultivar IRBB10, which is susceptible to Xoc. The TALE gene tal5d of Xoc strain RS105 was previously characterized (Ji et al., 2014; Wilkins et al., 2015). Tal5d contains 17.5 central repeat units that are nearly identical to Tal2g in Xoc strain BLS256; the obvious difference is that the 10th RVD of Tal5d is ND rather than HD in Tal2g (Figure 1a). We speculated that Tal5d might bind to the same EBE as Tal2g (EBETal2g) based on prediction of the two TALE-binding sites (Figure 1a). Furthermore, no other variants of Tal2g and Tal5d were identified in the available sequences of Xoc (Ji et al., 2014; Wilkins et al., 2015). To test this hypothesis, plasmid pET30a-tal5d was constructed and used for purifying His-Tal5d. Electromobility shift assays (EMSA) showed that the His-Tal5d bound to the Cy5-labelled pEBEtal2g fragment, and binding was reduced by adding unlabelled pEBEtal2g (Figure 1b). These results demonstrated that Tal5d binds the OsSULTR3;6 promoter at the EBEtal2g locus, which was then designated EBETal2g/Tal5d (Figure 1a). In order to disrupt Tal2g and Tal5d binding, we designed a sgRNA targeting the OsSULTR3;6 promoter near EBETal2g/Tal5d and constructed binary vector pCas9-gRNA4-SU (Zhou et al., 2014) to edit the EBETal2g/Tal5d sequence (Figure 1c,d). Five homozygous rice lines of IRBB10 (T1 generation) were obtained and named SU-1 to SU-5 (Figure 1d). PCR amplification and sequencing of the EBETal2g/Tal5d region showed the following changes relative to the wild-type EBE: SU-1 contained a 14-bp deletion and 12-bp insertion; SU-2 contained a 1-bp insertion (cytosine); SU-3 was missing a single nucleotide (thymine deletion); and SU-4 and SU-5 had 17- and 3-bp deletions, respectively (Figure 1d). The wild-type IRBB10 and five edited lines (SU-1 to SU-5) were inoculated by pin-pricking method (Pan et al., 2018) with Xoc strains BLS256 and RS105, respectively, to investigate potential resistance. At 14 days post-inoculation (dpi), the lesions induced by BLS256 and RS105 on the five edited rice lines (SU-1 to SU-5) were significantly smaller than those on IRBB10 (Figure 1e, 1f). It should be noted that SU-2 line showed less resistance significantly than IRBB10, but the lesion length formed was obviously longer than those in SU-1, SU-3, SU-4 and SU-5 lines (Figure 1e,f). Sequence analysis showed that the single nucleotide insertion in SU-2 occurred at the terminal nucleotide of the EBE (Figure 1d), suggesting that the modification may have less effect on TALE binding than the changes in the other four edited lines. The expression levels of Os01g52130 in SU-1 to SU-5, infiltrated with BLS256, were significantly lower than those in IRBB10 (Figure 1g), suggesting that the edited EBE loci disrupt the activation of Os01g52130 by Tal2g of BLS256 strain. For this, Xoc strains BLS256 and RS105, containing tal2g and tal5d, respectively, were then inoculated to IRBB10, SU-1 and SU-4 and bacterial growth was measured. Growth of Xoc BLS256 and RS105 was remarkably reduced in rice lines SU-1 and SU-4, respectively, as compared to the wild-type IRBB10 (Figure 1h). These results indicated that the homozygous mutations in the EBETal2g/Tal5d locus disarmed the recognition of Tal2g and Tal5d in rice nuclei, resulting in resistance to Xoc infection. In summary, we first generated mutations in the EBE of the OsSULTR3;6 promoter in IRBB10 and created new germplasm that exhibits resistance to Xoc strains containing virulence factors either Tal2g or Tal5d. Our findings show that genetic modification of the EBETal2g/Tal5d sequence via CRISPR/Cas9 technology may be used to develop rice lines with broad-spectrum resistance to bacterial leaf streak in rice by disrupting the EBEs of TALE-matched S genes in rice. We are grateful to Dr. Bing Yang (University of Missouri) for providing the CRISPR/Cas9 system. This work was supported by the National Natural Science Foundation of China (31830072), the National Key Research and Development Program of China (2016YFD0100601) and the National Transgenic Major Program (2016ZX08001-002). The authors declare no conflicts of interest. X.X. and G.C. designed the experiments. X.X., Z.X., Z.L. and M.Z. performed the experiments. X.X. and Z.X. wrote the manuscript. L.Z. and G.C. revised the manuscript.

Xa23 as an executor mediates broad-spectrum resistance to Xanthomonas oryzae pv. oryzae (Xoo), which contains a matching avirulence gene avrXa23, in rice for bacterial leaf blight (BLB). avrXa23 encodes a transcription activator-like effector (TALE) protein which binds to the EBE (effector-binding element) of the Xa23 promoter. It is unclear whether the considerable pressure of Xa23 leads to an emerging Xoo strain that overcomes Xa23 resistance. This study aimed to uncover new Xoo isolate(s) that overcome Xa23-mediated resistance and to investigate how the pathogen evades the resistance. Totally 185 Xoo isolates were used to screen possibly compatible strain(s) with Xa23-containing rice CBB23 by pathogenicity test. Genome Sequencing, Southern blot, tal gene cloning, Western blot, qRT-PCR and electrophoretic mobility shift assays (EMSA) were conducted to determine the mechanism of one Xoo isolate being compatible with Xa23-containing rice. One isolate AH28 from Anhui province is compatible with CBB23. AH28 strain contains an ortholog of avrXa23, tal7b and has 17 tal genes. The 4th RVD (repeat-variable diresidue) in Tal7b are missed and the 5th and 8th RVDs changed from NG and NS to NS and S*, respectively. These alternations made Tal7b unable to bind to the EBE of Xa23 promoter to activate the expression of Xa23 in rice. The ectopic expression of tal7b in a tal-free mutant PH of PXO99A did not alter the virulence of the strain PH, whereas avrXa23 made AH28 from compatibility to incompatibility with Xa23 rice. Best to our knowledge, this is the first insight of a naturally-emerging Xoo isolate that overcomes the broad-spectrum resistance of Xa23 by the variable AvrXa23-like TALE Tal7b. The RVD alteration in AvrXa23 may be a common strategy for the pathogen evolution to avoid being “trapped” by the executor R gene.

The Gram-negative bacterium Xanthomonas translucens infects a wide range of gramineous plants with a notable impact on small grain cereals. However, genomics-informed intra-species population structure and virulence repertories of the pathogen have rarely been investigated. In this study, the complete genome sequences of seven X. translucens strains representing an entire set of genetic diversity of two pathovars X. translucens pv. undulosa and X. translucens pv. translucens is provided and compared with those of seven publicly available complete genomes of the pathogen. Organization of the 25 type III secretion system genes in all the 14 X. translucens strains was exactly the same, while TAL effector genes localized singly or in clusters across four loci in X. translucens pv. translucens and five to six loci in X. translucens pv. undulosa . Beside two previously unreported endogenous plasmids in X. translucens pv. undulosa , and variations in repeat variable diresidue (RVD) of the 14 strains, tal1a of X. translucens pv. translucens strain XtKm8 encode the new RVDs HE and YI which have not previously been reported in xanthomonads. Further, a number of truncated tal genes were predicted among the 14 genomes lacking conserved Bam HI site at N-terminus and Sph I site at C-terminus. Our data have doubled the number of complete genomes of X. translucens clarifying the population structure and genomics of the pathogen to pave the way in the small grain cereals industry for disease resistance breeding in the 21st century’s agriculture.