Genome editing of a dominant resistance gene for broad‐spectrum resistance to bacterial diseases in rice without growth penalty

Bacterial blight (BB) and bacterial leaf streak (BLS), caused by Xanthomonas oryzae pv. oryzae (Xoo) and Xanthomonas oryzae pv. oryzicola (Xoc), respectively, are the two most devastating bacterial diseases of rice worldwide. Both bacterial pathogens infect rice plants relying on type III secreted transcriptional activation-like effectors (TALEs) that bind to specific effector binding elements (EBEs) in the promoter of susceptibility (S) genes, and activate its expressions for disease development (Chen et al., 2010). To counteract BB, rice has evolved a unique type of executor resistance (R) genes (Xa7, Xa10, Xa23, etc.), which can specifically trap certain Xoo TALEs via EBEs within their promoters and trigger strong hypersensitive response (HR). Previously, we verified that gene correction of xa23 with EBEAvrXa23 restored its function in triggering defence responses against Xoo invasion (Wei et al., 2021). Considering Xa23-mediated resistance to BB has been overcome by new Xoo isolates in the field in recent years, and no natural R genes against Xoc have been identified in rice; here, we investigated whether an EBE-stacking-in-the-promoter strategy could be employed in the molecular rice breeding for durable and broad-spectrum resistance to both BB and BLS by genome-editing technology. Sequencing analysis of the xa23 locus in the commercial rice cultivar Nangeng 46 (N46) revealed that it shared an identical coding region with Xa23 in CBB23, whereas it lacked the complete EBEAvrXa23 sequence in the promoter (Figure 1a). We presumed that introducing multiple EBEs, which were responsive to TALEs from various Xoo and Xoc strains, into the xa23N46 locus by genome editing would render host broad-spectrum and durable resistance to both pathogens. Thus, 10 EBEs (Table S1) responding to PthXo1, PthXo3, AvrXa23, Tal9aBLS256, etc. were selected and constructed into a 220-bp EBEXoc/Xoo array (Figure 1a; Figure S1). It was synthesized, PCR amplified using a pair of chemically modified primers and used as the DNA donor (Figure S1) for targeted DNA insertion in the xa23N46 promoter through CRISPR/Cas9-induced non-homologous end joining (NHEJ) repair pathway as described previously (Lu et al., 2020). After a screen of 79 hygromycin-resistant independent lines by PCR amplification (Figure S2) and Sanger sequencing, we identified three homozygous lines (#3, #12, and #60) which carried intact EBEXoc/Xoo array in the forward orientation (Figure 1b, Table S2). Transgene-free edited plants were identified in T1 populations (Figure S3; Table S4). Among them, the identity of homozygous plant #3–7 [named N46(Xa23R)] was verified by whole-genome sequencing. T2 and T3 progenies of N46(Xa23R) were planted in the paddy field, and no significant differences in major economic traits were observed compared with N46 (Figure 1c,d; Figure S4; Table S3), suggesting that the promoter-editing of xa23 does not cause growth penalty in N46. Semi-quantitative RT-PCR analysis and HR assay revealed that the edited xa23 gene with the EBEXoc/Xoo array insertion is responsive to the different TALEs derived from the corresponding bacterial pathogens and regains the HR-eliciting activity in N46(Xa23R) (Figure 1e,f). Thus, 20 standard Xoo strains (seven from China, 10 from Philippines and three from Japan), which were known for their virulence or weakly virulence in N46, were selected and inoculated to N46(Xa23R) plants. Two weeks after inoculation, N46(Xa23R) exhibited robust resistance (reduced lesion lengths) against all strains compared with N46 (Figure 1g–j). Also, we challenged N46(Xa23R) plants with another 30 virulent Xoo isolates, which were collected from the paddy fields in China during the BB epidemic in recent years and known for virulence in N46. As a result, N46(Xa23R) plants were highly resistant to these field isolates (Figure S5). To examine the BLS resistance of N46(Xa23R) plants, two representative Xoc strains (RS105 and BLS256) were first tested. We observed typical HR in N46(Xa23R), whereas leaf streak lesions in N46 for both strains (Figure 1k,m). Next, another 28 virulent field isolates of Xoc were included. We observed that bacterial leaf streak lesions in N46(Xa23R) plants were significantly reduced compared with the control (Figure 1l,n). Combined, these data suggest that the EBEXoc/Xoo array renders N46(Xa23R) plants broad-spectrum resistance to a wide range of virulent Xoo and Xoc strains tested here (Figures S6 and S7). In other words, engineering the endogenous xa23 gene by EBE stacking in its promoter broadens Xa23's function in defence responses against different bacterial pathogens. In the co-evolutionary arms race between hosts and pathogens, resistance conferred by a single R gene is often short-lived and is rapidly overcome by new pathogen strains bearing mutation events in relevant effector genes (Xu et al., 2022). In this regard, we assume that N46(Xa23R) line and its derived resistant rice lines by cross-breeding would be virtually impregnable to pathogen evolution and exhibit rice durable resistance given that the engineered Xa23 is responsive to multiple core virulence TALEs of both Xoo and Xoc. Overall, here we present a feasible promoter-editing strategy of executor gene for generating novel rice germplasms and upgrading current commercial japonica rice cultivar with robust, durable and broad-spectrum resistance to BB, especially to BLS, without defence-growth trade-off. The project was supported by the Nanfan special project of the Chinese Academy of Agricultural Sciences (YBXM2313) and the Hainan Yazhou Bay Seed Laboratory (B21HJ0215) to H.Z. The authors have filed a patent application based on the results reported in this study. H.Z., C.S., C.Z. and X.Z. designed the research. M.W., Z.J., H.L., S.L. and W.X. conducted the experiments and analysed the data. H.Z., C.S., S.Z., C.Z. and X.Z. supervised the research. Z.J., M.W. and H.Z. wrote the original draft; all authors participated in the discussion and revision of the manuscript. Figure S1 Sequence of chemically modified donor of the designed EBEXoc/Xoo array. Figure S2 Representative T0 transgenic rice lines containing inserts of varying sizes identified by PCR amplification. Figure S3 Isolation of T1 plants without T-DNA insert. Figure S4 Growth performance of N46(Xa23R) and N46 plants in paddy field. Figure S5 N46 and N46(Xa23R) challenged by Xoo isolates from China. Figure S6 Statistical analysis of the lesion length of N46 and N46(Xa23R) challenged by 20 standard Xoo strains from China, Philippines, Japan and 30 Xoo field isolates from China. Figure S7 Statistical analysis of the lesion length of N46 and N46(Xa23R) challenged by 2 standard Xoc strains, and 28 Xoc field isolates from China. Table S1 Information of 10 EBEs used in this study. Table S2 The editing patterns of T0 plants. Table S3 Analysis of the agronomic traits of N46(Xa23R). Table S4 Primers used in this research. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.