Precision editing of a susceptibility gene promoter to alter its methylation modification for engineering rice resilience to biotic and abiotic stresses

Rice is a primary food crop, and its yield is threatened by biotic and abiotic stresses. Xanthomonas oryzae pv. oryzae (Xoo) causes bacterial blight, a chief bacterial disease of rice. Xoo infects rice depending on its transcriptional activation-like effectors (TALEs), which specifically target effector binding elements (EBEs) in the promoter of host susceptibility (S) genes and regulate S genes' expression for disease development. Editing S gene EBEs is an efficient approach for engineering disease-resistant rice (Oliva et al., 2019; Xu et al., 2019). Cold stress is a major abiotic factor that limits rice growth and productivity (Liu et al., 2019). Therefore, engineering rice resilience to biotic and abiotic stresses is a powerful strategy to enhance rice yield. Here, we precisely edited an S gene EBE to engineer an elite rice variety exhibiting broad-spectrum resistance to Xoo and to enhanced cold tolerance.

OsTFX1 is an S gene targeted by the major Xoo TALE (Römer et al., 2010; Sugio et al., 2007). Analysing TALEs of Xoo whose genomic sequences are available, it was found that all Xoo strains contain a TALE that targets OsTFX1 EBE (Figure S1) and activates OsTFX1 expression (Yuan et al., 2016), suggesting that OsTFX1 is a Xoo-dependent major S gene. Comparing OsTFX1 sequence in 3339 rice accessions from the RiceVarMap database, its EBE was found to have a unique sequence (Figure S2), indicating that there were no natural resistant alleles of OsTFX1 for breeding. Therefore, we designed sgRNA specifically targeting OsTFX1 EBE and generated the OsTFX1^ebe mutants via CRISPR/Cas9-mediated mutagenesis. By screening 34 hygromycin-resistant independent lines by Sanger sequencing, we identified five types of OsTFX1^ebe mutants (Figure 1A). These OsTFX1^ebe mutants harbouring none off-target events were backcrossed with wild type (WT) and transgene-free plants were generated for analysis (Figure S3). The OsTFX1^ebe mutants exhibited enhanced resistance to a set of Xoo than WT (Figure 1B,C; Figure S4). OsTFX1 did not respond to Xoo infection in the OsTFX1^ebe mutants (Figure 1D), suggesting that EBE-edited OsTFX1 had attenuated induction to Xoo, causing the OsTFX1^ebe mutants to exhibit broad-spectrum resistance.

Interestingly, OsTFX1 has significantly higher expression in OsTFX1^ebe mutants than in WT (Figure 1E). Diversification of transcription factor-targeted regulatory elements and modification of DNA methylation at the promoter can alter gene expression (Zhu et al., 2016). On analysing the OsTFX1 promoter, neither putative transcription factors that might target the EBE nor new potential transcription factor-targeted regulatory elements were detected in the EBE (Figure S5). However, analysis of RiceENCODE methylation data (Figure S6) showed that the OsTFX1 EBE contained three CHH-type methylation sites. Our bisulfite sequencing assay validated that the three methylation sites of OsTFX1 EBE were methylated in WT, but not in the OsTFX1^ebe mutants (Figure 1F). DNA methylation in the promoter has a repressive function on promoter activity (Schmitz et al., 2019). Thus, EBE-edited OsTFX1 has a lack of methylation sites, resulting in increased OsTFX1 expression in the OsTFX1^ebe mutants.

vOsTFX1 (alternative name OsbZIP73) is positively involved in cold tolerance. Co-overexpression of OsTFX1 and OsbZIP71 could improve rice tolerance to cold stress at both the seedling and reproductive stages (Liu et al., 2019). To assess whether the OsTFX1^ebe allele had a comparable effect as the overexpression of OsTFX1 when combined with the overexpression of OsbZIP71, we generated the OsbZIP71-overexpressing transgenic lines (OsbZIP71-OE) (Figure S7) and then crossed two representative OsbZIP71-OE lines showing increased expression with two OsTFX1^ebe mutants to generate OsTFX1^ebe/OsbZIP71-OE lines. Both OsTFX1 and OsbZIP71 had significantly increased expressions in the T2 homozygous seedlings of OsTFX1^ebe/OsbZIP71-OE lines (Figure S8). To evaluate the response of OsTFX1^ebe/OsbZIP71-OE lines to cold stress at the seedling stage, three-week-old seedlings were subjected to chilling treatment along with OsTFX1^ebe mutant, OsbZIP71-OE, and WT. The OsTFX1^ebe/OsbZIP71-OE lines exhibited dramatically enhanced cold tolerance, supported by twofold increased survival rate and 1.8-fold decreased ion leakage rate, compared to the OsTFX1^ebe mutant, OsbZIP71-OE or WT (Figure 1G,H). Cold stress in rice largely results in pollen sterility, leading to decreased yield at the reproductive stage (Liu et al., 2019). During cultivation upto the booting stage, half of the OsTFX1^ebe/OsbZIP71-OE, OsTFX1^ebe, OsbZIP71-OE, and WT were subjected to a chilling treatment at 16°C for 9 days and then returned to normal temperature, whereas the other half were consistently planted under normal temperature. All the plants had comparable phenotypes including over 92% pollen viability (Figure S9), 86% seed setting rate (Figure 1J,K), and about 22 g yield per plant under normal temperature (Figure S10), suggesting that EBE-edited OsTFX1 does not has a negative effect on the yield. Although cold stress can compromise pollen viability, the OsTFX1^ebe/OsbZIP71-OE lines still had 1.5-, 1.8-, and 1.7-fold higher pollen viability, seed setting rate, and yield per plant, respectively, than OsTFX1^ebe mutant, OsbZIP71-OE, or WT under cold stress (Figure 1J,K; Figure S8). Collectively, combining EBE-edited OsTFX1 and overexpressed OsbZIP71 could improve rice tolerance to cold stress at both the vegetative and reproductive stages. In addition, the OsTFX1^ebe/OsbZIP71-OE lines also exhibited broad-spectrum resistance to Xoo (Figure 1L,M).

In summary, we here present a feasible epigenetic editing approach of a susceptibility gene to modify its methylation modification for engineering bacterial blight-resistant and cold-tolerant rice, without defence-growth trade-off.

The authors declare no conflict of interest.

M.Y. conceived the project. J.T., H.Z., Y.L., L.X. and M.Y. performed the research and analysed the data. J.T. and M.Y. wrote the manuscript.