Developing super rice varieties resistant to rice blast with enhanced yield and improved quality

Rice blast, caused by the Magnaporthe oryzae, is the most detrimental disease to rice. Yield losses caused by this disease were from 10% to 30% in rice planting areas (Skamnioti and Gurr, 2009); severe cases may even lead to complete cessation of production (Parker et al., 2008). Cloning rice blast resistant genes and applying them to cultivate resistant varieties is a practical and effective method for controlling rice blast disease. So far, more than 70 disease resistance genes and QTLs have been identified, of which at least 25 have been cloned and used in disease resistance breeding (Luo et al., 2017). The Pi2 gene encodes a protein with a nucleotide-binding site and leucine-rich repeat (LRR) domain (Zhou et al., 2006), and is one of the most broad-spectrum and efficient resistance genes to rice blast, exhibiting resistance to 36 out of 43 rice blast strains from 13 countries (Liu et al., 2002). However, many elite varieties do not carry Pi2 gene, which seriously hinders the widespread and sustainable promotion of these varieties. Marker-assisted selection (MAS) is one of the important technologies in modern breeding. It uses molecular markers to quickly detect individual plants carrying target genes, greatly improving breeding efficiency and saving breeding costs. Previously, Jiang et al. (2015) developed blast-resistant thermosensitive genic male sterile (TGMS) lines by employing MAS. High-density whole genome single nucleotide polymorphism (SNP) chips are also a technology that accelerates the breeding process. Therefore, combining MAS and SNP chips can make breeding improvement faster and more accurately. For example, Guo et al. (2024) used MAS and Green Super Rice 40K (GSR40K) detection to screen improved strains with high genetic similarity to the receptor in BC1F2. In this study, we successfully generated an improved rice germplasm with blast-resistant by integrating MAS and GSR40K detection technology. F1 plants obtained from the cross of Zhonghui261 (ZH261; a good quality restoring line of Huazheyou 261 [HZY261], which was selected as a demonstration and promotion variety of super rice in 2024) and Yuenongsimiao (YNSM; carrying Pi2), were backcross with ZH261 to generate BC1F1. Among 1025 BC1F1 plants, we selected 70 strains that similar to ZH261 and 26 individual plants were identified to carry Pi2 (Heterozygous) using the marker Pi2-CM1. Then, the single plant out of 26 with 89.72% of genetic background reversion was selected by detecting with GSR40K. The single plant was backcrossed again with ZH261 to generate 964 BC2F1 plants. We selected 10 strains that are similar to ZH261 and three individual plants were identified to carry Pi2 (Heterozygous) using the marker Pi2-CM1. The single plant out of three with 97.86% of genetic background reversion was selected to selfing and generate BC2F2. The single plant named ZH261-Pi2 was selected from BC2F2 by employing MAS and detecting with GSR40K (Figure 1a). ZH261-Pi2 exhibited 97.81% genetic similarity to ZH261 and 99.97% purity (Figure 1b). Based on 48 simple sequence repeats (SSR) markers followed by the protocol NY/T 1433-2014, only one pair of SSR marker (RM176) was different between ZH261 and ZH261-Pi2 (Figure 1c), indicating that ZH261 and ZH261-Pi2 were the same variety. Finally, we hybridised with Huazhe2A using ZH261 and ZH261-Pi2, respectively, to produce HZY261 and HZY261-Pi2, and evaluated their rice blast resistance, yield and quality traits (Figure 1a). We found that ZH261-Pi2 and HZY261-Pi2 significantly increased resistance to blast compared to ZH261 and HZY261 (Figure 1d–g). ZH261-Pi2 showed a similar grain numbers per panicle, but higher tiller numbers per plant, 1000 grain weight and grain yield per plant than that of ZH261 (Figure 1h–k). On the other hand, HZY261-Pi2 displayed lower 1000 grain weight, but higher grain numbers per panicle, tiller numbers per plant and grain yield per plant compared with HZY261 (Figure 1h–k). Together, our results indicate that introduction of Pi2 in ZH261 by integrating MAS and GSR40K detection can simultaneously enhance rice blast resistance and yield. Amylose content is the most critical parameter determining rice eating and cooking quality (Wang et al., 2024). ZH261-Pi2 showed a similar length–width ratio, alkali spreading value and amylose content, but higher gel consistency than that of ZH261 (Figure 1l–o). HZY261-Pi2 had higher length–width ratio, but lower amylose content and gel consistency than that of HZY261 (Figure 1l–n). As for alkali spreading value, no significant difference was observed between HZY261-Pi2 and HZY261 (Figure 1o). Balancing the yield, quality and resistance to disease is a daunting challenge in crop breeding due to the negative relationship among these traits (Xiao et al., 2021). However, it is feasible to improve the rice blast resistance, yield and quality of existing rice varieties through molecular design breeding (Mao et al., 2021). Our work provides a molecular design strategy to rapidly improve rice blast resistance, yield and quality by integrating MAS and GSR40K detection. In particular, the improved varieties can be directly applied to production, providing an important guarantee for food security. This research was financially supported by the Shanghai Agriculture Applied Technology Development Program, China (T20210103), the Project of Laboratory of Advanced Agricultural Sciences, Heilongjiang Province (ZY04JD05-005), the National Natural Science Foundation of China (32188102), the National Key Research and Development Program (2023YFD1201200), the Zhejiang Science and Technology Major Program on Agricultural New Variety Breeding (2021C02063-2), Zhejiang Provincial Natural Science Foundation of China (LDQ23C130001), Zhejiang Provincial Science and Technology Project (2020R51007) and the Key Research and Development Program of Zhejiang province (2022C02011). H.P., T.S. and S.Z. designed the research; G.N., A.R., R.Z., J.J., C.B., H.S., S.G., J.G., X.L., W.L. and Z.F. performed the experiments; S.Z. and G.N. analysed the results and wrote the manuscript. The authors declare no conflict of interests. Data sharing not applicable to this article as no datasets were generated or analysed during the current study.