Rice blast, which is caused by the filamentous ascomycetous (Magnaporthe oryzae, M. oryzae), is one of the most destructive diseases of rice on a global scale (Dean et al., Reference Dean, Kan, Pretorius, Hammond-Kosack, Pietro, Spanu, Rudd, Dickman, Kahmann, Ellis and Foster2012). Rice plants utilise a wide variety of disease resistance genes (R genes) to detect the presence of pathogens and initiate subsequent defence responses. Resistance mediated by the major R gene is effective at identifying fungal strains that carry the matching avirulence (AVR) gene (Woolhouse et al., Reference Woolhouse, Webster, Domingo, Charlesworth and Levin2002). With the exception of chromosome 3, at least 84 significant blast resistance genes (R gene) have been found and genetically mapped on 11 rice chromosomes. At least 26 of the identified blast R genes have been cloned and functionally investigated: Pb1, Pia, Pib, Pi-d2, Pid3, Pik, Pikh/Pi54, Pikm, Pikp, Pish, Pit, Pita, Pizt, Pi1, Pi2, Pi21, Pi25, Pi36, Pi37, Pi56, Pi63, PiCO39 (http://www.ricedata.cn), Pi64 (Ma et al., Reference Ma, Lei, Xu, Hao, Wang, Cheng, Ma, Ma, Zhou, Zhang, Guo, Wu, Lin, Wang, Zhai, Wang and Wan2015) and Pigm (Deng et al., Reference Deng, Zhu, Shen and He2006). The majority of cloned R genes are members of the nucleotide-binding site leucine-rich repeat (NBS-LRR) family, with the exception of Pi-d2 and pi21, which encode a receptor-like kinase and a proline-rich protein, respectively.
Long regarded as the most efficient and successful means of disease control, a deeper comprehension of the mechanisms behind the interaction between pathogens and host plants and the selection of stable, disease-resistant novel types are becoming increasingly important (Dodds and Rathjen, Reference Dodds and Rathjen2010; Peng et al., Reference Peng, Wersch and Zhan2018). However, AVR genes in M. oryzae are known to be highly variable and diverse (Selisana et al., Reference Selisana, Yanoria, Quime, Chaipanya, Lu, Opulencia, Wnag, Mitchell, Correll, Talbot, Leung and Zhou2017), which frequently threatens the efficacy of resistant cultivars with a single R gene a few years after their release. It is widely documented that mutations in the AVR gene AVR-Pita1 in historical and contemporary field isolates overcome a key R gene-mediated resistance (Dai et al., Reference Dai, Jia, Correll, Wang and Wang2010). Pyramiding more blast resistance genes in a rice cultivar (Arunakumari et al., Reference Arunakumari, Durgarani, Satturu, Sarikonda, Chittoor, Vutukuri, Laha, Nelli, Gattu, Jamal, Prasadbabu, Hajira and Sundaram2016; Jiang et al., Reference Jiang, Li, Liu, Shen, Gao, Zhang and He2019; Wu et al., Reference Wu, Xiao, Chen, Yu, Pan, Li, Zhang, Huang, Ji, Chen and Li2019) or mixture planting different varieties in a region (Zhu et al., Reference Zhu, Chen, Fan, Wang, Li, Chen, Fan, Yang, Hu, Leung, Mew, Teng, Wang and Mundt2000) may control rice blast effectively, but both methods have production limitations because pyramiding is a time-consuming and difficult task, and mixture planting is difficult to plant and harvest.
Some R genes with a relatively broad spectrum of blast resistance, such as Pi1 (Hua et al., Reference Hua, Wu, Chen, Wu, He, Lin, Wang, Ashikawa, Matsumoto, Wang and Pan2012), Pi37, Pit, Pikm, Pi5, and Pb1, have been extensively utilised in rice disease resistance breeding. Some blast R genes have a relatively narrow spectrum of resistance; however, their alleles and orthologs from various rice resources may exhibit varying blast resistance spectra. This is due to the coevolution of rice and the blast pathogen in various rice-growing locations, which has resulted in an abundance of allelic variety in rice resources. For example, Pi9/Pi2/Piz-t were cloned from the same chromosome location in different varieties and showed different resistance spectra (Zhou et al., Reference Zhou, Qu, Liu, Dolan, Sakai, Lu, Bellizzi and Wang2006). The same condition was also observed in the Pik locus, which contains five rice blast R genes (Pik, Pik-m, Pik-p, Pi1k-h and Pik-s) (Wang et al., Reference Wang, Xu, Lin and Pan2009). Therefore, studying natural variation in rice resistance genes from various rice resources can not only predict the stability of resistance to rice blast fungus, but can also be a crucial technique for rice blast resistance breeding. Four types of diversification of R genes can be defined on the basis of the polymorphism level (pi value) : conserved (type I; pi < 0.5%), intermediate-diversified (type II; 0.5% < pi < 5%), highly diversified (type III; pi > 5%), and present/absent genes (type IV; P/A).
Receptor-like kinases (RLK) are members of a vast gene family and participate in numerous cell-cell communication activities (Barre et al., Reference Barre, Herve, Lescure and Rouge2002; Bellande et al., Reference Bellande, Bono, Savelli, Jamet and Canut2017). In the preceding decades, a number of studies suggested a variety of possible functions for these genes. First, RLKs participate in plant growth, such as the CLV1 receptor in meristem signalling (Clark et al., Reference Clark, Williams and Meyerowitz1997) and the SRK receptor in pollen development signalling (Clark et al., Reference Clark, Williams and Meyerowitz1997). (Nasrallah, Reference Nasrallah2000). In addition, RLKs participate in biotic and abiotic stress signal transduction, such as XA21 (Song et al., Reference Song, Wang, Chen, Kim, Pi, Holsten, Gardner, Wang, Zhai, Zhu, Fauquet and Ronald1995) in rice blight resistance. The TM domains of RLKs play a significant role in the correct signal transduction from ligand reception by the extracellular region to the production of responses by the intracellular region. One amino acid variation in the TM region of Pi-d2 determines sensitivity or resistance to the AvrPi-containing rice blast fungus (Chen et al., Reference Chen, Shang, Chen, Lei, Zou, Zhai, Liu, Xu, Lin, Cao, Ma, Wang, Zhao, Li and Zhu2006).
Compared to Arabidopsis, RLKs exhibits expansionary patterns in rice, which are believed to be associated with recent lineage-specific expansions of resistance/defence-related genes (Shiu et al., Reference Shiu, Karlowski, Pan, Tzeng, Mayer and Li2004). The majority of plant RLKs involved in disease resistance, including XA21 (Song et al., Reference Song, Wang, Chen, Kim, Pi, Holsten, Gardner, Wang, Zhai, Zhu, Fauquet and Ronald1995), XA26 (Sun et al., Reference Sun, Cao, Yang, Xu, Li, Wang and Zhang2004), and FSL2, belong to this subclass (Gomez-Gomez et al., Reference Gomez-Gomez, Bauer and Boller2001). Pi-d2, a single copy gene identified in the rice genome encoding a potential serine-threonine RLKs membrane-spanning protein with 825 amino acids, gives resistance to M. oryzae strain ZB15 in the indica cultivar Digu (Chen et al., Reference Chen, Li, Xu, Zhai, Ling, Ma, Wang, Wang, Gao, Ma, Shang, Zhao, Zhou and Zhu2004, Reference Chen, Chen, Ma, Wang, Zhu and Li2010). At position 441 of the vulnerable pi-d2 allele, the amino acid methionine (M) was identified, while the resistant allele included isoleucine (I) (Chen et al., Reference Chen, Shang, Chen, Lei, Zou, Zhai, Liu, Xu, Lin, Cao, Ma, Wang, Zhao, Li and Zhu2006). Pi-d2 represents a new class of blast resistance genes since its structure differs from that of existing R genes, indicating that it may employ a novel pathogen recognition mechanism. The intracellular serine/threonine kinase domain of Pi-d2 has been discovered to interact directly with the E3 ligase OsPUB15 to regulate plant cell death and blast disease resistance (Wang et al., Reference Ma, Lei, Xu, Hao, Wang, Cheng, Ma, Ma, Zhou, Zhang, Guo, Wu, Lin, Wang, Zhai, Wang and Wan2015). Although some allelic variations have been discovered at the Pi-d2 locus in diverse rice varieties (Chen et al., Reference Chen, Shang, Chen, Lei, Zou, Zhai, Liu, Xu, Lin, Cao, Ma, Wang, Zhao, Li and Zhu2006; Li et al., Reference Li, Sun, Liu, Wang, Jia and Xu2015). However, the corresponding evolutionary analysis could not be performed due to the small number of the found Pi-d2 haplotypes. Moreover, the disease resistance spectra among the different haplotypes were not determined yet.
China's Yunnan province is one of the world's major hotspots of rice genetic variation (Zeng et al., Reference Zeng, Zhang, Li, Shen, Sun, Wang, Liao, Liu, Wang X, Xiao and Wen2007). Rice crop diversity has been influenced by the agricultural practises, customs, and traditions of many ethnic groups inhabiting regions of varying altitudes and climatic circumstances. The objective of this study was to investigate the DNA sequence of Pi-d2 haplotypes from Yunnan landrace varieties and to evaluate their blast resistance via gene transformation and blast inoculation. On the basis of the respective blast resistance spectra of the cloned orthologs, comparative analyses between the amino acid polymorphism sites of the discovered RLK proteins and their respective blast strain-specific resistances were done.
Materials and methods
Plant materials and rice blast strains
The Pi-d2 coding sequences for Digu (GenBank accession number FJ915121.1) and Nipponbare (NC008399) were retrieved from http://blast.ncbi.nlm.nih.gov. Six (6) wild rice lines' Pi-d2 haplotypes were retrieved from http://blast.ncbi.nlm.nih.gov. (online Supplementary Table S1), and four (4) wild rice lines' Pi-d2 haplotypes were amplified from plant seedlings cultivated in the lab's greenhouse. From the main germplasm bank of rice seed resources in Yunnan province, the Institute of Biotechnology and Genetic Resource of Yunnan Academy of Agricultural Sciences chose 299 rice (O. sativa) landraces, including 108 indica (Oryza sativa L. subsp. indica Kato) types and 191 japonica (Oryza sativa L. subsp. japonica Kato) variants (online Supplementary Table S2). The Agricultural Environment and Resources Research Institute of the Yunnan Academy of Agricultural Sciences supplied the 19 M. oryzae strains used in this investigation, which were dominant or high infectivity strains collected from Yunnan province (Table 1). Professor Cailin Lei of the Chinese Academy of Agricultural Sciences supplied ZB15 as the first isolate for the determination of Pi-d2.
Table 1. Rice blast resistance spectra of Pi-d2 haplotypes
5 scales were rated according to the visual number of lesions at the second youngest leaf, levels 0 and 1 were considered to be resistant (R); levels 2 and 3 represented medium resistance (MR) and medium susceptibility (MS), respectively; and levels 4 and 5 were considered to be susceptible (S).
Gene cloning
Fresh leaves of farmed and wild rice species were used to extract DNA. Forward primer Pid2F: 5′-ATAggatccATGGAAGCTCATGGCAATCG-3′ and reverse primer Pid2R: 5′-ATAcccgggTCATCTGGGACCAGAGAGCC-3′ were constructed and used to amplify the whole coding sequence of Pi-d2 based on the published data sequence (Li et al., Reference Li, Sun, Liu, Wang, Jia and Xu2015). A BamHI and a SmaI recognition sites with three protective bases (ATA) were added to their respective 5′ ends. Initial DNA denaturation was conducted at 95°C for 5 min, followed by 30 cycles of denaturation at 95 °C for 40 s, annealing at 60°C for 40 s, extension at 72°C for 3 min, and a final extension at 72°C for 5 min. The PCR products were respectively purified and sequenced.
Rice transformation
Because the Pi-d2 lacks an intron, the cloned gene fragments above were inserted into the binary vector pUN1301 via the BamHI and SmalI cloning sites. The completed construct was injected into Agrobacterium tumefaciens EHA105 after sequence verification.
Hiei et al. (Hiei et al., Reference Hiei, Ohta, Komari and Kumashiro1994) performed Agrobacterium-mediated transformation on calli obtained from mature embryos of the sensitive rice variety Nipponbare. Positive transgenic plants were identified by amplifying the marker gene hygromycin (HYG) in vector with the forward primer HYG-F: 5′-TGCGCCCAAGCTGCATCAT-3′ and reverse primer HYG-R: 5′-TGAACTCACCGCGACGTCTGT-3′. The following PCR amplification profile was used: initial DNA denaturation at 95°C for 5 min, followed by 30 cycles of denaturation at 95°C for 40 s, annealing at 60°C for 30 s, extension at 72°C for 20 s, and final extension at 72°C for 5 min.
Expression analysis of Pi-d2 haplotypes
TRIzol reagent (Invitrogen, Carlsbad, CA) was used to separate RNA from leaf sheath tissue, and cDNA was generated from poly(A) + RNA using a cDNA synthesis kit (Promega, Shanghai). The specific primer pair Pi-d2 was used in real time quantitative PCR (Pi-d2 D-F: 5′-CACAGGCTTCTTGCCTACGA-3′ and reverse primer Pi-d2 D-R: 5′-TATGCCAATCCCTTTGCCGT-3′) for 45cycles of amplification. Transcription of the Actin gene was utilised to normalise the cDNA levels using the primer pair 5′-CTGCGGGTATCCATGAGACT-3′ and 5′-GCAATGCCAGGGAACATAGT-3′.
Fungal inoculation
As detailed earlier by Jia et al. (Jia et al., Reference Jia, Valent and Lee2003), the reactions to rice blast were measured using a modified standard pathogenicity test. First, rice seedlings at the 3rd-4th leaf stage in a plastic sealed enclosure were spray-inoculated with a rice blast spore suspension at 1–5 × 105 spores/ml; second, the plastic sealed enclosure were sealed to maintain high humidity for 24 h; and finally, the plants were removed from the plastic sealed enclosure and kept in a greenhouse for an additional 6 days to allow for the development of notable disease symptoms. Based on the amount of visible lesions on the second youngest leaf, the reactions were classified as 0–1 for resistant, 2 for moderately resistant, 3 for moderately susceptible, and 4–5 for susceptible.
Computational analysis of DNA and protein sequences
Using DNAMAN (http://www.lynnon.com/) and Clustal X version 2.0, the sequences were aligned. Using version 7.0.1 of BioEdit, sequences were manually modified. The SMART tool (http://smart.embl-heidelberg.de) was utilised to search for protein motifs. The phylogenetic tree of Pi-d2 orthologs was constructed using MEGA5.0. DnaSP version 5.10 was used to conduct sliding-window analyses, polymorphism tests, and neutral-tests on different sections of Pi-d2 orthologs.
Results
Sequence characteristics of Pi-d2 haplotypes
The analysis of the Pi-d2 coding sequences of 309 rice varieties, comprising 299 landrace variants from Yunnan region and 10 wild rice varieties, revealed a total of 41 nucleotide changes. There were 28 informative parsimony sites and 13 variable singleton sites detected. In addition to the 25 synonymous mutation locations, 16 substitutional sites were observed.
Alignment of 299 Yunnan-assembled DNA sequences indicated 24 Pi-d2 haplotypes (H) (online Supplementary Table S3). H1 was compatible with the sequence of Pi-d2_digu, which was only observed in one indica variety, Jiugu (I1). H3, which includes 56 japonica types and 69 indica kinds, accounted for 41.67% of the total. H9, containing 88 japonica and 12 indica types, accounted for 33.33% of the total frequency (the second highest frequency, 31.7%). The third category, H2, contains 20 indica and 20 japonica types and accounts for 13.33% of the total. Each of the remaining haplotypes contains no more than ten variants. 14 of all haplotypes were only detected in Japonica varieties, while only 3 were found in indica varieties. However, the Yuanjiang wild rice variety had a unique H24 haplotype.
Nucleotide polymorphism of Pi-d2 haplotypes
Pi-d2 encodes a receptor-like kinase protein having an external domain anticipated to be a bulb-type mannose specific binding lectin (B-lectin) and an internal serine-threonine kinase domain. Figure 1 shows the nucleotide diversity distribution in the whole encoding sequences, as shown by a sliding window analysis of 309 Pi-d2 haplotypes, revealing the following characteristic: Firstly, 5 different variation peaks were found among all varieties, one in the unknown functional region, one in the PAN domain, one in the TM-spanning region, and two in the Serine-threonine kinase catalytic domain; secondly, the two peaks exists in the PAN domain and the TM-spanning region were significantly higher in Japonica, indicating that the diversity of Japonica is greater in these two regions than that of Indica. The PAN domain and the Serine-threonine kinase catalytic domain had greater nucleotide diversity than the other sections, which had lower nucleotide diversity.
Figure 1. Sliding-window analysis of diversities in Pi-d2 coding region. The nucleotide diversity (π; Y-axis) was generated by DNAsp5.0, and the X-axis represents the positions of nucleotides. The blue line stands for the result of cultivated rice varieties; the red line stands for the wild rice, and the black line stands for all rice. The map below the sliding window is the encoding structure of the Pi-d2 gene, the coloured box denotes the exon region.
Neutral tests were used to undertake additional examination of the different regions of Pi-d2 haplotypes (online Supplementary Table S5). In japonica and wild rice types, Tajima's D values were predominantly negative but not statistically significant, indicating that Pi-d2 orthologs may experience a slight tendency for positive selection but not balanced selection. However, the Tajima's D value was extremely negative for indica strains, particularly in the Serine-threonine kinase catalytic domain, indicating that the indica region may have been subjected to strong positive selection during evolution.
Amino acid sequence analysis of Pi-d2 haplotypes
Alignment of Pi-d2 protein amino acid sequences revealed a high degree of similarity; sequence consistency reached 99.69%. According to the alignment, the most significant amino acid substitution site is located in the TM domain, which is the resistance-determining site (441 I/M). In the remainder of the domain, only a few amino acid substitutions were detected, none of which were found in the region of sequence conservation (Fig. 2). Using the encoded amino acid sequences, a neighbour-joining tree was created to examine the evolutionary relationships of Pi-d2 haplotypes (Fig. 3). All of the materials were seen to be organised into four large groupings. On branch 1 were grouped five O.nivara varieties, one O.rufipogon variety, and ten haplotypes from farmed variations. This branch includes the majority of indica strains, including Pi-d2_Digu. Pi-d2_Nipponbare and eight additional haplotypes of 107 cultivated cultivars were assigned to branch 2. The percentage of japonica varieties within this category reached 88.79%. In branch 3, O.longistaminat, Hainan_O.rufipogon, and one japonica variety were grouped. Yuanjiang_O.rufipogon was placed on the same branch as Dongxiang_O.rufipogon and three haplotypes from cultivated rice cultivars. These outcomes showed that indica and japonica each had their own dominant haplotype, and that the majority of indica and japonica cultivars differed in the Pi-d2 coding region.
Figure 2. Conservatism analysis of amino acid sequences among Pi-d2 orthologs. The conservatism analysis of amino acid sequences was produced by the software WebLogo on line. The piled height of amino acid loci showed the degree of conservatism, and the piled height of different amino acids in single locus reflected the degree of correlation with this site. The underline indicated a conservative sequence.
Figure 3. Phylogenetic analysis of Pi-d2 orthologs.
Blast resistance of Pi-d2 haplotypes
Six Pi-d2 haplotypes, encoding for H1, H2, H11, H20, H23 and H24, containing a single amino acid I at position 441 were chosen for testing their resistance to a collection of rice blast strains. H1 was selected as a positive control since it codes for the same protein as Pi-d2_Digu. The coding protein of H2 was the most broadly distributed in Yunnan province, but the other four haplotypes were unique to Yunnan province.
To ensure uniformity, each Pi-d2 haplotype was placed into the binary vector pUN1301 under the control of the Zea mays ubiquitin 1 (Ubi) gene promoter and then converted into the susceptible rice variety Nipponbare. Pi-d2 haplotypes from Nipponbare Pi-d2_Nipponbare and empty vector UN1301 were also transformed into Nipponbare seedlings as a negative control. The independent primary transgenic lines (T0) were obtained and identified using the CAPS and Hyg transgene markers (Fig. 4). PCR results revealed that every putative transformant was transgene positive. The transgenic lines in which the individual transgenes were expressed at roughly the same levels (online Supplementary Fig. S1) and in which there were no visible side effects on phenotypes were chosen for co-segregation analysis in the T0 to T2 generations using the CAPS marker. In subsequent blast inoculation tests, the homozygous T2 lines of the six Pi-d2 haplotypes were used.
Figure 4. Detection of the DNA to the transformants. (a) HYG marker was used in the DNA detection. (b) (c) CAPS marker was used to confirm the positive transformants precisely and the restriction enzyme cutting site (Fau1) exit in Pi-d2_H1/ Pi-d2_H2/ Pi-d2_H12 and (Xmn1) exist in Pi-d2_H20/ Pi-d2_H23/ Pi-d2_H24.
Six Pi-d2 haplotype transgenic lines were separately injected with eleven M. oryzae strains. As a negative control, LTH exhibited susceptibility to all eleven strains, showing that the inoculation test was successful. Pi-d2_NP and UN1301 exhibited susceptibility to ZB15 and three other stains, including w1-117, 08-43-1a and 15-30-301. While exhibiting moderate resistance to w1-79, 16-2-1e, A23, w1-10, 08-55-1a and 08-35-1a, it exhibited resistance to w1-125 and 08-55-1a. All Pi-d2 haplotype transgenic lines exhibited resistance or moderate resistance to ZB15, as observed (Fig. 5). Nonetheless, various transgenic lines exhibited distinct spectra of resistance to other strains (Table 1). Notable is the fact that Pi-d2_H23 and Pi-d2_H24 transgenic lines were resistant to all strains. These outcomes suggest that the resistance of different Pi-d2 haplotypes transgenic lines has been significantly enhanced; however, there are variations in the resistance and spectrum of the various Pi-d2 haplotypes. To validate the disease resistance and resistance spectrum of Pi-d2_H23 and Pi-d2_H24, 18 additional M. oryzae strains were injected. Unlike the negative control, Pi-d2_H23 and Pi-d2_H24 exhibited nearly full resistance to all M. oryzae strains (online Supplementary Table S4). These results indicated that these two transgenic lines have a wider spectrum of resistance than the others.
Figure 5. Phenotyping of rice lines with Magnaporthe oryzae isolates. Disease reaction of transgenic plants inoculated with M. oryzae ZB15, W1-10, A23, 15-30-3a, 08-35-1a and W1-125. Columns: 1, UN1301 (empty vector control); 2, Pi-d2_NP; 3, Pi-d2_H1; 4, Pi-d2_H2; 5, Pi-d2_H12; 6, Pi-d2_H20, 7, Pi-d2_H23; 8, Pi-d2_H24.
By comparing the protein sequences of several Pi-d2 haplotypes, it was shown that the broad range of blast resistance was associated with an amino acid substitution at position 363 (online Supplementary Fig. S2), where Pi-d2_H23 and Pi-d2_H24 were V (valic acid) and others were A. (alanine). The results of the domain analysis revealed that the substitution site was present in the PAN domain of the extracellular structure of Pi-d2, indicating that the PAN domain coding region may have a role in detecting the unique blast strains.
Discussion
The present study showed that the pi value of the wild rice accessions belongs to type II (pi = 0.53%) while the Japonica accessions (pi = 0.191%) and the Indica accessions (pi = 0.085%) belong to type I, which indicated that the diversity of wild rice is larger than that in cultivated rice. This result shows that the encoding sequences of Pi-d2 gradually tend to unity during the rice domestication. Also Pi-d2_H24 from wild rice has a broader range of blast resistance than others. Plant disease resistance is connected with the expression pattern of R genes or genes involved in pathogenesis (PR genes) (Hayashi and Yoshida, Reference Hayashi and Yoshida2009; Amaranatha et al., Reference Amaranatha, Gambhire and Rajeswar2013; Li et al., Reference Li, Zhu, Chern, Yin, Yang, Ran, Cheng, He, Wang, Wang, Zhou, Zhu, Chen, Wang, Zhao, Ma, Qin, Chen, Wang, Liu, Wang, Wu, Li, Wang, Zhu, Li and Chen2017). Although our study did not consider the effect of the endogenous expression pattern of Pi-d2 on the disease resistance, the phenotypic determination results of transgenic plants with different Pi-d2 haplotypes in the same genetic background can also have a great reference value (Lv et al., Reference Lv, Xu, Shang, Jiang, Pang, Zhou, Wang, Liu, Li, Li, Xu, Chen, Zhao, Li and Zhu2013).
Our study show that there are two polymorphic loci in the TM domain, excluding the resistance determinant site (441 I/M), with the other site found at the 437th I/F, which is unique to one variety. To determine the biological significance of this polymorphism location, additional research is required. The majority of plant disease-resistant proteins are unstable. Similar to XA21, PI-D2's juxtamembrane domain contains a putative proteolytic motif (Ding et al., Reference Ding, Richter, Chen, Fujii, Seo, Xie, Zheng, Kanrar, Stevenson, Dardic, Li, Jiang, Zhang, Yu, Bartley, Chern, Bart, Chen, Zhu, Farmerie, Gribskov, Zhu, Fromm, Ronald and Song2009). It has been suggested that this motif contributed to the protein's instability (Xu et al., Reference Xu, Wang, Liu, Chen, Tinjuangjun, Pi and Song2006). All Pi-d2 haplotypes were found to be conserved in this motif range, indicating that the proteins coded by these Pi-d2 haplotypes remain unstable.
We show that, there are five polymorphic loci in the serine-threonine kinase catalytic domain. Specifically, at site 686H/R, nearly all rice varieties (R) differ from Pid2-digu (H). The significantly low value of Tajima’D in the Serine-threonine kinase catalytic domain of indica strains indicates that this region of indica may have been subjected to a strong positive selection during evolution. Although these polymorphism sites are not located in the kinase's ATP-binding region or serine-threonine specificity signature sequence region. However, these polymorphic sites may lead to conformational changes in the serine-threonine kinase domain, thereby altering the PI-D2 intracellular domain's ability to bind to substrates. And additional research is required to confirm this conclusion.
Therefore, it was hypothesised that the extracellular domain of PI-D2 could bind to hydrophobic macromolecules, such as plant hormones or proteins, with greater probability (Chen et al., Reference Chen, Shang, Chen, Lei, Zou, Zhai, Liu, Xu, Lin, Cao, Ma, Wang, Zhao, Li and Zhu2006). Among 309 rice varieties, our findings revealed two polymorphism loci in the b-lectin domain (including 299 landrace and 10 common wild rice lines). Nonetheless, these two loci are not within the core sequence of the b-lectin domain, indicating that this domain has been conserved throughout the evolution of rice. Interestingly, our investigation also suggested that resistance differences across Pi-d2 alleles are associated with the change of amino acids at position 363, with Pi-d2_H23 and Pi-d2_H24 containing valine and the others containing alanine (alanine).
Over all, in this study, we investigated the nucleotide polymorphism and evolution patterns of the Pi-d2 gene among 299 rice landraces of Yunnan province and six wild rice relatives, and 50 polymorphic sites and 24 haplotypes of Pi-d2 were found. Unlike the results of the previous studies, the functional site of Pi-d2 did not show strict differentiation between Indica subgroub and Japonica subgroub (Xie et al., Reference Xie, Liu, Lu, Zhang and Sun2022), which provides new clues on the origin and evolution of Pi-d2. Furthermore, we observed Tajima's D value was −1.87012 with statistical significance for Indica subgroub, which suggested that the rice blast resistant gene Pi-d2, especially Kinase domain of it indeed, may have undergone positive selection in Indica subgroub. Pi-d2 acts as a haploid gene and originated in early stage of Poales evolution (Vaid et al., Reference Vaid, Pandey and Tuteja2012; Xie et al., Reference Xie, Liu, Lu, Zhang and Sun2022). All of functional/non-functional haplotypes of Pi-d2 are present in the genome under strong directional selection during rice evolution, suggesting that Pi-d2 may palys other biological role besides pathogen recognition. Therefore, studying the mechanism of intracellular signalling mediated by kinase domain of Pi-d2 is likely to further reveal its biological function. In addition, we compared the blast resistance spectra of six Pi-d2 haplotypes carrying the functional site under Nipponbare's genetic background. Intriguingly, the data revealed that Pi-d2_H23 and Pi-d2_H24 have a broader range of blast resistance than others, including the Pi-d2 haplotype of Digu, which indicated that this two haplotype are beneficial for improving the breeding of rice blast resistance.
Supplementary material
The supplementary material for this article can be found at https://doi.org/10.1017/S1479262124000248
Data availability statement
The data that supports the findings of this study are available in the supplementary material of this article. These data are available from the corresponding author upon reasonable request.
Acknowledgements
We thank Professor Cailin Lei from Chinese Academy of Agricultural Sciences for providing M. oryzae strains ZB15. This research was supported by the China National Natural Sciences Foundation (#32260523/#31860369) and Agricultural Cooperative Research Programs of Yunnan of China (#202301BD070001-218/ #202101BD070001-002). The authors would like to express their gratitude to EditSprings (https://www.editsprings.cn) for the expert linguistic services provided.
Author contributions
Y.S. and M.X. conceived and designed the experiment and revised the paper. Y.Y. and J.M. conducted the main experiments, such as the RNA and DNA template preparation, gene cloning, vector construction, genetic transformation, material planting. J.L. performed the resistance to rice blast species identification of transgenic plants. J.T. provided assistance during paper revision. C.L. provided assistance during the planting of materials and genetic transformation. All authors have read and approved the final paper.
Ethical standards
This article does not contain any studies with animals performed by any of the authors. Informed consent was obtained from all individual participants included in the study. Neither the entire paper nor any part of its content has been published or has been accepted elsewhere. It is not being submitted to any other journal. All authors agree that there have no conflict of interest.
