A Molecular Identification and Resistance Evaluation of the Blast Resistance Genes in Japonica Rice in Northern China

Abstract

Rice blast is a fungal disease that seriously threatens rice production. It is of great significance to identify blast resistance genes and clarify their functions in rice varieties. In this study, 11 rice blast resistance genes in 80 Japonica rice varieties in northern China were investigated, including their resistance to rice blast. The results demonstrated that Pita, Ptr, Pib, Pik, and Piks were most widely found, accounting for 48.8, 48.8, 41.3, 20.0, and 18.8% of the tested varieties, respectively. Pi5-G2 at the Pi5 locus and Pik-G5 and Pik-G8 at the Pik locus were also commonly found, and these alleles accounted for 30.0, 10.0, and 3.8% of all the tested varieties, respectively. Pizt was identified only in two cultivars, and alleles Pi2, Pi9, and Pigm at the Piz locus on chromosome 6 were not detected. We found that Pi5 and Pita were relatively conserved, but the alleles of Pik were abundant. Besides Pik, Pikm, and Piks, we also found 10 new haplotypes, and Pikp and Pikh were not found in the japonica rice varieties in northern China. Among the tested varieties, 5 did not carry any of the tested genes, 30 carried only one blast resistance gene, 27 carried two, 14 carried three, and 4 carried four. The resistance of varieties carrying three or four resistance genes was better than those carrying none of the resistance genes or only one or two. There were no significant differences in the resistance characteristics among varieties from different provinces. Our study provided a reference for the molecular breeding of rice blast resistance.

1. Introduction

Rice blast, caused by Magnaporthe oryzae (M. oryzae), is a fungal disease that seriously threatens rice production. The most effective method for controlling the disease is via the breeding of rice varieties containing resistance genes. To date, more than 100 R genes or loci related to rice blast have been identified on all chromosomes [1]. Most of the cloned R genes encode proteins containing nucleotide-binding/leucine-rich-repeat (NLR). Many studies have shown that three resistance gene clusters located on chromosomes 6, 11, and 12 can mediate major resistance to rice blast [2]. In addition, the Pib and Pi5 genes have also been shown to have disease-resistant effects [3,4]. Pi2 [5], Pi9 [6], Pigm [7], and Piz-t [5] are alleles located in the resistance gene cluster on chromosome 6. The resistance gene cluster of chromosome 11 includes Pik [8,9], Pi1 [10], Pik-h/Pi54 [11], Pikm [12,13], Pik-p [14], Pik-s [8], and Pik-e [15]. Pita [16] and Ptr [17,18] are located on the resistance gene cluster of chromosome 12 [3,4].

Despite the numbers of rice blast genes having been cloned, the vital question remains as to how to use these R genes in breeding. Due to the complexity and diversity of M. oryzae, its pathogenicity in different regions varies significantly. Therefore, an accurate evaluation of the resistance genes and resistance phenotypes of different rice varieties is of great importance for the disease control of rice blast. In recent years, through the identification and application of blast-resistant genotypes in breeding resources, great achievements have been made in molecular-marker-assisted breeding. Rice varieties carrying different resistance genes have been applied to different degrees in production [19,20]. However, due to the complexity and variability of M. oryzae, the genes that play resistance effects are different in different areas. In addition, some alleles of rice blast resistance genes are abundant and difficult to distinguish. Therefore, common challenges faced by the majority of breeders include the selection of effective genes, the identification of resistant genotypes, and the evaluation of the effects of resistance genes in breeding.

In this study, 80 varieties from eight regions in northern China were used as test materials. The molecular markers of resistance genes were used in combination with PCR amplification and sequencing. Pib, Pi2/Pi9/Pizt/Pigm, Pi5, Pik/Pikm/Piks, Pita, Ptr, and their alleles in 80 northern Japonica rice varieties were systematically evaluated. The test materials were evaluated for their field resistance in Tonghua of the Jilin Province and Fengcheng of the Liaoning Province. At the same time, eight physiological races of blast isolated from the ZA group in the Liaoning Province were used to evaluate the in vitro resistance phenotypes of the tested materials. In order to provide a reference for blast resistance in the breeding of northern Japonica rice, the effects of different resistance genes and gene polymerization were analyzed based on an evaluation of the two different resistance phenotypes.

2. Materials and Methods

2.1. Plant Materials and Cultivation

In this study, 80 rice varieties were selected from 14 breeding units in eight provinces or municipalities including Liaoning, Jilin, Heilongjiang, Beijing, Henan, Shandong, Ningxia, and Xinjiang. Rice-blast-inducing nurseries were set up in Dandong, Liaoning Province (N 40°07′, E 124°23′) and Tonghua, Jilin Province (N 40°52′–43°3′, E 125°10′–126°44′) in 2015 and 2016, respectively (Table S1). Each plot was 2 m long, with 4 rows in total and plant spacings of 30 cm × 13.3 cm. Mongolian rice was used as the inducible variety. One inducible variety plot was set in every ten plots, and two lines of inducible varieties were planted outside the plots. The incidence of leaf blast was investigated in mid-July and neck blast in late September. The experimental materials for in vitro inoculation were planted in 10 × 7.0 × 8.5 cm³ seedling boxes and placed in the potted field of the Liaoning Academy of Agricultural Sciences.

2.2. Fungal Materials and Disease Evaluation

Eight M. oryzae strains were identified and provided by Professor Songhong Wei (Shenyang Agricultural University), belonging to the ZA group and numbered J1–J8. M. oryzae was inoculated on potato dextrose agar (PDA, Solarbio Cat# P8931) medium and cultured under light at 25 °C for 5–7 days. The active mycelia of M. oryzae were inoculated on tomato oatmeal agar (TOA, Oatmeal: SEAMILD NY/T 1510; Fresh tomato; Agar: Coolaber Cat# CA1331-500G) medium and further cultured under light at 25 °C for 2–3 days. Following the incubation period, the aerial mycelia on the TOA medium were gently erased with cotton swabs and cultured under light at 25 °C for 2–3 days. The spores of M. oryzae were collected and prepared as 10⁵/mL spore suspensions using Tween-20 (Coolaber Cat# CT11551-100 mL). Three 5 cm leaves were taken from each test variety of the three-and-a-half leaves stages and inoculated with three equal punch inoculations. A total of 5 μL of sporosporine suspension was placed at each point on the leaves and placed in 0.1% 6-benzylaminopurine (Coolaber Cat# PH104-100 mL) [21]. The incidence was observed 5–7 days following inoculation. Seven days after inoculation, disease symptoms were evaluated using a standard 0–9 scale [22], where 0–1 indicated highly resistant (HR), 2–3 = resistant (R), 4 = moderately resistant (MR), 5–6 = moderately susceptible (MS), 7 = susceptible (S), and 8–9 = highly susceptible (HS).

2.3. DNA Extraction and PCR Amplification

Rice leaf DNA was extracted using the CTAB method [23]. Primers of the corresponding genes were used for PCR amplification. The PCR reaction system was as follows (25 μL total volume per reaction): 10 μL of 2× Taq PCR Master Mix (Servicebio Cat# G3304), 0.5 μL of 10 μm primer, 1 μL of 50 μg/mL template DNA, and 8 μL of ddH₂O. A total of 4 μL of each of the PCR products was run on 1% agarose gel (Coolaber Cat# PH104-100 mL) and stained by Goldview (Servicebio Cat# G3304). The results were recorded using a gel imaging system. The PCR products were sent to Tsingke Biotechnology Co., Ltd. (Beijing, China) for sequencing. The DNAMAN software (DNAMAN 6.0.3.99) was used for a comparative analysis of the sequencing results.

2.4. Molecular Identification of Pi2/Pi9/Piz-t/Pigm

In order to better distinguish the genotypes of Pi2/Pi9/Piz-t/Pigm, we used different primers to distinguish them. PCR amplification was performed with Pi9 Pro F/R primers. Band typing was performed using 10% polyacrylamide gel to obtain 111 bp for the Pi2/Piz-t allele, 128 bp for the Pi9 allele, and 138 bp for the non-Pi2/Piz-t and non-Pi9 alleles (Table S2). The samples with 111bp bands were amplified with PCR using primer Pi2 F/R and detected using agarose gel electrophoresis. The bands were Pi2 with 322bp and Piz-t with no bands [24].

2.5. Molecular Identification of Ptr

The rice samples in this study were amplified with PCR using the primer Z12 F/R (Table S2), which can distinguish the susceptibility alleles of Ptr. Strip typing could be performed using 4% agarose gel electrophoresis or 10% polyacrylamide gel electrophoresis, and 226 bp was the susceptibility alleles of Ptr and 214 bp was the resistance alleles of Ptr [18].

2.6. Molecular Identification of Pita

Two pairs of primers, YL155/YL87 and YL183/YL87, can clearly distinguish the alleles of Pita. The primers YL155/YL87 were used for the PCR amplification and detection of Pita (Table S2). The 1042 bp product was the Pita resistance allele, while the absence of a band indicated the Pita susceptible allele. In order to further determine the accuracy of the detection, the primers YL183/YL87 were used to detect Pita via PCR amplification (Table S2). The product was a 1042 bp Pita-susceptible allele, and no band was a disease-resistant allele [25,26].

2.7. Molecular Identification of Pi5

The rice samples tested in this study were amplified with PCR using the Pi5 F /R primer to detect whether they carried the Pi5 resistance gene, and then tested using 1% agar-gel electrophoresis (Table S2). The product size of 1105 bp was sequenced and analyzed. The obtained sequence was consistent with that of the Pi5 gene, indicating that it contained Pi5 resistance alleles [4].

2.8. Molecular Identification of Pib

Pib F/R primers can clearly distinguish the rice blast resistance gene Pib. The rice samples in this study were amplified with PCR using the primer Pib F/R to detect whether they carried Pib, and then tested using 1% agarose gel electrophoresis (Table S2). The product size was 915 bp. The product sequence was the same as the Pib gene sequence, which meant that the Pib disease resistance allele was contained [18].

2.9. Molecular Identification of Pikh/Pik-ku/Pikm/Piks/Pikp/Pi1

The primers RGA4 were used for the PCR amplification of the samples. The amplified product was 1326 bp and sequenced (Table S2). The results were compared with the Pikh/Pik-ku/Pikm/Piks/Pikp gene sequence, and the resistance gene was determined according to the sequencing results [9]. The trees were constructed based on the whole predicted DNA sequences using neighborhood linkage algorithms. The bootstrap values corresponding to the number of branch order (1000 replicates) matches are displayed on the nodes at each branch point. The unit branch length is 0.0025.

3. Results

3.1. Distribution of Blast Resistance Genes in Japonica Rice in Northern China

3.1.1. Pi5

Pi5 consists of two NBS LRR genes (Pi5-1 and Pi5-2). In total, 75 varieties were successfully identified using functional markers of the Pi5-2 LRR region and 25 amplified the target fragment. Two haplotypes, Pi5-G1 and Pi5-G2, were found by sequencing with detection rates of 30 and 1.25%, respectively. Zhengdao20 carried Pi5-G1 with a stop codon on the sixth exon of Pi5-2. The other 24 varieties were Pi5-G2 genotypes with 4 SNPs, which were all missense mutations. The 24 varieties were collected in all the provinces, except for Shandong. The sequence alignment results of the Pi5 alleles are presented in Figure 1.

3.1.2. Pib

A pair of primers was designed to detect the differences in the Pib alleles according to the variations in the functional regions of the third exon of Pib. The PCR results showed that 37 varieties amplified the target bands. The sequencing results showed that there were only two genotypes, Pib and Pib-G1, with detection rates of 41.25 and 2.5%, respectively. Tiejing17 and Jingjing3 were the Pib-G1 genotypes and there were 60 SNPs within Pib (Figure 2). The PCR product sequences of the other 33 varieties that were distributed throughout all eight provinces were identical with Pib.

3.1.3. Pita and Ptr

The functional molecular markers of Pita were designed based on the SNP differences between the alleles at the Pita locus, and the functional molecular markers of Ptr were designed based on the Indel differences between the Ptr-resistant alleles [16,17]. In this study, these two functional markers were used to identify the presence of Pita and Ptr in the rice varieties. Thirty-nine of the varieties carried the Pita and Ptr disease resistance alleles. The two genes were closely linked in all the varieties, with a detection rate of 48.75%. These alleles were distributed in all the provinces except Ningxia (Figure 3).

3.1.4. Pi2/Pi9/Piz-t/Pigm

Pi2/Pi9/ Piz-t/Pigm were cloned at the Piz locus on chromosome 6. These alleles have been widely used in the southern rice regions of China. However, no northern Japonica rice varieties carrying the Pi2/Pi9/Pigm alleles were identified from the tested varieties. Only Longjing 31 and Liaokai 79 carried the Pizt locus, with a detection rate of 2.5%. The results showed that the alleles at this locus were less distributed in the main Japonica rice varieties in northern China (Figure 3).

3.1.5. Pikh/Pik-ku/Pikm/Piks/Pikp/Pi1

The Pik locus is highly polymorphic and six alleles (Pikh/Pik-ku/Pikm/Piks/Pikp/Pi1) were cloned at this locus. According to the sequence differences in the Pik locus resistance genes, one pair of primers was developed to distinguish different alleles. In this study, 28 out of the 80 varieties did not have PCR amplification products, suggesting that they did not carry known resistance alleles at the Pik locus. In addition to the known alleles for disease resistance, ten different haplotypes were detected in the Pik locus, including Pik-G1, Pik-G2, Pik-G3, Pik-G4, Pik-G5, Pik-G6, Pik-G7, Pik-G8, Pik-G9, and Pik-G10. Among these genotypes, Pik, Piks, and Pik-G5 were most widely distributed, with 16, 15, and 8 genotypes detected, respectively. Pikm detected two genotypes, while the other genotypes were detected with a low frequency.

These alleles were divided into three subgroups according to the sequence differences in the Pik functional loci (i.e., Pik subgroup, Piks/Pikm/Pi1 subgroup, and Pikh/Pikp subgroup) (Figure 4). The Pik subsets included Pik, Pik-G1, Pik-G2, Pik-G3, and Pik-G4, among which, Pik-G1 was closest to the Pik gene sequence, but Pik-G4 was relatively distant. The largest number of Pik alleles was 16 from the Liaoning, Jilin, and Heilongjiang Provinces. The Piks/Pikm/Pi1 subgroup included seven alleles (Piks, Pikm, Pi1, Pik-G7, Pik-G8, Pik-G9, and Pik-G10), among which, Pik-G7, Pik-G9, and Pik-G10 had the closest genetic distances with Piks. Pikm and Pi1 had the closest genetic distances. The Pik-G8 sequence was different from that of the other alleles. Fifteen rice varieties carrying Piks were from the Liaoning, Jilin, Heilongjiang, Henan, Beijing, and Ningxia Provinces. Only two rice varieties were carrying Pikm, one from Liaoning and one from Xinjiang. The Pikh/Pikp subgroup contained Pikh, Pikp, Pik-LTH, Pik-IR64, Pik-G5, and Pik-G6. Among these, the genetic distance of Pikh and Pikp was relatively close, while Pik-G5 and Pik-G6 were relatively close. Eight varieties carrying Pik-G5 originated from the Liaoning, Beijing, and Shandong Provinces. No rice varieties carrying Pikh and Pikp were detected in this study.

3.2. Aggregation of Resistance Genes and Their Contribution to Rice Blast Resistance

There were 39 rice varieties carrying Pita, 39 carrying Ptr, 33 carrying Pib, 16 carrying Pik, 15 carrying Piks, 2 carrying Pikm, and 2 carrying Pizt (Figure 5A). In addition, some new alleles were found in Pi5-G2, Pik-G5, and Pik-G8, consisting of 23, 8, and 3 varieties, respectively, accounting for 30.0%, 10.0%, and 3.8% of the total tested varieties. However, further verification is needed to determine whether these three alleles are new disease-resistant alleles. In total, 5 varieties did not carry any of the tested genes, 30 carried only one blast resistance gene, 27 carried two blast resistance genes, 14 carried three blast resistance genes, and 4 carried four blast resistance genes. Except for a few varieties, the incidence of disease was lower when inoculated in vitro than when identified in the field (Figure 5B).

The varieties carrying three or four disease-resistant genes were generally more resistant than those carrying none or only one or two disease-resistant genes (Figure 6A). There were no significant differences in the resistance among varieties in different regions (Figure 6B). Twenty-five rice varieties were artificially inoculated with eight strains and the susceptibility grades were all below the three in three field identification tests. The species used in this study were no disease resistance genes, Pita, Pik, Pi5-G2, Piks, Pik-G8, Pi5-G2 + Ptr, Pi5-G2 + Piks, Piks + Pita, Piks + Pizt, Pik + Pib, Pib + Pita, Pib + Pik-G5 + Pita, Pi5-G2 + Pib + Piks, Pita + Pizt + Pik, Pi5-G2 + Pik, Pi5-G2 + Pib + Pikm + Pita, and Pi5-G2 + Pib + Piks + Pita (Figure 6C). When the varieties contained only a single resistance gene, Pita, Pi5-G2, Pik, Piks, and Pik-G8 were overall more resistant than the Pib varieties (Figure 6C).

4. Discussion

Based on the need for screening resistant parents in disease resistance breeding, many breeders have carried out the identification of blast resistant genes in germplasm resources. However, comprehensive analyses have found that there were significant differences in the results. For example, some research found that, in the Heilongjiang Province, Pi9 had the highest distribution frequency, followed by Pita and Pikm, while Pib showed a low distribution frequency [27]. However, other research found that Pikh and Pi54 had the widest distribution [28]. The difference in these identification results was related to the experimental materials and identification methods used. Compared with previous studies, we summarized the identification methods for blast-resistant genes and found a set of effective methods for identifying 18 blast-resistant genes.

The degree of differentiation of the resistance genes was closely related to their positions on chromosomes. For example, Pik is located at the end of chromosome 11, which is prone to recombination and exchange in the process of hybridization. Therefore, there are abundant haplotypes of Pik in the cultivated species, with 13 types detected in this study alone. For such genes, it is not enough to use only a certain molecular marker in the genotyping of variety blast resistance. The accurate identification of the genotypes at this locus should be combined with PCR product sequencing, or, according to the reported results, multiple markers should be developed in the mutation-prone area to increase the accuracy of the genotyping. In this study, we identified one mutation hotspot region that can distinguish different alleles at the Pik locus. In addition to the known alleles, 10 unknown alleles were identified in the varieties of this study.

Northern China is the primary production area of Japonica rice, with an annual cultivated area of ten million hectares. In recent years, rice blast has become more common. The accurate identification of blast resistance genes is of great significance for the breeding of Japonica rice. In this study, functional molecular markers of Pib, Pi2/Pi9/Pizt/Pigm, Pi5, Pik/Pikm/Piks, Pita, and Ptr were used to detect the resistance genes carried by the northern Japonica rice from 14 breeding units in eight provinces or municipalities. It was found that Pib, Pita, Ptr, Piks, and Pi5-G2 were most widely distributed in these Japonica rice varieties. There were four SNPs between them, however, further research is needed to determine whether Pi5-G2 and Pi5 perform the same function. Pik was only detected in the varieties from the Heilongjiang and Liaoning Provinces, Pikm was only found in the varieties from the Liaoning and Xinjiang Provinces, Pikh was not detected on the allele at the Piz locus on chromosome 6, and Pizt was detected only in two varieties, while Pi2, Pi9, and Pigm were not detected. These results suggest that the genes at the Piz locus were less used in northern Japonica and would be beneficial for future breeding.

In the field investigation, many varieties often carried known resistance genes, but their field resistance is not stable at different times and regions, due to the fact that the interaction between rice blast fungus and rice conforms to the “gene to gene” hypothesis. Only when the resistance genes carried by rice varieties can recognize the effector of M. oryzae can the rice varieties exhibit disease resistance. However, M. oryzae is complex and unpredictable. Once the pathogenic gene of the pathogen mutates into a resistance gene that cannot be recognized, the resistance gene cannot play its resistance function, resulting in a loss of resistance. Therefore, an accurate assessment of the main resistance genes in a certain area can be a challenge. The genes that can affect resistance will change with the variation in the endemic rice M. oryzae races in this area. In this study, eight physiological races of M. oryzae were inoculated onto the experimental materials. The results showed that three varieties carrying Pib + Pik-G5 were resistant to the eight artificially inoculated strains, but showed moderate resistance in the field identification. Not were the results of the artificial inoculation and field induction significantly different, but the resistance effects of the same gene in different genetic backgrounds also varied significantly. For example, three of the ten varieties carrying only Pita and Ptr showed resistance to the disease during the artificial inoculation and field identification, while the other seven were susceptible to different degrees of the disease. In addition, for Pita + Pib + Piks carriers, Tiejing 1603 was significantly more sensitive to M. oryzae than Tongyu256. Kendao3 and Tiejing1601 both carried the Pi5-G2 + Pib + Pik + Pita gene, but Kendao34 had a significantly better resistance than the Tiejing1601 variety.

The results showed that the aggregation of multiple resistance genes in the same variety not only broadened the resistance spectrum, but also improved the resistance level to some physiological races. In this study, the susceptibility degree of the test materials carrying only Ptr was generally higher, while, when Ptr and other resistance genes co-existed, Pik-G5, Pik-G2, Piks, Pik, and Pita could all improve the resistance. Furthermore, the materials carrying three or four disease-resistant genes were generally more resistant than the materials carrying no disease-resistant genes or only one or two disease-resistant genes. However, there were exceptions. In this study, the resistance of Tiejing1603 and Liaonong979 to Pib + Pita + Piks was inferior to Liaoxing21 carrying Pib + Pita alone. In conclusion, the resistance of rice to blast was very complex, and the aggregation of the resistance genes was not the effect of 1 + 1 ≥ 2. Therefore, the number of genes should not be blindly pursued when carrying out resistance polymerization breeding. In other words, more genes may not be necessarily better. According to the combination of the genotypes and the resistance phenotypes of the parents, the materials carrying the major blast resistance genes and showing a relatively stable resistance to rice blast in many years should be selected. Taken together, the introduction of different types of blast resistance genes on different chromosomes should be strengthened to avoid the yield loss caused by the decline in disease resistance caused by the large-scale application of single blast resistance genes.

5. Conclusions

In conclusion, Pib, Pita, Ptr, and Piks were most widely distributed in northern Japonica rice varieties, while Pi2/Pi9/Pizt/Pigm alleles at the Piz locus were less distributed and require an urgent introduction. In terms of disease resistance, among the tested genes, the two varieties carrying Pizt had strong resistance, which deserves further research. Pik had abundant loci, among which, Pikm demonstrated a better resistance. Many varieties were carrying the Pita and Ptr loci. Although some of them were not resistant, the resistance contribution of Pita and Ptr cannot be ignored. Further identification of the resistance of antagonistic parents and selection for application in disease-resistant molecular breeding is needed.