Virulence Pattern of Pyricularia oryzae Pathotypes Towards Blast Monogenic Lines

Rice blast caused by Pyricularia oryzae (P. oryzae) is one of the most serious diseases infecting rice worldwide. In the present study, virulence pattern of six P. oryzae pathotypes (P0.0, P0.2, P1.0, P3.0, P7.0 and P9.0) identified from the blast pathogen collected in Peninsular Malaysia, were evaluated using a set of 22 IRRI-bred blast resistance lines (IRBL) as well as to determine the resistance genes involved. The information on the virulence of the blast pathotypes and the resistance genes involved is important for breeding of new rice variety for durable resistance against blast disease. The IRBL was established from 22 monogenic lines, harbouring 22 resistance genes [Pia, Pib, Pii, Pit, Pi3, Pi5(t), Pish, Pi1, Pik, Pik-s, Pik-m, Pik-h, Pik-p, Pi7(t), Pi9, Piz, Piz-5, Piz-t, Pi19, Pi20(t), Pita-2, and Pita=Pi4(t)]. Based on the disease severity patterns, the tested pathotypes were avirulence towards seven IRBLs [IRBLi-F5, IRBLk-Ka, IRBLkh-K3, IRBLz-Fu, IRBLsh-S, IRBLPi7 (t) and IRBL9-W] of which these IRBLs harbouring Pii, Pik, Pik-h, Piz, Pish, Pi7(t) and Pi9 resistance genes, respectively. Therefore, the results suggested that the seven IRBLs carrying seven resistance genes [Pii, Pik, Pik-h, Piz, Pish, Pi7(t) and Pi9] would be suitable candidates of resistance genes to be incorporated in new breeding lines to combat the current blast pathotypes in the field.


INTRODUCTION
Blast disease caused by Pyricularia oryzae (P. oryzae) (synonym Magnaporthe oryzae) is one of most serious disease of rice worldwide including Malaysia. The most effective and practical method to control blast disease is the use of resistant rice varieties. However, the resistance often breakdown or lost in a few years after the rice variety was released. This is caused by the appearance of new virulence pathotypes or races of the blast pathogen that overcome the resistance (Zhou et al. 2007) as well as due to high variability of rice blast pathogen (Wang et al. 2013;Poonsin & Parinthawong 2020). In order to develop more effective resistance against blast pathogen, it is important to have the knowledge of the host resistance and the pathogen as well as to determine the resistance genes.
The interaction between rice plant and blast pathogen is based on genefor-gene theory by Flor (1971) of which for every resistance gene (R gene) in the host, there is a corresponding avirulence gene (AVR) in the pathogen. Thus, in blast disease pathosystem, a major resistance gene confers specific resistance to a blast pathogen pathotype that contains a specific corresponding avirulence gene. In other words, every resistance gene in the host corresponds to an avirulence gene in the pathogen that acts as an effector to triggers the defence response (Wang et al. 2008;Huang et al. 2014).
To study the resistance and avirulence genes interaction, differential system comprising rice differential varieties and blast pathogen pathotypes are often used. This system provide systematic method to characterise and to postulate the resistance genes as well as to determine the relationships between pathotypes and the resistance genes (Koide et al. 2011). The information can be used to strategise effective and lasting method to manage blast disease of rice. Differential varieties distinguish pathotypes by their differential reactions to the blast pathogen. Current pathotype identification by using local differential varieties is useful to study the pathological diversity of P. oryzae population for pathotype identification and to characterise the blast isolates, but it is not sufficient to distinguish and characterise in detail blast pathotypes as the system lack the information on the genes involved in these local differential varieties. Thus, a set of international differential varieties harbouring resistance genes is frequently used to determine blast pathotypes virulence patterns.
A set of international blast differential varieties were developed by collaborative effort between International Rice Research Institute (IRRI) and Japan International Research Center for Agricultural Sciences (JIRCAS). The international blast differential varieties consisted of 23 monogenic lines of IRRIbred blast-resistant line (IRBL) that representing 23 resistance genes, namely Pish, Pia,Pib,Pit,Pii,Pi1,Pi3,Pi5(t),Pik,Pi7(t),Pi9,Piz,Pita,Pi12(t), Pi19(t) and Pi20(t) with the genetic background of a Chinese Lijiang Xintuan Heigu (LTH) rice variety from blast-susceptible Japonica variety (Tsunematsu et al. 2000;Kobayashi et al. 2007). This differential system is used for identification of pathotype virulence, virulence pattern between the blast pathotypes and predicting the resistance gene(s) in rice varieties (Telebanco-Yanoria et al. 2010).

Blast Monogenic Lines
In this study, a set of international differential varieties consisting of 22 IRBLs carrying 22 resistance genes [Pia,Pib,Pii,Pit,Pi3,Pi5(t),Pish,Pi1,Pik,Pi7(t),Pi9,Piz,Pi19,Pi20(t), Pita-2 and Pita=Pi4(t) and a susceptible control, LTH (Table 1) were used to evaluate the virulence of six pathotypes identified in Peninsular Malaysia. The resistance genes present in each line was predicted based on their reaction patterns to the IRBL. Only 22 IRBL lines were used in the present study as the seeds of IRBL12-M did not germinate and thus was not included in the study.
Sowing of the seeds and inoculum preparation were synchronised. Seeds were pre-germinated and sown in a plastic tray (26 cm × 37 cm) with sieved top soil mixed with cow dung. The seedlings were fertilised with urea (150N kg/ha) 2-3 days before inoculation.

Designation of IRBLs
Target resistant gene
The inoculum preparation was done according to Hayashi et al. (2009). Mycelia grown on OMA were scraped with spatula and the plates were left open in a tray covered with wrapping plastic. The plates were then placed under exposure of fluorescent light at 25°C ± 2°C for 4-7 days in order to induce sporulation. To prepare a conidial suspension, the plates were flooded with distilled water and the conidia were gently scraped using a brush. The conidial suspensions were filtered through nylon mesh and the concentrations of the suspensions were adjusted to 1×10 5 spores/ml by using a haemocytometer.

Inoculation of P. oryzae pathotypes
Rice seedlings at 3-4 leaf stage were inoculated with different pathotypes of blast isolates by spraying the conidial suspension simultaneously using a motorised sprayer. The inoculated seedlings were placed in a dark chamber with a moisturesaturated atmosphere at 25°C-30°C for 24 hr. The seedlings were then transferred to a mist room with high humidity for 6 days (Hayashi et al. 2009). The seedlings were arranged in complete randomised design with three replicates in a plant house at MARDI Seberang Perai, Pulau Pinang, Malaysia. The inoculation and evaluation were repeated twice.

Disease assessment
Disease severity was assessed 7 days after inoculation using disease scale as described by Goto and Yamanaka (1968) and Mackill and Bonman (1992) ( Table 2). Disease assessment scored as 0 to 2 is categorised as resistant (R) and scored 3 to 5 is categorised as susceptible (S) as described by Hayashi et al. (2009)

Statistical analysis
SAS statistical software package version 9.4 (SAS Institute, Cary, NC) was used for statistical analysis. Disease score for virulence analysis was found to be nonnormally distributed, therefore the data was analysed using nonparametric method: Kruskal-Wallis test (p ≤ 0.05).
Cluster analysis was conducted to classify the pathotypes based on the disease severity patterns. For cluster analysis, disease scores were converted to binary system where the resistant reaction was scored as 0 while the susceptible reaction was scored as 1. The binary data of the disease score was entered for analysis using NTSYS-pc 2.2 software (Rohlf 2005). Dendogram of cluster analysis was performed based on matrix of similarities between all pair of isolates (Jaccard coefficient) by Unweighted Pair Group Method with Arithmetic Mean (UPGMA). Jaccard coefficient was chosen as the coefficient measured similarity between all pair of isolates and negative matches are not counted (Romesburg 1984). 2 Brown specks about 0.5 mm-1.0 mm in diameter. Small lesions with distinct tan centre which surrounded by a darker brown margin approximately 1 mm in diameter. No sporulation. 3 Roundish to elliptical lesion about 1 mm-3 mm in diameter with gray centre surrounded by brown margins. Small eyespot lesions less than one and a half times the interval between thin veins or less than 1.5 mm in diameter surrounded by dark brown, lesions capable of sporulation.

4
Typical spindle shaped blast lesion capable of sporulation, 3 mm or longer with necrotic gray centre and water soaked brown margins with little or no coalescence of lesion. Intermediate size eyespot lesions less than twice the size of interval between thin veins or less than 2 mm in diameter.

5
Lesions as in scale 4 but about half of one or two leaf blade killed by coalescence of lesion. Large eyespot lesions sized more than twice the size of interval between thin veins or more than 2 mm in diameter.
Meanwhile, IRBL carrying resistance genes, Pib (IRBLb-B) and Pit (IRBLt-K59) were susceptible to all pathotypes tested. This suggested that the resistance genes are not suitable to be incorporated in developing new resistant variety.
Note: R = Resistant, S = Susceptible , '-' = information on the resistance gene is not available.
Pathotype P0.0 was avirulent to susceptible control variety, LTH. The results suggested that LTH may harboured the genes that confer specific resistance to pathotype P0.0. Similar results were reported by Fukuta et al. (2014) of which 3.3% of blast isolates tested in Cambodia showed avirulence to LTH, suggesting that the susceptible variety may contain resistance genes in its genetic background. However, according to Ling et al. (1995) the resistance genes in LTH were of minor importance, and Tsunematsu et al. (2000) reported major resistance genes have not been identified in LTH.
Based on the disease severity pattern, each pathotype was either virulent or avirulent on specific monogenic lines and the monogenic lines were also susceptible or resistant against specific pathotypes. Ellingboe and Chao (1994) described that the ability of a plant to express resistance depends on the genotype of the pathogen. Thus, the results from this study was an agreement with the genefor-gene hypothesis by Flor (1971) of which the presence of major resistance gene (R gene) in the plant is effective in recognising avirulence gene (AVR gene) in P. oryzae pathotypes (Jones & Dangl 2006;Lu et al. 2019). Subsequently, the race-specific pathogen recognition will trigger the signal transduction events that lead to pathogen invasion and their virulence functions.
Results from this study shown the avirulence of all pathotypes evaluated to IRBL harbouring R genes, Pii, Pik, Piz,Pish,Pi7(t) and Pi9. The results suggested that the six pathotypes may contain the AVR gene, avr-Pii, avr-Pik, avr-Pik-h, avr-Piz, avr-Pish, avr-Pi7(t) and avr-Pi9 that recognised R gene in their respective IRBL. Therefore, these R genes, namely Pii, Pik, Piz,Pish,Pi7(t) and Pi9 could be suitable candidates of resistance genes to be incorporated in the new breeding lines in terms of combating the current pathotypes in the field.
Cluster analysis carried out using binary data of disease score based on Jaccard similarity coefficient is presented in a dendrogram (Fig. 1). The cluster analysis suggested the disease score can be grouped into two main clusters, I and II at similarity coefficient of 0.25. Cluster I was divided into two sub-clusters, A and B, while cluster II comprising five sub-clusters, C, D, E, F and G.
Pathogenicity and virulence patterns of blast isolates in many countries have been identified using differential varieties with different types and numbers of resistance gene. The present study was similar with a study conducted by Khan et al. (2014) on the inoculation of blast isolates from fragrant rice in Bangladesh, on a set of international differential varieties consisting of 32 monogenic lines. In the study, Pish, Pi9, Pita-2 and Pita were estimated as the effective resistance genes against tested blast isolates, and 80%-90% resistance frequencies were observed.
In Japan, pathogenicity study was carried out on 310 blast isolates from eight regions then challenged to a set of differential varieties consisted of 23 monogenic lines (carrying 21 resistance genes) and two near-isogenic line with the LTH genetic background. From the 310 blast isolates tested, 306 isolates (98.7%) were found to be virulent to IRBLsh-S (Pish) . In contrast, the results of the present study revealed that all the pathotypes tested were avirulent to IRBLsh-S (Pish).
The resistance gene Pii which could be one of suitable candidates of resistance genes to be used to develop new breeding lines, has been deployed for more than two decades in China (Duan et al. 1990). However, Zhu et al. (2000) and Li et al. (2007) reported the rice cultivars that harbored Pii gene were resistant to only 54.6% of blast isolates evaluated in Yunnan province and 15.1% isolates from Guangdong province, respectively. Subsequently, studies conducted by Huang et al. (2014) on blast isolates from various rice-producing regions in China revealed that 80% of the isolates were found to have complete deletions of AVR-Pii. The results suggested breakdown of Pii resistance in the rice cultivars. The Pii gene composed of five exons encoding a putative CC-NBS-LRR protein with 1025 amino acids (Takagi et al. 2013). The AVR-Pii gene of blast pathogen corresponds with the host resistance gene Pii, subsequently triggers the defence response (Yoshida et al. 2009).
Studies conducted by Silva et al. (2004) and Yasuda et al. (2006) revealed that a loss of the AVR-Pii gene transforms an avirulent to a virulent fungal strain. Moreover, the AVR-Pii gene is located on chromosome 7 which is a highly unstable chromosome segment (Yasuda et al. 2006;Yoshida et al. 2009) which suggested a risk of gene loss and horizontal transfer events (Silva et al. 2004;Rehmeyer et al. 2006).
Although in the present study, the blast isolates used were limited, determination on the virulence patterns of the six pathotypes of P. oryzae (P0.0, P0.2, P1.0, P3.0, P7.0 and P9.0) help to determine the response of the resistance genes against specific lines. Thus, it is an initial step to elucidate the diversity and differentiation of blast pathotypes virulence using differential varieties. The knowledge on the virulence of the blast pathotypes and the resistance genes is important for breeding of new rice variety for durable resistance against blast disease.
The variability in virulence patterns of the blast pathotypes would greatly help researchers in selecting suitable donors in breeding for resistance of rice blast of which particular functioning set of genes are incorporated into a desirable rice variety for durable resistance against the blast pathogen.

CONCLUSION
In conclusion, to combat rice blast against current high frequencies of pathotypes P7.0, P0.0, P9.0 and P1.0, suitable resistance genes or donor to be incorporated for developing future rice blast resistant variety are Pii, Pik,Piz,Pish,Pi7(t) and Pi9. Meanwhile, Pib and Pit are not suitable resistance genes to be incorporated for developing new resistant variety considering their susceptible response to all pathotypes evaluated.