Overexpression of BSR1 confers broad-spectrum resistance against two bacterial diseases and two major fungal diseases in rice.

Broad-spectrum disease resistance against two or more types of pathogen species is desirable for crop improvement. In rice, Xanthomonas oryzae pv. oryzae (Xoo), the causal bacteria of rice leaf blight, and Magnaporthe oryzae, the fungal pathogen causing rice blast, are two of the most devastating pathogens. We identified the rice BROAD-SPECTRUM RESISTANCE 1 (BSR1) gene for a BIK1-like receptor-like cytoplasmic kinase using the FOX hunting system, and demonstrated that BSR1-overexpressing (OX) rice showed strong resistance to the bacterial pathogen, Xoo and the fungal pathogen, M. oryzae. Here, we report that BSR1-OX rice showed extended resistance against two other different races of Xoo, and to at least one other race of M. oryzae. In addition, the rice showed resistance to another bacterial species, Burkholderia glumae, which causes bacterial seedling rot and bacterial grain rot, and to Cochliobolus miyabeanus, another fungal species causing brown spot. Furthermore, BSR1-OX rice showed slight resistance to rice stripe disease, a major viral disease caused by rice stripe virus. Thus, we demonstrated that BSR1-OX rice shows remarkable broad-spectrum resistance to at least two major bacterial species and two major fungal species, and slight resistance to one viral pathogen.

Introduction

In the natural environment, plants encounter many species of pathogenic microorganisms, such as fungi, bacteria and viruses. The damage caused by microbial diseases is one of the most important limiting factors for crop production. To solve this problem, improvement of host resistance against these pathogens is the most economical and environmentally friendly approach. Rice (Oryza sativa L.) is one of the most important food crops and is a staple food for approximately 50% of the world’s population (Liu ). Moreover, it is a model plant of monocotyledonous species. Bacterial leaf blight caused by bacterial pathogen Xanthomonas oryzae pv. oryzae (Xoo), blast by fungus Magnaporthe oryzae, brown (leaf) spot by fungus Cochliobolus miyabeanus, and stripe mosaic caused by viral pathogen rice stripe virus are the major rice diseases worldwide and result in serious losses of rice production.

To breed blast-resistant rice, efforts have aimed to introduce the resistance (R) genes into susceptible cultivars. The R gene is a key component of disease resistance to a particular pathogen and is often associated with a hypersensitive response (HR) (Flor 1971). In most cases, the resistant cultivars with R genes remain effective for only a few years in agricultural production (Dean ) because new biotypes of the pathogen that can overcome the R gene often appear after release of the resistance cultivar. By contrast, although large numbers of quantitative trait loci (QTLs) (or quantitative genes) for bacterial leaf blight or blast resistance have been identified, these sources have not been used effectively in rice improvement because the genetic control of quantitative resistance is complex. Therefore, breeding for cultivars that exhibit broad-spectrum and durable disease resistance is a top priority in rice improvement programs around the world.

In addition to bacterial leaf blight and blast, there are several other important diseases in rice, such as brown spot and bacterial seedling rot. Brown spot disease is caused by the fungus C. miyabeanus, a representative necrotrophic pathogen, and is one of the most prevalent diseases in all rice-growing areas. C. miyabeanus infects plant tissues such as leaves and spikelets in all development stages. Fungicides, such as iprodione and propiconazole, are effective means to manage this disease (Moletti ). However, the use of resistant varieties would be preferable because fungicides are expensive and not environmentally friendly. Meanwhile, rice cultivar ‘Tadukan’ offers quantitative resistance to brown spot disease. However, no major genes conferring immunity to this disease have been identified, although three QTLs for disease resistance have been identified (Sato ). So far, genetic studies of resistance to brown spot disease have been superficial.

Bacterial seedling rot and bacterial grain rot (bacterial panicle blight) in rice are caused by bacterial pathogen Burkholderia glumae, which is also a necrotrophic pathogen (Iwai , Mizobuchi ). Recently, these diseases have become an increasingly serious problem in global rice production because of global warming and climate change; B. glumae prefers high temperature and humidity (Ham ). Some studies reported partially resistant varieties for these diseases and several QTLs for resistance to bacterial seedling rot and bacterial grain rot have been identified (Mizobuchi , 2013b, 2015, Pinson ). However, these resistances are not strong. Meanwhile, Iwai reported that transgenic rice lines overproducing Asthi1, an oat leaf thionin gene, showed enhanced resistance to B. glumae. However, it has not been applied to actual breeding.

Rice stripe disease, caused by rice stripe virus (RSV), is one of the major viral diseases in East Asia. The majority of japonica cultivars in East Asia are highly susceptible to RSV (Wang ) although a few rice cultivars/lines that show resistance to RSV have been described (Noda , Zhang ).

Breeding crops with broad-spectrum disease resistance using genetic resources is one of the principal goals of crop improvement. However, only a few genes have been identified as genetic resources for broad-spectrum disease resistance in rice. Hence, the transgenic approach could be a viable alternative. In recent years, several gain-of-function transgenic mutant populations have been developed in rice (Hsing , Jeong , Mori , Nakamura ). Meanwhile, Kondou produced more than 20,000 independent Arabidopsis transgenic lines overexpressing rice full-length cDNAs (rice-FOX Arabidopsis lines) to enable high-throughput screening for rice genes. We performed screening for pathogen resistance using these lines (Dubouzet ). As a result, we identified several rice genes conferring resistance to both bacterial Pseudomonas syringae pv. tomato DC3000 and fungal Colletotrichum higginsianum in Arabidopsis. One of the genes named BROAD-SPECTRUM RESISTANCE1 (BSR1), encoding a receptor-like cytoplasmic kinase, conferred resistance to Xoo and M. oryzae when overexpressed in rice (Dubouzet ).

In this paper, we report that overexpression of BSR1 not only conferred non-race-specific resistance to Xoo and M. oryzae, but also extended resistance to B. glumae and C. miyabeanus. Moreover, overexpression of BSR1 is likely to confer partial resistance to RSV.

Materials and Methods

Plant materials

Rice (Oryza sativa L.) wild-type (WT) cultivar ‘Nipponbare’, Xoo-resistant cultivar ‘Asominori’, RSV-resistant cultivar ‘Sainokagayaki’ and two transgenic plant lines (BSR1-OX-5 and -9) were grown under greenhouse conditions at 27°C to 30°C. BSR1-OX-5 and BSR1-OX-9 correspond to the previously reported AK070024:OX-5 and AK070024:OX-9 (Dubouzet ), respectively.

For disease resistance tests, except for B. glumae, dehusked seeds were surface sterilized, sown on one-half-strength MS medium (Wako Pure Chemicals, Osaka, Japan), containing 3% (w/v) sucrose and 0.4% (w/v) Gelrite (Wako Pure Chemicals), in Agripots and grown in the growth chamber at 28°C in the dark for 3 days, then at 25°C under long-day conditions (16 h light [60–70 μmol m−2 s−1]/8 h dark) for 4–7 days. For transgenic seeds, Hygromycin B (30–50 μg/ml; Wako Pure Chemicals) was added to the medium. WT seedlings and hygromycin-resistant transgenic seedlings were transplanted into soil (Bonsol No. 2, Sumitomo Kagaku Kougyo, Osaka, Japan) and used for disease resistance tests.

Pathogens and pathogen cultures

The bacterial isolates used in this study were T7147 (MAFF311019, race II) and T7133 (MAFF311020, race III) of Xoo and AZ8204 (MAFF301682) of B. glumae, and the fungal isolates were Hoku1 (MAFF101512, race 007.0) of M. oryzae and H11-42-1 of C. miyabeanus.

Culture procedures for the various pathogens for inoculum were as follows. Xoo were cultured on PSA agar plates (1% proteose peptone, 1% sucrose, and 1.5% bacto agar) for 2 days at 28°C under dark conditions. B. glumae were cultured on King B agar plate (2% proteose peptone, 0.15% K2HPO4, 0.15% MgSO4, 1% glycerin and 1.5% agar, Eiken chemical, Tochigi, Japan) at 28°C for 2 days under dark conditions. M. oryzae was grown on oatmeal agar plates (3% oatmeal, 0.5% sucrose, and 1.6% bacto agar) at 25°C in the dark for 10 days, then under continuous illumination for 4 days to induce sporulation. C. miyabeanus was grown on V8 agar plates (20% V8 juice (Campbell soup company, Camden, NJ, USA), 0.3% CaCo3, and 1.5% bacto agar) at 25°C in the dark for 5 to 6 days, and then under a 12/12 h light/dark regime for 3 to 4 days to induce sporulation. The cultured pathogens were scraped and used to produce inocula.

Expression analysis of BSR1 by quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR)

Total RNA was isolated from rice leaves using Isogen (Wako Pure Chemicals) followed by further purification with the RNeasy mini kit (Qiagen, Valencia, CA, USA). First-strand cDNAs were synthesized from equal amounts of total RNA using a ReverTra Ace qPCR RT Master Mix with gDNA Remover Kit (Toyobo, Osaka, Japan) in a total volume of 10 μl, as described by the manufacturer. qRT-PCR was performed with the Thermal Cycler Dice TP800 system (Takara, Tokyo, Japan) using a Kapa SYBR FAST qPCR kit (Kapa Biosystems, Cape Town, South Africa) as described by the manufacturer. The primers used for qPCR were as follows: BSR1 5′-AGGTGAGGTTGCACTCTGCT-3′ and 5′-CCAAGAATCCACCAACTCGT-3′ as described Dubouzet ; those for Rubq1 were 5′-GGAGCTGCTGCTGTTCTAGG-3′ and 5′-TTCAGACACCATCAAACCAGA-3′ as an internal control, as described Jiang . Transcript levels of BSR1 were normalized to the endogenous rice reference gene (Rubq1).

Test for resistance to Xoo

Rice seedlings transplanted in soil were grown in a growth chamber until the 6–8-leaf stage at 25°C under long-day conditions (16 h light/8 h dark) and then used for evaluation. The suspensions of Xoo (isolate T7147 or T7133) for inocula were adjusted to OD600 = 0.3 with sterile water. The top leaf blades of the tested plants were cut with scissors pre-wetted with inoculum at about 5 cm from the tip, and the cut ends (about 5 mm from the end) were dipped in inoculum for 10 s. The disease symptoms (lesion length) of inoculated plants were assessed 2 weeks after inoculation, as described previously (Dubouzet ).

Test for resistance to M. oryzae

Rice seedlings transplanted in soil were grown in a greenhouse until the four-leaf stage at 28°C under a natural photoperiod. The plants were inoculated by spraying a spore suspension of M. oryzae (isolate Hoku1 (MAFF101512, race 007.0)). The detailed procedure for producing spore suspension is as follows: The mycelia of M. oryzae were scraped and the gel surface was flooded with sterile water containing 0.01% Tween 20. The suspension was filtered through a Kimwipe, and the resulted spore suspension was collected. The spore suspension was adjusted to a concentration of 6.7 × 105 spores/ml and used for inoculation. After inoculation, the plants were placed in a dark chamber at 26°C and 100% humidity for 20 h, and then further cultured in the greenhouse. Evaluation of resistance was based on the total number of compatible lesions that appeared on the 3rd and 4th leaf blades of each plant 5 days after inoculation.

Evaluation of bacterial pathogen B. glumae resistance

WT ‘Nipponbare’ rice seeds and T3–5 seeds of BSR1-OX-5 and -9 were sterilized by soaking in 70% ethanol for 30 s and Antiformin (available chlorine 5%) for 20 min. The seeds were then rinsed with sterilized water. The sterilized seeds were soaked in sterilized water at 28°C for 2 to 3 days in the dark and pre-germinated to 1–2 mm of sprout. The pre-germinated seeds were soaked in suspensions of B. glumae adjusted to OD520 = 0.004 and held under a vacuum for 1 min. The inoculated seeds were dried and sown in sterilized soil (Bonsol No. 2, Sumitomo Kagaku Kougyo). The inoculated seeds were incubated in a growth chamber at 28°C with 100% humidity under a 14-h photoperiod. Plant phenotypes were classified as ‘healthy’ or ‘diseased’ at 7–10 days after inoculation and the percentage of healthy plants among the total seeds used for inoculation was calculated as the survival ratio.

Evaluation of fungal pathogen C. miyabeanus resistance

Rice seedlings transplanted in soil were grown in a greenhouse until the six-leaf stage at 28°C under a natural photoperiod. The inoculation method of C. miyabeanus was the same as that of M. oryzae, except as follows. The concentration of the spore suspension was adjusted to 104 spores/ml. Evaluation of resistance was based on the total number of compatible lesions that appeared on the 5th and 6th leaf blades of each plant 5 days after inoculation.

Assessment of resistance to RSV

BSR1-OX seedlings selected by Hygromycin resistance (30–50 μg/ml), WT and cv. ‘Sainokagayaki’ seedlings were transplanted into soil at about 10 seedlings per pot, and grown in a greenhouse until the two-leaf stage at 28°C under a natural photoperiod. The rice seedlings were exposed to approximately 10 viruliferous small brown planthoppers (Laodelphax striatellus Fallén) per plant in an inoculation cage for 1 day to inoculate RSV, as described previously (Satoh ). After the inoculation period, the insects were killed with insecticide and the seedlings were transferred to the same greenhouse. The virus infection was evaluated by enzyme-linked immunosorbent assay (ELISA) using antiserum against RSV nucleocapsid protein, as described previously (Shimizu ). The pieces (about 1 cm) of leaf sheath plus stem tissue from inoculated seedlings were harvested for ELISA at 16 days after inoculation. Resistance to RSV was calculated by the ratio of diseased seedlings detected by ELISA among all inoculated seedlings.

Results

Transcript level of BSR1 in BSR1-OX rice lines

The cDNA of BSR1 was inserted downstream of the constitutive maize ubiquitin promoter (Fig. 1A), and the construct was used to generate transgenic rice lines overexpressing BSR1, as described previously (Dubouzet ). The resulting two transgenic lines, BSR1-OX-5 (former name, AK070024:OX-5) and -9, were used for various disease resistance tests. To gain sufficient seeds for the disease resistance tests, the two transgenic lines were subjected to acceleration of advanced generations. To confirm overexpression of BSR1, we examined the transcript level of BSR1 by qRT-PCR in T3–4 generations of the BSR1-OX lines (Fig. 1B). Transcript levels of BSR1-OX-5 and -9 lines were 159- and 130-fold higher than that of ‘Nipponbare’ (WT), respectively. Thereafter, we used the plants of T3–5 lines for various disease resistance tests.

Fig. 1

Schematic representation of pRiceFOX-BSR1 and the transcript level of BSR1 in BSR1-OX rice. (A) The pRiceFOX-BSR1 construct for overexpression of BSR1. (B) Transcript levels of BSR1 in T3–4 generations of BSR1-OX lines. Second youngest leaf blades of BSR1-OX and wild-type (WT) plants at the eight-leaf stage were used for measurement. Transcript levels of BSR1 were normalized to that of an endogenous Rubq1 housekeeping gene (Jiang ). Values are means ± SD (n = 4).

Overexpression of BSR1 confers resistance to multiple races of Xoo and M. oryzae in rice

We have reported that BSR1-OX rice shows strong resistance to isolate T7174 (race I) of Xoo, a bacterial pathogen for rice bacterial leaf blight, and to isolate Kyu89-246 (race 003.0) of M. oryzae, a fungal pathogen for rice blast (Dubouzet ). Hence, it would be plausible that BSR1 also confers resistance to other races of Xoo and M. oryzae. First, we examined whether BSR1-OX rice extended resistance to isolates T7147 (race II) and T7133 (race III) of Xoo. The WT and the resistant control, cv. ‘Asominori’, inoculated with isolate T7147 (race II) developed extended lesions from the cut (inoculated) end of the leaves, whereas BSR1-OX-5 and -9 plants showed restricted lesions (Fig. 2A). Lesion lengths in the inoculated BSR1-OX-5 and -9 were about 10 mm and 7 mm long, whereas those in WT and ‘Asominori’ were about 106 mm and 95 mm long (Fig. 2A). Two BSR1-OX lines showed 10- and 15-fold reductions in lesion length compared with the WT. Similarly, the results of inoculation of isolate T7133 (race III) are shown in Fig. 2B. The two BSR1-OX lines showed 10- and 12-fold reductions in lesion length compared with the WT. These results indicated that overexpression of BSR1 confers strong resistance to both T7147 (race II) and T7133 (race III), as well as to previously shown T7174 (race I). The results suggested that overexpression of BSR1 conferred non-race-specific resistance to Xoo. Although ‘Asominori’ has very strong resistance to the isolate T7174 (race I), the resistances to T7147 (race II) and T7133 (race III) were moderate (Kaku and Kimura 1989). Resistance levels of ‘Asominori’ shown here were similar to the report and ‘Asominori’ showed more race-specific resistance.

Fig. 2

Disease resistance to multiple races of X. oryzae pv. oryzae. Isolates T7147 (race II; A) and T7133 (race III; B) of Xoo were used for infection. Arrowheads indicate the point of inoculation. Lesion lengths in BSR1-OX plants were significantly shorter than those in wild-type (WT) and ‘Asominori’, the resistance control cultivar. Values are means ± SD, n = 6–18. Different letters indicate significant differences (P < 0.05 by Tukey-Kramer’s test).

Next, to examine whether overexpression of BSR1 also confers resistance to another race of M. oryzae in rice, isolate Hoku1 (MAFF101512, race 007.0) was used for inoculation. Lesion numbers in the BSR1-OX-5 and -9 lines were significantly smaller than those in the WT plants (Fig. 3). Thus, because BSR1-OX lines conferred strong resistance to isolate Hoku1 (race 007.0) in addition to the previously shown isolate Kyu89-246 (race 003.0), we hypothesized that overexpression of BSR1 conferred non-race-specific resistance to M. oryzae.

Fig. 3

Disease resistance to another race of M. oryzae. Isolate Hoku1 (race 007.0) of M. oryzae was used. Lesion numbers in BSR1-OX plants were significantly lower than those in wild-type (WT) plants (***P < 0.001 by t-test). Values are means ± SD, n = 3–7.

Extended resistance to another bacterial pathogen, Burkholderia glumae

Bacterial seedling rot and bacterial grain rot (bacterial panicle blight) are caused by the bacterial pathogen B. glumae. The latter is an increasingly important disease problem in global rice production (Ham ). Many genetic studies for resistance to bacterial grain rot have been reported (Mizobuchi , 2015, Pinson , Sayler , Wasano and Okuda 1994). However, there are few reports on resistance to bacterial seedling rot, because such resistance is a complex characteristic influenced by environmental factors (Iwai , Mizobuchi ). We were interested in whether overexpression of BSR1 conferred resistance to bacterial seedling rot in rice. In the test for resistance to bacterial seedling rot, non-germinated seeds are usually used for inoculation by soaking. However, it was difficult to evaluate the resistance by this method because BSR1-OX rice displayed a decreased germination rate (Dubouzet ). Therefore, we gathered only pre-germinated seeds for use in the disease resistance test. To determine the condition for inoculation of pre-germinated seeds, we performed a preliminary experiment using various concentrations of B. glumae suspension and WT seeds. The disease symptoms were classified as shown in Fig. 4A. Browning of the leaf sheath was usually detected together with a dwarf phenotype in diseased plants. Disease resistance was evaluated by the survival ratio, indicating the ratio of healthy plants to total seeds used for infection and shown as a percentage. When non-germinated (NG) seeds were inoculated by soaking in suspensions of three different concentrations of B. glumae, no healthy plant survived in all concentrations (Fig. 4B). In contrast, when pre-germinated seeds were used, 98%, 43% and 2% of healthy plants survived after soaking in suspensions of low (OD520 = 0.0004), medium (OD520 = 0.004) and high (OD520 = 0.04) concentrations of the bacteria, respectively. Thus, the optimal concentration range to evaluate resistance in this experiment was OD520 = 0.004–0.04. In subsequent experiments, we evaluated disease resistance by this method using pre-germinated seeds.

Fig. 4

Preliminary experiments to determine the optimal concentration of bacterial pathogen, Burkholderia glumae, for inoculation. (A) The classification of disease symptoms at 7 days after inoculation. Healthy, normal phenotype; Diseased, dwarf or dead phenotype. Bar = 50 mm. (B) Results of the preliminary experiment. Photograph shows plants at 7 days after inoculation (n = 50–60). Non-germinated seeds (NG) and pre-germinated seeds (PG) were used for inoculation. Bar = 50 mm. Survival ratio (%) = (number of healthy plants/number of total seeds) × 100

Resistance to B. glumae was evaluated for BSR1-OX pre-germinated seeds. Survival ratios in BSR1-OX-5 and -9 lines were two times higher than that in the WT (Fig. 5). These results indicated that BSR1-OX lines displayed resistance to B. glumae in rice.

Fig. 5

Disease resistance to another bacterial pathogenic species, Burkholderia glumae. Pre-germinated seeds of BSR1-OX and wild-type (WT) were inoculated with B. glumae. Concentration of inoculation was OD520 = 0.004. Survival ratio was calculated 8 days after inoculation (n = 17). Tests were performed three times with similar results.

Extended resistance to another major fungal pathogen, Cochliobolus miyabeanus

Brown spot disease is caused by the fungal pathogen, C. miyabeanus. We hypothesized that BSR1 conferred resistance to brown spot disease as well as to rice blast in rice. First, we investigated whether the expression of BSR1 changed by inoculation with C. miyabeanus in WT plants (Fig. 6A). C. miyabeanus was spray-inoculated onto WT plants, and the transcript level of BSR1 was measured by qRT-PCR. As a result, inoculated plants showed inducible expression of BSR1 compared with mock control after inoculation (Fig. 6A), although the transcript levels of BSR1 in the inoculated plants were much lower than those in BRS1-OX lines. This result suggested that BSR1 is involved in innate immunity against brown spot in rice. Hence, we examined whether overexpression of BSR1 conferred resistance to C. miyabeanus (Fig. 6B). Lesion numbers in BSR1-OX-5 and -9 plants were significantly lower than those in the WT plants (***P < 0.001 by t-test, Fig. 6B). Thus, overexpression of BSR1 conferred significant resistance to C. miyabeanus.

Fig. 6

Disease resistance to another fungal pathogenic species, Cochliobolus miyabeanus. (A) Relative expression levels of BSR1 in wild-type (WT) plants on inoculation with mock or C. miyabeanus. Seventh-leaf blades were used for inoculation. Total RNAs at 0 to 3 days after inoculation were extracted. Transcript levels of BSR1 were normalized to that of Rubiq1. Expression levels of BSR1 in inoculated leaves were up-regulated compared with mock at 1 to 3 days. Values are means ± SD, n = 3. An asterisk indicates a statistically significant difference from the mock at 2 days (P < 0.05 by t-test). (B) Resistance to C. miyabeanus in BSR1-OX rice. Lesion numbers in BSR1-OX plants were significantly lower than those in WT plants (***P < 0.001 by t-test). Values are means ± SD, n = 4–12.

BSR1-OX rice were slightly resistant to rice stripe virus (RSV)

We examined whether overexpression of BSR1 could confer resistance against a viral pathogen, RSV, in rice, because the majority of japonica cultivars, including cv. ‘Nipponbare’, are susceptible to RSV. The results are shown in Fig. 7. After inoculation of RSV, the percentages of diseased seedlings detected by ELISA in BSR1-OX-5 and -9 seedlings were slightly lower than those in WT seedlings, but were higher than that in cv. ‘Sainokagayaki’, the RSV-resistant control cultivar (Fig. 7). Thus, overexpression of BSR1 could confer slight resistance to RSV, although its resistance level was weaker than that of ‘Sainokagayaki’, which possesses the highly resistant Stvb-i gene to RSV (Shimizu ).

Fig. 7

Disease resistance to a viral pathogen, rice stripe virus. Diseased seedling ratio, percentage of diseased seedlings detected by enzyme-linked immunosorbent assay (ELISA) in all inoculated seedlings at 16 days after inoculation. n = 8 to 10. Tests were performed three times with similar results.

Discussion

From the point of view of breeding, the quality of broad-spectrum resistance against two or more different pathogen species is an agronomically important trait. We previously reported that overexpression of BSR1, encoding a receptor-like cytoplasmic kinase, confers remarkable resistance to both bacterial and fungal pathogens in Arabidopsis, and that it confers resistance to the bacterial pathogen, Xoo, and fungal pathogen, M. oryzae, in rice (Dubouzet ). In this study, we first showed that BSR1 confers resistance to multiple races of Xoo and M. oryzae in overexpressing rice. In addition, BSR1-OX rice showed extended resistance to another bacterial pathogen, B. glumae, and another fungal pathogen, C. miyabeanus. Thus, overexpression of BSR1 conferred broad-spectrum disease resistance to at least two bacterial and two fungal pathogenic species in rice. Therefore, it would be natural to consider that the resistance to bacterial and fungal pathogens by BSR1 is non-race specific. Furthermore, BSR1-OX rice showed slight resistance to RSV. To the best of our knowledge, there are no other genes that confer such multi-disease resistance.

As a typical example of interactions with host plants and pathogens, the plant defense against pathogens, such as bacteria and fungi, is provided through cell-surface-localized pattern recognition receptors that detect pathogen-associated molecular patterns (PAMPs), and result in pattern-triggered immunity (PTI). To counteract this innate immunity, pathogens deliver effector proteins into the host cell to suppress PTI. In some cases, plants use intracellular resistance (R) proteins to detect race-specific effectors, which results in effector-triggered immunity, often associated with hypersensitive response and programmed cell death. The BSR1 gene, OsRLCK278, is one of 379 RLCK genes encoding receptor-like cytoplasmic kinases in rice (Dubouzet ). BSR1 is classified into the RLCK-VII protein family, which belongs to the same family as BIK1 and PBS, the well-characterized Arabidopsis RLCKs involved in plant defense (Dubouzet ). Arabidopsis BIK1 was originally isolated as a gene involved in the defense against necrotrophic fungal pathogens (Veronese ). BIK1 also associates with a flagellin receptor complex, FLS2/BAK1, which is rapidly phosphorylated upon perception of flagellin. In addition, BIK1 is also phosphorylated by another PAMP, translation elongation factor (EF-Tu). Hence, BIK1 is considered to mediate PTI signal transduction, such as production of a reactive oxygen species (ROS) burst, activation of mitogen-activated protein kinases (MAPKs) and calcium-dependent protein kinases (CPKs or CDPKs), transcriptional reprogramming, and ultimately immunity, from multiple PAMP receptor complexes (Chinchilla , Heese , Lu , Roux , Schulze , Schwessinger , Sun ). Furthermore, BIK1 regulates calcium influx, phosphorylates and activates NADPH oxidase RBOHD, and not only induces the defense response (including the ROS burst that is a poisonous factor directly attacking pathogens), but also prevents invasion of pathogenic bacteria by closing stoma (Kadota , Li ). Thus, BIK1 plays pivotal roles in recognition of PAMPs and subsequent signal transductions. In rice, OsRLCK185, which belongs to the RLCK-VII family, was reported recently (Yamaguchi ). OsRLCK158 interacts with a pattern recognition receptor OsCERK1, which recognizes chitin and peptidoglycan at the plasma membrane, and regulates a MAP kinase cascade that leads to a PTI defense response. Taken together, it is likely that the BSR1 protein interacts with various transmembrane receptors that recognize PAMPs of Xoo, M. oryzae, B. glumae and/or C. miyabeanus, which functions in linking multiple PAMP receptor complexes to downstream intracellular signaling and enhances a part of PAMP-mediated defense.

From the viewpoint of the pathogen character, the lifestyles of pathogens used in this study are classified as hemibiotrophs or necrotrophs. The blast fungus, M. oryzae, and bacterial leaf blight bacteria pathogen, Xoo, are considered hemibiotrophs (Van Bockhaven ). Hemibiotrophs are characterized by an initial infectious period of the biotrophic stage in which the pathogens grow within host cells before switching to a necrotrophic growth stage when lesions become apparent (Wilson and Talbot 2009). Xoo has been considered mostly as a biotroph but is probably best classified as a hemibiotroph (De Vleesschauwer ). Furthermore, the pathogens of bacterial seedling rot, B. glumae, and of brown spot, C. miyabeanus, have been classified as necrotrophs (Iwai , Su’udi ). Thus, we demonstrated that BSR1-OX rice conferred resistance not only to hemibiotrophs but also necrotrophs.

Salicylic acid (SA), jasmonic acid (JA) and ethylene (ET) are typical defense hormones, and the regulation of their signaling networks against pathogen infection is well established, especially in the dicotyledonous model plant Arabidopsis (Van Bockhaven ). In Arabidopsis, SA-dependent defenses are generally associated with resistance to biotrophs, whereas JA/ET-dependent defenses are generally associated with necrotrophs (Van Bockhaven , Thomma , De Vos ). By contrast, the defense system in rice is considered to be controlled by a more complicated signaling network (De Vleesschauwer ). SA-, JA- and ET-dependent defenses all enhanced resistance to hemibiotrophic M. oryzae (Ahn , Helliwell , Iwai , Mei , Peng , Qiu , Schweizer , Shimono , 2012). Meanwhile, SA- and JA-dependent defenses were involved in disease resistance against hemibiotrophic Xoo (Babu , Qiu , Shimono , Tanaka , Taniguchi ), whereas ET-dependent defense played a negative role (Shen ). Among the hemibiotrophs, the role of ET is the opposite of Xoo and M. oryzae. Similarly, ET-dependent defense is involved negatively in the resistance against C. miyabeanus, a necrotrophic pathogen (De Vleesschauwer ), while no reports have implicated SA- and JA-dependent defenses against C. miyabeanus (Van Bockhaven ). More interestingly, ABA, which antagonizes the SA pathway, is involved in the resistance to C. miyabeanus (De Vleesschauwer ), although ABA promotes susceptibility against Xoo and M. oryzae (Jiang , Xu ). The plant hormone defense network against B. glumae is unknown in rice. RSV resistance is implicated in enhancing the SA signal (Wang ). Taken together, understanding the mechanism of broad-spectrum resistance of BSR1 from the point of view of plant hormones is complex.

Currently, several genes have been reported to confer broad-spectrum disease resistance in rice. For example, overexpression of OsWRKY13 or OsWRKY45 enhanced resistance to Xoo and M. oryzae in rice by mediating SA signaling (Qiu , Shimono , 2012). Unlike BSR1-OX rice, however, there are few reports that overexpressed genes enhanced resistance to necrotrophic pathogens, B. glumae and C. miyabeanus in rice. Meanwhile, overexpression of OsWRKY30 or OsACS2 encoding a key enzyme of ET biosynthesis enhanced resistance to fungal pathogens M. oryzae and necrotrophic Rhizoctonia solani in rice. The resistance conferred by OsWRKY30 was associated with the activation of JA synthesis-related genes and the increased accumulation of endogenous JA (Peng ), and that by OsACS2 was associated with the increased level of endogenous ET (Helliwell ). Hence, we speculate that the broad-spectrum disease resistance against two hemibiotrophs (Xoo and M. oryzae) and two necrotrophs (B. glumae and C. miyabeanus) in BSR1-OX rice is based on an SA and JA/ET combined pathway.

Here, BSR1-OX rice displayed strong resistance to three tested races (races I to III) of Xoo. In contrast, ‘Asominori’ showed more race-specific resistance. Although the resistance to isolate T7174 (race I) on ‘Asominori’ was very strong (Dubouzet ), the resistances to T7147 (race II) and T7133 (race III) were moderate (Fig. 2), almost similar to the report of Kaku and Kimura (1989). ‘Asominori’ has been reported to have an Xa1-as(t) gene at the Xa1 locus (Ise ). Xa1-as(t) is implicated in the strong resistance to T7174 (race I). ‘Asominori’ has also been suggested to have minor-affected loci that are involved in the quantitative resistance to T7133 (race III) (Yoshimura ). BSR1-OX rice showed strong resistance to both T7174 (race I) and T7133 (race III); therefore, the defense mechanism of BSR1-OX rice would be different from that of ‘Asominori’. From the viewpoint of breeding, overexpression of BSR1 could confer more useful non-race-specific resistance to Xoo in rice compared with using the resistance genes of ‘Asominori’.

In the previous report, BSR1-OX rice displayed strong resistance to race 003.0 (isolate Kyu89-246) and the resistance level was stronger than that in cv. ‘Sensho’ (Dubouzet ), which has a strong non-race-specific resistance to M. oryzae associated with pi21 (Fukuoka ). In this paper, BSR1-OX rice displayed extended strong resistance to race 007.0 (isolate Hoku1; Fig. 3). Hence, non-race-specific resistance or field resistance against M. oryzae is also promising in BSR1-OX rice. Taken together, overexpression of BSR1 could confer more promising leaf blight and blast resistances compared with the resistant cultivars ‘Asominori’ and ‘Sensho’, respectively, in many useful O. sativa varieties. Furthermore, it is plausible that BSR1-OX rice also shows non-race-specific resistance to B. glumae and C. miyabeanus, and exhibits resistance to other pathogen species, because BSR1-OX rice showed resistance to all pathogens tested.

In conclusion, BSR1, when overexpressed in rice, conferred broad-spectrum disease resistance against at least four diseases: bacterial leaf blight, blast, bacterial seedling rot and brown spot, and slight resistance against rice stripe disease by RSV. BSR1 represents a highly valuable and convenient genetic resource because it confers resistance to various diseases by a single gene. In the future, the defense mechanism conferred by BSR1 will be clarified to use the BSR1 gene effectively.