Identification of QTLs for rice brown spot resistance in backcross inbred lines derived from a cross between Koshihikari and CH45.

Rice brown spot (BS), caused by Bipolaris oryzae, is one of the major diseases of rice in Japan. Quantitative resistance has been observed in local cultivars (e.g., CH45), but no economically useful resistant variety has been bred. Using simple sequence repeat (SSR) polymorphic markers, we conducted quantitative trait locus (QTL) analysis of BS resistance in backcross inbred lines (BILs) from a cross between indica CH45 (resistant) and japonica Koshihikari (susceptible). On the basis of field disease evaluations in 2015 and 2016, four QTLs contributing to BS resistance were identified on chromosomes 2 (qBSR2-kc), 7 (qBSR7-kc), 9 (qBSR9-kc), and 11 (qBSR11-kc). The 'CH45' alleles at qBSR2-kc, qBSR7-kc, and qBSR11-kc and the 'Koshihikari' allele at qBSR9-kc increased resistance. The major QTL qBSR11-kc explained 23.0%-25.9% of the total phenotypic variation. Two QTLs (qBSR9-kc and qBSR11-kc) were detected in both years, whereas the other two were detected only in 2016. Genetic markers flanking these four QTLs will be powerful tools for marker-assisted selection to improve BS resistance.

Introduction

Rice brown spot (BS) caused by the fungus Bipolaris oryzae (Breda de Haan) Shoemaker was the main cause of the Bengal famine of 1943 (Padmanabhan 1973) and remains one of the most serious diseases causing severe yield losses. The severe epidemic of BS in Bengal was attributed to the higher temperatures than in a normal year (20–30°C) and rainfall during rice flowering and grain-filling stages (Padmanabhan 1973). In temperate regions such as Japan, high temperatures during these rice growth stages also tend to increase the severity of BS epidemics (Oohata and Kubo 1974). The BS epidemic area in Japan has been steadily increasing over the last decade; in 2015, it reached 178,121 ha and was the fourth largest area after those of sheath blight (586,208 ha), leaf blast (330,186 ha), and neck blast (266,145 ha) (Japan Plant Protection Association 2016). The spread of BS in Japan was promoted by two factors. First, the doses or application frequency of fungicides have decreased to permit low-input sustainable agriculture (Niigata Agricultural Research Center 2010, Yamaguchi ). Second, so-called global warming or climate change including high temperatures, heavy rain, and droughts has affected Japan (Ministry of Agriculture, Forestry and Fisheries 2015, Mizobuchi ).

Although chemical control of BS has evolved, it is too expensive and hence host resistance is given priority in disease control strategies (Sato ). No major gene confers immunity to the pathogen (Sato ). Several studies have examined the genotypic variability of BS resistance in rice, and some research groups, including us, have started to identify and use quantitative trait loci (QTLs) for BS resistance (reviewed by Mizobuchi ). An indica cultivar CH45 from India shows a high level of partial resistance (Misra 1985), and we recently confirmed its resistance in Japan (Matsumoto ). We selected this cultivar as a standard cultivar and a useful donor for the BS resistance breeding program in Japan (Matsumoto ).

The objectives of this study were to conduct QTL analysis for BS resistance using backcrossed inbred lines (BILs) derived by backcrossing a regionally adapted japonica cultivar, Koshihikari, to CH45, and to introduce BS resistance QTLs into Koshihikari.

Materials and Methods

Fungal strain

Bipolaris oryzae strain Iga-2 (deposited as MAFF 245177 at the Genebank Project of National Agriculture and Food Research Organization (NARO), Tsukuba, Japan) was used. Inoculation, culture of the mycelia, and induction of conidiophore formation under irradiation with black-light lamps were based on the methods of Kihara and Kumagai (1994).

Plant materials

The BS-resistant donor, Oryza sativa L. ssp. indica, cv. CH45 (acc. no. JP12893), was provided by the Gene Bank Project, NARO. BILs were developed by first crossing CH45 with the susceptible parent O. sativa L. ssp. japonica, cv. Koshihikari. An F1 plant was backcrossed to Koshihikari and 60 BC1F1 plants were obtained. Successive backcrossing with Koshihikari was performed to produce 229 BC2F1 seeds. Generations were advanced by means of single-seed descent. A population of 190 BILs at the BC2F5 generation was obtained in 2015.

Field evaluation of brown spot resistance

In 2015 and 2016, BS resistance in the BIL population was evaluated for QTL mapping in a research field at the Mie Prefecture Agricultural Research Institute (MPARI, Iga, Mie, Japan) with two replications, according to Matsumoto . Spreader plants inoculated with the fungus were planted around experimental plots as described by Matsumoto . Disease scores using a scale from 0 (no incidence) to 9 (severe) were recorded 108 and 105 days after transplanting in 2015 and 2016, respectively.

Linkage and QTL analysis

Total DNA was extracted from leaves by using the CTAB method (Murray and Thompson 1980). To construct a linkage map, we used 126 polymorphic rice SSR markers. SSR analysis was carried out according to McCouch . Linkage groups and marker order were determined using version 3 of the MAPMAKER/EXP software (Lander ). The resulting genetic linkage map was visualized by using a Microsoft Excel macro, Map Draw (Liu and Meng 2003). QTL analysis was performed using version 2.5 of Windows QTL Cartographer (Wang ) with the default composite interval mapping and control parameters, standard model 6, five control markers, a 10-cM window size, and the forward and backward regression model. We used the genome-wide threshold value α = 0.05 to detect putative QTLs based on the results of 1000 permutations.

Results

Phenotypic analysis of parental lines and their progeny

Disease severity of BS at MPARI was greater in 2016 than in 2015, but distinct differences in BS field resistance were observed between the parental cultivars, CH45 and Koshihikari. The mean disease score of CH45 was 2.0 in 2015 and 2.6 in 2016 and that of Koshihikari was 4.1 in 2015 and 5.9 in 2016 (Figs. 1, 2). The disease scores of the BIL population in 2015 and 2016 were normally distributed, and some lines transgressively segregated in both directions (Fig. 2), indicating quantitative inheritance of field resistance. Significant negative correlations (P < 0.01) between disease scores and days to heading (DTH) were found in the BIL population (r = −0.289 in 2015 and −0.373 in 2016) (data not shown).

Fig. 1

BS lesions on the leaves of Koshihikari and CH45. Photos were taken 122 days after transplanting in 2016.

Fig. 2

Frequency distribution of BS disease score in 190 BILs (Koshihikari*3/CH45) in 2015 and 2016. Arrowheads indicate the mean values for the parents.

QTLs for BS resistance

We used 126 SSR polymorphic markers for map construction. The ratio of genotypes in the population was as follows: 83.3% homozygous for Koshihikari, 10.3% homozygous for CH45, 6.0% heterozygous, and 0.4% missing data. Heterozygous regions remained in the BILs, but segregation without heterozygous regions fitted a ratio of 7/8 (homozygous for Koshihikari): 1/8 (homozygous for CH45), as expected in a BC2 population. We constructed a linkage map for all chromosomes, which covered a total genetic distance of 427.2 cM (Fig. 3). The genome coverage of the map was 94% (350.5 Mb/373.2 Mb of Nipponbare pseudomolecule, IPGSP-1.0), as estimated from the physical positions of markers at the distal end of each chromosome (Fig. 3).

Fig. 3

Genetic linkage map of 126 SSR markers and locations of QTLs for BS resistance and DTH detected in 190 BILs (Koshihikari*3/CH45). Positions (Mb) are indicated according to the Nipponbare pseudomolecule (IPGSP-1.0) on the left side of SSR markers. QTLs with one-LOD confidence intervals for BS resistance and DTH are represented as black ovals and a striped oval, respectively.

Four QTLs (qBSR2-kc, qBSR7-kc, qBSR9-kc, and qBSR11-kc) for BS resistance were identified on chromosomes 2, 7, 9, and 11 (Fig. 3, Table 1); qBSR11-kc had the highest logarithm of odds (LOD) score (12.6 in 2015 and 10.1 in 2016) and was considered as a major QTL. The alleles from CH45 explained 17.1% (qBSR2-kc), 7.8% (qBSR7-kc), and 23.0%–25.9% (qBSR11-kc) of the total phenotypic variation, whereas qBSR9-kc from Koshihikari accounted for 6.3%–6.5%. Two QTLs (qBSR9-kc and qBSR11-kc) were detected in both years, whereas the other two were detected only in 2016.

Table 1

Putative QTLs for BS resistance and DTH

Trait	Years	QTL	Chromosome	Marker interval1)	LOD score	Variance explained of total (%)	Additive effect2)
Brown spot resistance	2015	qBSR9-kc	9	RM3919-RM6797	3.7	6.5	0.2
		qBSR11-kc	11	RM6534-RM4112	12.6	25.9	−0.5
	2016	qBSR2-kc	2	RM5578-RM5614	5.8	17.1	−0.6
		qBSR7-kc	7	RM1353-1-RM5672	3.0	7.8	−0.4
		qBSR9-kc	9	RM3919-RM6797	3.3	6.3	0.3
		qBSR11-kc	11	RM6534-RM4112	10.1	23.0	−0.8
Days to heading	2015, 2016	qDTH4-kc	4	RM16739-RM5586	22.3	32.7	34.6

The nearest markers are underlined.

Negative values mean that the ‘CH45’ allele decreased the disease score.

As significant negative correlations between disease scores and DTH were found in both years, we also mapped the QTL for DTH. A single QTL associated with DTH (qDTH-kc4) was mapped on chromosome 4 (Fig. 3, Table 1). Thus, despite the correlation between the two traits, no DTH loci were identified on the same chromosomes as the disease resistance loci.

Discussion

No major genes conferring immunity to BS have been identified, but a few cultivars, such as CH45, have sufficiently high quantitative resistance that may be agriculturally useful (Eruotor 1985, Sato ). In the present study, we conducted QTL analysis for BS quantitative resistance with a BIL population derived from a cross between CH45 and Koshihikari.

A negative correlation between disease score and DTH was observed in the population, but the genes controlling disease scores and DTH appear to map on different chromosomes (Fig. 3). Therefore, BS resistance is not a pleiotropic effect of delayed heading.

We identified a total of four QTLs for BS resistance on chromosomes 2, 7, 9, and 11; each QTL explained 6.3% to 25.9% of the phenotypic variation (Fig. 3, Table 1). Except for qBSR9-kc, the resistant parent CH45 contributed the resistance alleles of these QTLs. Two QTLs (qBSR9-kc and qBSR11-kc) were stable in both years, whereas the other two (qBSR2-kc and qBSR7-kc) were detected only in 2016. Monthly average temperatures at MPARI from June to September 2016 were 1.0–2.3°C higher than in 2015, and heavy rainfalls were frequent in September 2016 (Japan Meteorological Agency 2016), which may have increased the disease severity in 2016 in comparison with 2015 and thus facilitated the detection of additional QTLs.

The BS-resistance QTLs on chromosomes 2, 9, and 11 were already reported (Katara , Sato , 2015). The marker intervals of these QTLs were similar to those of the QTLs detected in this study. To the best of our knowledge, qBSR7-kc on chromosome 7 is a novel QTL for BS resistance.

Our research group previously reported the QTLs qBS2, qBS9, and qBS11 (the latter is the same as the major QTL qBSfR11) for BS resistance in RILs derived from a cross between Tadukan (indica) and Hinohikari (japonica); the resistance alleles at qBS9 and qBS11 were provided by the indica parent and at qBS2 by the japonica parent (Sato , 2015). In the present study, we found that qBSR11-kc from the indica parent CH45 was a major BS resistance QTL; however, unlike in the previous study, the resistance allele at qBSR2-kc was from the indica parent and that at qBSR9-kc was from the japonica parent. The existence of different resistance alleles in the same QTL region may be important for the introduction of resistance QTLs from indica germplasms to materials with the japonica background through gene pyramiding in breeding programs. Now we are developping a program to breed practical japonica cultivars with strong BS resistance.