High-resolution mapping and characterization of xa42, a resistance gene against multiple Xanthomonas oryzae pv. oryzae races in rice (Oryza sativa L.).

Improvement of resistance against rice bacterial blight (BB) disease is an important breeding strategy in breeding programs across the world, especially in Africa and southern Asia where BB is more prevalent. This report describes a high-resolution map and characterization of xa42 at XA42 locus, a rice BB resistance gene in XM14, a mutant line originating from IR24. The candidate gene region was narrowed down from 582 kb, which had been obtained in our previous study, to 57 kb. XM14 shows brown spots in its leaves like lesion mimic mutants. This line also shows a shorter stature than the original cultivar IR24. In XA42 gene segregating populations, homozygotes of xa42 allele were consistently resistant to the six Japanese Xanthomonas oryzae pv. oryzae races used for this study. They also showed brown spots and markedly short stature compared with the other genotypes, suggesting that xa42 gene exhibits pleiotropic effects.

Introduction

Bacterial blight (BB) caused by Xanthomonas oryzae pv. oryzae (Xoo) is undoubtedly a severe and broadly distributed disease of cultivated rice (Oryza sativa and O. glaberrima). BB affects rice production in both rain-fed and irrigated environments (Furuya ). According to a review by Mew , reduction in rice yield might be at high as 50% in fields where the crop is infected severely. Infection at the tillering stage can engender total crop losses. More commonly, however, plants are affected at the maximum tillering stage. Yields are reduced by 10–20%. BB is prevalent in both tropical and temperate climates, and reportedly in all rice growing regions worldwide except North America (Niño-Liu , Ou 1985).

BB caused by Xoo has been ranked fourth among the ten most destructive bacterial plant pathogens (Mansfield ) because of the great losses it causes to rice production. Many control measures have been assessed and practiced to combat BB disease. They encompass biological control, chemical control, cultural control, and physical control, as reviewed by Ezuka and Kaku (2000). However, varietal resistance is regarded as the most effective and economical means of controlling BB (Khan , Verdier ).

About 40 known resistance genes (R genes) confer resistance to various strains of Xoo (http://www.shigen.nig.ac.jp/rice/oryzabase/locale/change?lang=en, Busungu , Khan , Xia ). These R genes have been identified in both cultivated rice cultivars and wild relatives of rice, and some from mutation induction. However, rapid changes in the pathogenicity of Xoo and the continuous evolution of pathogenic races are causing the breakdown of resistance in many improved varieties (Khan , Xia ). Consequently, novel BB R genes are still highly demanded for rice. More analyses of these R genes are needed to combat this destructive disease effectively in the future. A report of our previous study (Busungu ) describes the identification of a recessive R gene from the ‘XM14’ line. This gene was named and registered as xa42 at the XA42 locus according to the gene nomenclature system for rice (McCouch and CGSNL 2008). Linkage analysis showed that XA42 is located around centromeric regions of chromosome 3 between DNA markers KGC3_16.1 and RM15189 (Busungu ). To identify this gene at molecular level, one must narrow down the candidate chromosomal region. As reported from our previous study (Busungu ), XM14 exhibits resistance against the six Japanese Xoo races used. However, the genetic base of the resistance against multiple races remains unknown. The XM14 line shows brown spots in its leaves similarly to lesion mimic mutants (Fig. 1). This line also presents a shorter stature than the original cultivar IR24 (see below). No information indicates that these phenomena are caused by the pleiotropic effect of xa42 or not. This study was conducted to provide high-resolution mapping and characterization of the XA42 gene.

Fig. 1

Leaf appearance of parental lines (XM14 and IAS16) and F2, F3 plants from the cross between them. XA14 showed brown spots on its leaves. IAS14 showed normal leaves. In the F2 and F3 populations, both plants with normal leaves and those with brown spots on their leaves appeared.

Materials and Methods

Bacterial races, inoculation and scoring

Xoo have been classified into many races according to virulence and origin. The present study used six Japanese races for inoculation tests: race I (strain T7174), race IIA (strain T7147), race IIB (strain H9387), race III (strain T7133), race IV (strain H75373) and race V (strain H75304). To conduct the inoculation tests, each Xoo inoculum was prepared and cultured using potato semi-synthetic agar medium (Wakimoto 1954) and was incubated at 28°C for 48 hr. The inoculum was then diluted with distilled water. Subsequently, the absorbance was adjusted to A = 0.05 (620 nm) using a spectrophotometer. This value corresponds to the concentration of 108 colony-forming units per milliliter (cfu/ml), which is the optimum concentration required to cause BB disease. Xoo was inoculated using clipping method, a procedure described by Kauffman , during booting to the flowering stage. BB severity was scored by measuring the lesion length (LL) of the inoculated leaves of rice plants with measurement using a ruler 18 days after Xoo inoculation.

Plant materials

‘IR24’ is an elite Indica rice cultivar produced by the International Rice Research Institute. IR24 is highly susceptible to all Philippines and Japanese Xoo races (Busungu , Taura ). ‘XM14’ is a resistant mutant line produced by induction of MNU mutagen to IR24 (Busungu ). ‘IAS16’ is a chromosome segment substitution line carrying segments from a Japonica cultivar ‘Asominori’ with the background of IR24 (Kubo ). IAS16 carries the chromosome segment containing XA42 locus from Asominori. It is susceptible to the six Japanese Xoo races described above (Busungu ).

Progeny of the cross from the cross between XM14 and IAS16 was used for this study. All plant materials were grown and analyzed in the following manner, with some exceptional cases described in the next subsections. Germinated seeds were sown in seedling boxes in a greenhouse in May. About two weeks after sowing, seedlings were transferred out of the greenhouse. Then, after about three weeks, seedlings were transplanted to a paddy field in the experimental farm of the Faculty of Agriculture, Kagoshima University, Kagoshima, Japan. Fertilizers were applied one week after transplantation at the rate of 6 g of N, 3 g of K2O, 3 g of P2O5/m2. Two weeks before Xoo inoculation test, N was applied at a rate of 3 g/m2. The plant spacing was 15 × 30 cm.

High-resolution mapping of XA42

In 2015, about 1000 F2 plants and 20 plants from each parental line were subjected to high-resolution mapping of XA42. In 2016, 2950 F2 plants and 30 plants from each parental line were subjected to high-resolution mapping in the same manner as 2015. In addition to F2 plants, 17 F3 lines (30 plants per line with some exceptional cases) from recombinants obtained in 2015 were subjected to the progeny test for the genotype of XA42 locus. Xoo race IIA (strain T7147) was inoculated at the booting stage. Some F3 lines were also inoculated with the other five Xoo races for resistance against multiple races (see below). The LL of one representative leaf from each plant was measured after careful visual observation.

Test for resistance of xa42 against multiple Xoo races

In 2016, six F3 lines were selected so that XA42 gene segregation was expected and that recombination had occurred very close to XA42 gene (Table 2). About 150 plants per F3 line were transplanted to the paddy field. At the booting stage, the tillers of each plant from each line were separated into two halves (right and left) using polypropylene ropes. Different Xoo races were inoculated to each half of tillers. Each F3 line was divided into three sublines of approximately fifty plants each. One subline was inoculated with race I (strain T7174) and race IIA (strain T7147). Another subline was inoculated with race III (strain T7133) and race IV (strain H75373). The other subline was inoculated with race V (strain H75304) and race IIB (strain H9387). The LL of one representative leaf from bundle of half tillers of each plant was measured after careful visual observation.

Table 2

Genotypes of informative recombinants for the DNA marker loci linked with XA42 in the F2 population (XM14 × IAS16), brown spots, and reaction against Xoo Japanese race IIA (T7147) inoculation in the F2 and F3 generations

F2 Individualb	Year	Lesion length	Reactionc	Brown spotsd	Genotypes of the DNA marker locia																	No. of F3 plants Reaction R : Se
					KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	KGC3	RM
					16.1	16.180	16.209	16.255	16.278	16.3	16.342	16.370	16.371	16.399	16.407	16.421	16.514	16.594	16.613	16.636	15189
1	2015	18.8	S	N	A	A	A	A	A	H	H	H	H	H	H	H	H	H	H	H	H	7 : 22
2	2015	14.6	S	N	A	A	A	A	A	A	A	A	A	H	H	H	H	H	H	H	H	0 : 30
3	2015	19.4	S	N	H	H	H	H	A	A	A	A	A	A	A	A	A	A	A	A	A	0 : 30
4	2015	31.5	S	N	H	H	H	H	H	A	A	A	A	A	A	A	A	A	A	A	A	0 : 30
5	2015	18.3	S	N	H	H	H	H	H	H	A	A	A	A	A	A	A	A	A	A	A	0 : 30
6	2015	21.6	S	N	H	H	H	H	H	H	H	H	H	H	A	A	A	A	A	A	A	6 : 24
7	2015	0.1	R	B	H	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	30 : 0
8	2015	2.8	R	B	H	H	H	H	X	X	X	X	X	X	X	X	X	X	X	X	X	30 : 0
9	2015	12.7	S	N	H	H	H	H	H	H	H	H	H	X	X	X	X	X	X	X	X	31 : 128
10	2015	14.5	S	N	X	X	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	31 : 109
11	2015	22.1	S	N	H	H	H	H	H	H	H	H	H	H	H	X	X	X	X	X	X	42 : 117
12	2015	1.1	R	B	X	X	X	X	X	X	X	X	X	H	H	H	H	H	H	H	H	50 : 0
13	2015	0.3	R	B	X	X	X	X	X	X	X	X	X	H	H	H	H	H	H	H	H	30 : 0
14	2015	23.3	S	N	X	X	X	H	H	H	H	H	H	H	H	H	H	H	H	H	H	43 : 110
15	2015	18.5	S	N	H	H	H	H	H	H	H	H	H	H	H	H	H	X	X	X	X	42 : 117
16	2015	23.4	S	N	X	X	X	X	X	H	H	H	H	H	H	H	H	H	H	H	H	34 : 116
17	2015	18.2	S	N	X	X	X	X	X	X	X	H	H	H	H	H	H	H	H	H	H	5 : 24
18	2016	0.2	R	B	H	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	Not tested
19	2016	0.1	R	B	H	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	Not tested
20	2016	0.1	R	B	H	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	X	Not tested
21	2016	0.1	R	B	H	H	H	H	H	H	X	X	X	X	X	X	X	X	X	X	X	Not tested
22	2016	0.1	R	B	H	H	H	H	H	H	H	X	X	X	X	X	X	X	X	X	X	Not tested
23	2016	22.8	S	N	H	H	H	H	H	H	H	H	H	H	X	X	X	X	X	X	X	Not tested
24	2016	0.1	R	B	X	X	X	X	X	X	X	X	X	H	H	H	H	H	H	H	H	Not tested
25	2016	0.2	R	B	X	X	X	X	X	X	X	X	X	X	X	H	H	H	H	H	H	Not tested
26	2016	13.0	S	N	X	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	Not tested
27	2016	26.5	S	N	X	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	H	Not tested
28	2016	19.4	S	N	X	X	X	X	H	H	H	H	H	H	H	H	H	H	H	H	H	Not tested
29	2016	25.0	S	N	X	X	X	X	H	H	H	H	H	H	H	H	H	H	H	H	H	Not tested
30	2016	13.6	S	N	X	X	X	X	X	H	H	H	H	H	H	H	H	H	H	H	H	Not tested

X, H, and A respectively denote homozygotes for XM14, heterozygotes, and homozygotes for IAS16.

Progeny of Plant Nos. 9, 10, 11, 14, 15, and 16 were subjected to test for resistance of xa42 against multiple Xoo races.

R and S respectively denote resistant and susceptible.

B and N respectively denote plants showing brown spots on their leaves and those showing normal leaves.

LL shorter than 3.0 cm were scored as R, whereas those longer than 5.0 cm were scored as S. As for the materials for resistance against multiple races, reactions to multiple races were summed up (see Fig. 5).

Test for pleiotropic effects of xa42 gene on brown spots and agronomic traits

The XM14 line exhibits brown spots in its leaves (Fig. 1). In all, 982 F2 plants for xa42 mapping in 2015, 2950 F2 plants for xa42 mapping and the six F3 lines for test for resistance of xa42 against multiple Xoo races in 2016 were evaluated for their brown spots immediately before Xoo inoculation test because the presence or absence of brown spots on leaves was most clearly recognizable at the stage. The ‘B’ score was given to plants with brown spots on their leaves; the ‘N’ score was given to those with normal leaves. At the maturity stage in 2016, three agronomic traits, culm length, plant height and number of tillers, were evaluated in F2 generation (250 plants), F3 generation (297 plants), parental lines (20 plants per line), and IR24 (20 plants per line). The F2 plants in this test were selected randomly in 2950 F2 plants. The F3 came from the progeny of recombinant plant Nos. 10 and 14 (Table 2), two of the six F3 lines used for multiple Xoo races test of xa42 gene. Statistical analyses were conducted using software (SPSS statistics 23; IBM Inc. New York, USA).

Molecular techniques

Two weeks before inoculation tests, leaf samples from each F2 and F3 plant were collected for DNA extraction. DNA was extracted from 50 mg leaf of each plant in segregating populations according to the method described by Dellaporta with some modifications. The polymerase chain reaction (PCR) mixture (6 μL) consisted of 10 ng genomic DNA, 200 μM dNTPs, 0.2 μM of each primer, 0.25 U of Taq polymerase (AmpliTaq Gold; Applied BioSystems, CA, USA), and 10× buffer containing MgCl2. PCR was performed under the following conditions: 4 min at 94°C for initial denaturation, followed by 35 cycles of 94°C for 30 s, annealing temperature for 30 s, 72°C for 30 s, with final extension at 72°C for 7 min. The PCR products were separated in 10% (29 : 1) polyacrylamide gels, stained with ethidium bromide, and visualized using ultraviolet light (GelDoc-It® TS Imaging System; UVP, CA, USA). Before electrophoresis, the PCR products of CAPS and dCAPS markers were digested using the respective restriction enzymes.

DNA markers/restriction enzyme

For this study, we used 19 DNA markers (Table 1), 3 of which have already been described by Busungu . The other 16 markers, that is, 1 SSR, 2 dCAPS, 2 CAPS markers and 11 Insertion/deletion (Indel), were designed according to procedure elaborated below. In Busungu , XA42 gene was located between DNA markers KGC3_16.1 and RM15189. Apparently, a single sequence repeat (SSR) located between them was available. We checked for uniqueness of the DNA sequences surrounding the SSR using BLAST with BLASTScope in Oryzabase (Yamazaki , http://www.shigen.nig.ac.jp/rice/oryzabaseV4/blast/search). Primer pairs were designed automatically using software (PRIMER 3 ver. 4; http://bioinfo.ut.ee/primer3-0.4.0/primer3/, Untergasser ). We adopted the following primer design parameters: Primers were 20–35 nucleotides long, with optimum set at 25. Primer Tm was 55–65°C with optimum set at 60°C. The maximum Tm difference was 2°C. Primer GC% contents of 20–80.

Table 1

Primer sequences of DNA markers designed in high-resolution mapping of XA42 gene

Marker name	Kind of DNA marker	Restriction enzyme	Primer sequence	Location on IRGSP 1.0 pseudomolecule chromosome 3
				From	to	Source
KGC3_16.1	Indel		F GTTTAGATATCGCTTTCAGGCATGT	16117085	16117109	Busungu et al. 2016
			R CGGTTTATAAGGGTAGCCGC	16117216	16117235
KGC3_16.180	Indel		F TGTTTCACATCGGGGACTTGAATTT	16180443	16180467	This study
			R GACAAACGGGACAAGGCTAAATTAT	16180526	16180550
KGC3_16.209	Indel		F ACATAACGTGGCACCAAAACTAG	16208945	16208967	This study
			R GCTCGATTTATAGGGTGCAAAATCT	16209086	16209110
KGC3_16.255	Indel		F TTCTAGGTCGATCGGTGCATTTATG	16255712	16255736	This study
			R CGTCTGGTAATTGTGAAACATGACT	16255856	16255880
KGC3_16.278	Indel		F TAATTGGCGCTGAGATATGTCCGAT	16278421	16278445	This study
			R GAATCGAACCTGGACCTTTTACTTG	16278578	16278602
KGC3_16.3	Indel		F ATTAGAGTATCCACCAATAAGCCCG	16323299	16323323	Busungu et al. 2016
			R GAGGTAAGATGAGATCGTGTAGGAG	16323522	16323546
KGC3_16.342	CAPS	Hsp92	F GGTTATTTTCTTAAACCCGCTATTGG	16342875	16342900	This study
			R ATAAATAGTGTATAGCAGCGGGTTG	16342934	16342958
KGC3_16.370	Indel		F GATGCCATCCTTCTTCACCCTT	16370161	16370182	This study
			R GGTAGAAGTAGAGCCATGGAACTTG	16370289	16370313
KGC3_16.371	Indel		F GATAATGCGAAAGAGAATCAAGGGG	16371045	16371069	This study
			R AGAAGCACGAGTTTAATAAGGCCTA	16371279	16371303
KGC3_16.399	dCAPS	BseG I	F ATTTTACCAATTATAGTTTTCGTCCGGAT	16399678	16399703	This study
			R GCTTTACCTGGCATATTCTAGACCT	16399803	16399827
KGC3_16.407	SSR		F CTCGTCTCCATCAAGTTTTCAGGTC	16407423	16407447	This study
			R TAGTGTACCATGATTTGCAAGCCTT	16407536	16407560
KGC3_16.421	dCAPS	AluI	F CCCTCCGTACAAGTACAATACATAGC	16421768	16421790	This study
			R CCCTCTCCAAGTAAATCCATGTCTT	16421914	16421938
KGC3_16.422	CAPS	BssSI	F AGTGCATATTCTTTCGTCCAACTTT	16422886	16422910	This study
			R TTAACCTTCCTTTACTATCGGTGTC	16423007	16423031
KGC3_16.514	Indel		F GCTTTATATCTGTCAACTGGATTAGAATTC	16514854	16514883	This study
			R AGGTGAGTATAATAAGCAAGTTGAGT	16514952	16514977
KGC3_16.552	CAPS	HaeIII	F GGTACAATATAACAGTTCCACCAAGA	16552916	16552941	This study
			R GTTTGGAATTGACTGATTAGCCACA	16553069	16553093
KGC3_16.594	Indel		F GTTTTGATAGAGCGCAATTTGTCAT	16594790	16594814	This study
			R ATCCCAAGCTGCCAGTATAAATTAA	16594863	16594887
KGC3_16.613	Indel		F TGCCTGTAAAAGTTCTTGATGGAAT	16613174	16613198	This study
			R TTTAGCTTCACAGTGTAAAAGTTAG	16613383	16613407
KGC3_16.636	Indel		F CTTATTGGTTGGCGTGGTGTTT	16636613	16636634	This study
			R AGAAGCACGAGTTTAATAAGGCCTA	16636819	16636845
RM15189	SSR		F CAGTAAGTGTCTCTGGAAGCTTG	16699297	16699319	IRGSP 2005, Redisigned in Busungu et al. 2016
			R TGCTGAGTAGGTACCTTTCTTAAAAC	16699440	16699465

For Indel markers, we used Indel information released by Xu or looked for Indel polymorphism (5–50 bp difference) between a Japonica cultivar ‘Nipponbare’ and an Indica cultivar ‘93-11’ and/or an Indica cultivar HR12. Oryza sativa (rice) Nucleotide BLAST (http://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch&PROG_DEF=blastn&BLAST_PROG_DEF=megaBlast&BLAST_SPEC=OGP__4530__9512) was used for Indel information. Blastn search optimized for highly similar sequences was applied using one thousand to ten thousand base Nipponbare sequence (Os-Nipponbare-Reference-IRGSP-1.0) as query and 93-11 sequence (GCA_0000046551) or HR12 sequence (GCA_000725085) as subject. Uniqueness of the DNA sequences surrounding Indel was confirmed using BLAST with BLASTScope in Oryzabase (Yamazaki ).

As for CAPS and dCAPS markers, we used SNP information from Huang and the Rice SNP seeker database (Alexandrov ). Using various SNP datasets available in a single interface (Mansueto ), we were able to detect SNP which can distinguish Indica (IR24) and Japonica (Asominori). We confirmed the SNP information using the genome information of five rice cultivars: three Japonica cultivars, Nipponbare, Hitomebore (GCA_000321445.1), and Koshihikari (GCA_000164945.1), and two Indica cultivars, 93-11 and HR12 (https://blast.ncbi.nlm.nih.gov/Blast.cgi?PAGE_TYPE=BlastSearch&BLAST_SPEC=OGP__4530__9512). The DNA sequences surrounding the SNP were checked for uniqueness using BLAST with BLASTScope in Oryzabase (Yamazaki ). DNASIS Pro software (ver. 2.10; Hitachi Software Engineering Co. Ltd., Tokyo, Japan) was used to detect restriction enzymes that can recognize SNP.

Primers surrounding Indel and SNP were designed using Primer 3 (Untergrasser ) with the parameters above. When SNPs were not detected by restriction enzymes, dCAPS markers were designed using dCAPS finder software (Neff ).

Mapping and gene annotation

F2 plant recombinants were used to determine the exact position of the nearest recombination event to the target region on chromosome 3 centromeric region. The high-resolution mapping of the target XA42 gene was constructed according to the physical distance of the DNA markers used in the study (Table 1). In addition to F2 plants, the progeny test of all recombinants obtained in 2015 was applied to confirm the recombination events, segregation or non-segregation reaction against Xoo. Candidate genes of XA42 were sought within the candidate chromosomal region using ‘the rice annotation project database (RAP-DB)’ (http://rapdb.dna.affrc.go.jp/) (Kawahara , Sakai ) and ‘rice genome annotation project’ () (Kawahara ).

Results

High-resolution mapping of XA42 gene

In 2015, 982 F2 plants from the cross between IAS16 and XM14 were evaluated for LL inoculated with Xoo race IIA (strain T7147). The distribution of LL showed a bimodal distribution (Fig. 2). The dividing point was set at 5 cm LL because of the clear gap of LL at 5 cm. LL shorter than 5 cm was regarded as resistant, whereas LL longer than 5 cm was regarded as susceptible to BB. Based on that dividing point, the resistant and susceptible plants were, respectively, 288 and 694. In 2016, the 2950 F2 plants from the same cross combination were evaluated for LL inoculated with Xoo race IIA (strain T7147). The segregation indicated a clear bimodal distribution of LL (Fig. 3). We observed a clear gap of LL around 4 cm and classified the 2950 F2 plants into 885 resistant plants with LL shorter than 3 cm and 2065 susceptible plants with LL longer than 5 cm. The ratio in 2015 and 2016 both deviated from 1 resistant: 3 susceptible, one-gene segregation (2015: χ2 = 9.810, P < 0.01, 2016: χ2 = 39.333, P < 0.001). However, the tight linkage between the resistance and DNA markers KGC3_16.370 (see below) confirmed one-gene segregation. Our earlier report (Busungu ) described that XA42 was located between DNA markers KGC3_16.1 and RM15189. From initial recombinant screening using these two markers in 2015, a total of 17 recombinants between them were selected. Non-recombinants between them were treated as non-recombinants among all inside markers because the genetic distance between the two markers was only 0.7 centiMorgan. To draw a high-resolution map of XA42 gene, 16 new polymorphic DNA markers were developed. Reaction to Xoo and DNA marker genotypes of Plants 7–17 in Table 2 all suggested that XA42 was located between KGC3 16.1 and KGC3_16.594. Therefore, in 2016, the DNA marker combination of KGC3_16.1 and KGC3_16.594 was used for recombinant screening. In all, 29 recombinants between them were selected. Then they were subjected to genotyping of internal DNA markers (Table 1). In addition to F2 tests, we performed F3 test of 17 recombinants. The results of genotyping of informative F2 plants and their F3 tests facilitated detailed mapping of the XA42 gene (Table 2).

Fig. 2

Lesion length distribution of 982 F2 plants (planted in 2015) from the cross between XM14 and IAS16 after Xoo Japanese race IIA (strain T7147) inoculation. Three classified genotypes were assessed for KGC3_16.370 as indicated: black, homozygotes for XM14; gray, heterozygotes; dark gray, homozygotes for IAS16. Horizontal lines at the top of the figure show the ranges of parental lines. Vertical lines crossing the horizontal line show means of the parental lines.

Fig. 3

Lesion length distribution of 2950 F2 plants (planted in 2016) from the cross between XM14 and IAS16 after Xoo Japanese race IIA (strain T7147) inoculation. Three classified genotypes were assessed for KGC3_16.370 as indicated: black, homozygotes for XM14; gray, heterozygotes; dark gray, homozygotes for IAS16. Horizontal lines at the top of the figure show the ranges of parental lines. Vertical lines crossing the horizontal line show means of the parental lines.

Genetic information from each informative recombinant was mutually consistent (Table 2). The discontinuous LL distribution like Fig. 3 was observed in F3 lines in Table 2 in 2016: LL shorter than 3.0 cm were scored as resistant (R), whereas those longer than 5.0 cm were scored as susceptible (S). Important recombination events occurred between DNA markers KGC3_16.3 and KGC3_16.407. Plant Nos. 12 and 13 both show that the XA42 gene is located on the left side of DNA marker KGC3_16.399 because the F3 lines of the two plants were fixed for resistant plants with brown spots. Plant No. 17 showed that the gene was located on the right side of DNA marker KGC3_16.341 because the F3 line of the plant is composed of 5 resistant plants with brown spots and 24 susceptible plants with no brown spots. The result for Plant No. 22 was consistent with that, although its progeny test was not performed. Therefore, XA42 gene was located in the 57 kb-chromosomal region between DNA markers KGC3_16.341 and KGC3_16.399. Results for other plants in Table 2 all support this idea. A linkage map comparing results obtained from our previous study (Busungu ) and this study is portrayed in Fig. 4.

Fig. 4

Location of XA42. A, comparative linkage map of XA42 gene between RFLP framework map of chromosome 3 modified from Harushima and the linkage map of XA42 gene constructed from F2 population from XM14 and IAS16 (n = 194). Modified from Busungu . B. Candidate chromosomal region of XA42 on Os-Nipponbare-Reference-IRGSP-1.0, based on results of high resolution mapping obtained from this study.

The search for candidate gene using RAP-DB (http://rapdb.dna.affrc.go.jp/) found five ORFs in the candidate region of XA42 (Table 3). The rice genome annotation project (rice.plantbiology.msu.edu/) found seven ORFs, three of which might correspond to those found in RAP-DB, and two of which encoded retrotransposon proteins (data not shown). The other two are LOC_Os03g28430 described as encoding hypothetical protein and LOC_Os03g28420 described as 3-oxoacyl-synthase. Its first exon corresponds to Os03g402200.

Table 3

Annotation data by RAP-DB of putative ORFs in the candidate chromosomal region of XA42

ORF in RAP-DB	Location on IRGSP 1.0 pseudomolecule of chromosome 3	Description in RAP-DB	Predicted length (bp)
Os03g0401951	16358834-16359470 (+ strand)	Hypothetical gene	637
Os03g0402000	16359486-16362281 (− strand)	TRAPP I complex, Bet3 domain containing protein	993
Os03g0402200	16379079-16378724 (− strand)	Hypothetical protein	274
Os03g0402400	16384695-16386794 (− strand)	Similar to ribosome-associated protein p40-like	1146
Os03g0402600	16397246-16397680 (− strand)	Predicted gene	435

Resistance of xa42 to multiple Xoo races

Inoculation of Xoo to parental lines showed that IAS16 line was susceptible, whereas the XM14 line was resistant to the six Japanese Xoo races used for this study. All F3 lines showed segregation in the reaction against the six races (Fig. 5). For this study, we inoculated two races to one plant. The reaction against one race was almost identical to that against the other race. When a plant showed short LL against one race, it showed short LL against the other race. A plant showing long LL against one race showed long LL against the other race. The parental F2 plants of F3 lines used in this experiment were all heterozygous at KGC3_16.370 locus, which proved to co-segregate with XA42 in the previous subsection. Homozygotes of XM14 allele at KGC3_16.370 locus showed very short LL (shorter than 1 cm in most cases), indicating that they were resistant against inoculated Xoo races. Heterozygotes and homozygotes of IAS16 allele at the locus showed long LL (longer than 5 cm in most cases), indicating that they were susceptible to inoculated Xoo races (Fig. 5). When the six F3 lines were summed up, the segregating chromosomal region around XA42 gene was limited to 121 kb (Tables 1, 2).

Fig. 5

Lesion length distribution of the F3 plants from the cross between XM14 and IAS16 after six Japanese Xoo races. Each F3 line, the progeny of recombinants in Table 2, was divided into three sublines comprising ca. 50 plants. Sublines were inoculated with two Xoo races shown along axes of subfigures. X, solid circle and open triangle respectively denote homozygotes for XM14, heterozygotes and homozygotes of IAS16 at the KGC3_16.370 locus. Dotted lines denote the dividing point between resistant and susceptible plants.

Relation between resistance to Xoo and brown spots

In 2015, 982 F2 plants were examined for the presence of brown spots on their leaves. All 288 resistant plants were found to exhibit brown spots, although the 694 susceptible plants showed no signs of brown spots. In 2016, we used 2950 F2 and 920 F3 plants for the presence of brown spots (Tables 4, 5). Results showed that all the F2 and F3 plants which had brown spots consistently exhibited a resistant reaction to Xoo races used for this study. Genetic analysis at DNA marker KGC3_16.370 locus showed that these plants were homozygotes of XM14 allele. Therefore, they proved to be homozygotes of xa42 allele. In contrast, all the F2 and F3 with normal leaves exhibited susceptible reactions to the Xoo races used for this study and were either homozygotes of IAS16 allele or heterozygote at KGC3_16.370 locus. These results suggest strongly that brown spots are caused by xa42.

Table 4

Relations between genotypes at KGC_16.370 locus, brown spots and reaction against Xoo Japanese race IIA (strain T7147) inoculation in the F2 population from the cross between IAS16 and XM14

Year	Genotype at KGC3_16.370 locus	Reaction against Xoo race IIA		Brown spots
		Resistant	Susceptible	Ba	N
2015	Homozygote for XM14 allele	288	0	288	0
	Heterozygote	0	506	0	506
	Homozygote for IAS14 allele	0	188	0	188
2016	Homozygote for XM14 allele	885	0	885	0
	Heterozygote	0	1529	0	1529
	Homozygote for IAS14 allele	0	536	0	536

B and N respectively denote plants showing brown spots on their leaves and those showing normal leaves.

Table 5

Relations between brown spots and genotypes at KGC_16.370 locus in the F3 lines used for resistance against multiple Xoo strains

Genotype at KGC3_16.370	Brown spots in F3 lines aderived from recombinant plant No.
	9		10		11		14		15		16
	B	N	B	N	B	N	B	N	B	N	B	N
Homozygote for XM14 allele	31		31		42		43		42		34
Heterozygote		80		73		80		80		87		79
Homozygote for IAS16 allele		48		36		37		30		30		37
Total	159		140		159		153		159		150

B and N respectively denote plants showing brown spots on their leaves and those showing normal leaves.

Pleiotropic effect of xa42 gene on agronomic traits

Table 6 presents the culm length, plant height and number of tillers of parental lines (XM14, IR24, IAS16), the F2 population and sum of the two F3 lines derived from the cross between XM14 and IAS16. Analysis of variance (ANOVA) revealed significant difference among parental lines, and among genotypes at KGC3_16.370 locus in the segregating populations for the entire three traits (data not shown). According to the multiple mean comparisons, XM14 showed significantly smaller value than IAS16 and IR24 for culm length and plant height. In the segregating population, homozygotes of XM14 showed a much smaller value than heterozygotes and homozygotes of IAS16 allele for culm length and plant height. We compared homozygotes of XM14 allele and the combination of the other genotypes, which corresponded respectively to resistant plants and susceptible plants in F2 and F3 populations, by applying t-tests: resistant plants showed significantly smaller values than susceptible plants for all three traits (data not shown).

Table 6

Tukey HSD mean comparisons of culm length (cm), plant height (cm) and the number of tillers of parental lines (XM14, IR24, and IAS16), and F2 and segregating F3 lines at KGC3_16.370 locus derived from the cross between XM14 and IAS16

Group	Subgroup	Number of plants	Culm length (cm)	Plant height (cm)	Number of tillers
Parental lines
	XM14	20	60.5 aa	79.1 a	9.5 a
	IR24	20	63.1 b	84.0 b	10.6 ab
	IAS16	20	64.2 b	86.2 b	11.6 b
F2
	Homozygote for XM14 allele	73	59.7 a	81.3 a	10.1 a
	Homozygote for IAS16 allele	45	67.5 b	91.0 b	12.3 ab
	Heterozygote	132	69.5 b	92.6 b	14.0 b
F3
	Homozygote for XM14 allele	74	60.7 a	81.2 a	10.2 a
	Homozygote for IAS16 allele	66	65.5 b	86.0 b	11.6 ab
	Heterozygote	153	67.2 b	88.0 b	13.0 b

Values followed by the same letter in each trait in the same group are not significantly different at P = 0.05 according to Tukey’s HSD test.

Discussion

In this study, we first performed high-resolution mapping of XA42, narrowing the area of interest to 57 kb located between the two DNA markers, KGC3_16.342 and KGC3_16.399. This gene cosegregated with two DNA markers, KGC3_16.370 and KGC3_16.371. Five candidate genes were found using RAP-DB. With the aid of this DNA-marker assisted analysis, the recessive allele xa42 was thought to be resistant to the entire Japanese Xoo races used for this study, and to cause brown spots on leaves and shorter stature.

The XA42 gene candidate region was narrowed down from 582 kb (between KGC3_16.1 and RM15189) (Busungu ) to 57 kb (KGC3_16.342 and KGC3_16.399) (Table 2). The high-resolution map depicted in this study (Fig. 4) established the foundation for XA42 gene map-based cloning and function characterization. As discussed in our previous study (Busungu ), based on IRGSP 1.0 genome and RAP-DB, the candidate chromosomal region of XA42 does not contain genes encoded by three isolated recessive BB R genes: xa5, xa13 and xa25. No isolated plant disease R gene encodes 3-oxoacyl-synthase or TRAPP I complex, Bet3 domain containing protein, or ribosome-associated protein (p40-like), which was annotated in XA42 candidate region in the present study (Table 3). Similarly, no isolated plant lesion mimic mutant (see below) gene encoded these proteins (for a review, Bruggeman ). These data suggest that xa42 is a new R gene. However, we downloaded the chromosome 3 sequences of nine cultivars (‘Nipponbare’ (IRGSP 1.0, NC_029258.1), ‘Nipponbare’ (assembled by Beijing Genomics Institute, CM000140.1), ‘HEG4’ (CM003066.1), ‘A123’ (CM003078.1), ‘Hitomebore’ (DG000055.1), ‘Koshihikari’ (DG000027.1), ‘RP Bio-226’ (CPO12611.1), ‘93-11’ (CM000128.1), ‘IR8’ (CM007598.1), ‘Kasalath’ (no accession name, downloaded from http://rapdb.dna.affrc.go.jp/download/irgsp1.html)), and found that the DNA length surrounded by the two DNA markers KGC3_16.342 and KGC3_16.399 ranged from 57 kbp (NC_029258.1) to 143 kbp (CM003066.1). Both NC_029258.1 and CM000140.1 are Nipponbare chromosome 3 sequences, but the XA42 candidate region length of CM000140.1 is 132 kbp. Our preliminary analysis of these genome alignments suggests that most of the sequences not covered by IRGSP 1.0 (NC_029258.1) are repetitive sequences scattered in rice genome (data not shown). However, some of these sequences could be unique, and contain genes not found in IRGSP 1.0 (data not shown). These data suggest that there might have been other candidate genes for XA42 than listed in Table 3. To identify XA42 gene, DNA sequencing of the candidate XA42 region of both XM14 and IR24 is necessary.

Because the probability of identifying Xoo resistant mutant is small (Taura ), the existence of simultaneous plural resistance mutations on one mutant line seems improbable. Therefore, these results strongly support the idea that xa42 is resistant to the six Japanese Xoo races. Suzuki induced mutation in a Japonica cultivar Taichung 65 using 1 mM of MNU, the same method of obtaining XM14 (Busung ), and estimated the mutation frequency as 3.2 × 103 nucleotide changes in a 4.3 × 108 rice genome corresponding to one mutation in every 135 kb. If the estimate is applied to our experimental results and the above rice genome information, there is high possibility that only one mutation occurred in the XA42 candidate chromosomal region. Therefore, these findings strongly support the idea that one mutation in XA42 locus induced resistance to the six Japanese Xoo races, brown spots on leaves and shorter stature.

Results show a strong and significant correlation between BB resistance and brown spots, which suggests that xa42 exhibits a pleiotropic effect. Brown spot mutants are generally called spotted leaf (spl) in rice. They are regarded as lesion mimic mutants. Yin reported that four mutants (spl1, spl5, spl9, and spl11) show enhanced resistance to blast, and that spl11 shows resistance to four Philippine Xoo races. Lesion mimic mutants have been studied extensively in light of programmed cell death leading to resistance to pathogens in many plants such as wheat (Li and Bai 2009) and Arabidopsis, as reviewed by Lorrain . Therefore, xa42 might be a kind of lesion mimic mutant. However, the brown spots on its leaves are not as dense and thick as those of typical spl mutants (Fig. 1). According to Oryzabase, two spl genes, SPL3 and SPL30, were registered as located on chromosome 3. A comparative linkage map of chromosome 3 (https://shigen.nig.ac.jp/rice/oryzabase/marker/mapCirn/3) based on findings reported by Harushima , Tsunematsu , and Yoshimura shows that SPL3 is located on the short arm, not close to the centromeric region. Therefore, SPL3 and XA42 differ. SPL30 was found to cosegregate with a DNA marker RM15380 in the segregating population comprising 2890 plants (Huang ). RM15380 is located in 18,632 kb region in Rice_IRGSP_Ver1.0, which is more than 2000 kb from the XA42 candidate region (Fig. 4). Therefore, XA42 differs from two reported spl genes on chromosome 3. Of about 40 reported rice BB resistance genes (Busungu , Khan ), only Xa3 gene was reported to exhibit brown spots resembling disease symptoms (Kaku and Hori 1977). Cultivars carrying the Xa3 resistance gene develop the brown spots as a necrotic resistant response after inoculating the plant with Xoo inoculums (Kaku and Hori 1977, Kaku and Ogawa 2001). The positive relation between brown spots and BB resistance reported in Xa3 seems similar to our observation in xa42. However, the brown spots in Xa3 appear after inoculation in contrast to those of xa42 with brown spots appearing even without inoculation of Xoo.

The pleiotropic effect of xa42 on agronomic traits can sometimes be negative in rice production when xa42 is used for breeding rice cultivars with multiple Xoo resistance. Therefore, the combination of xa42 and genes masking the negative effect of xa42, for example, allele conditioning tall stature on QTL controlling plant height, represents a possible solution.

XA42 gene segregation in F2 generation was distorted. However, when reactions to multiple strains were summed up for each F3 line, the ratio of resistant: susceptible all fitted to 1 : 3 (0.10 < P < 0.70 for χ2 (1 : 3)), was expected from one-gene segregation (Table 2, the progeny of Plant Nos. 9, 10, 11, 14, 15, and 16). According to Fukuta , segregation of chromosomal regions of both the short arm and long arm on chromosome 3 were skewed in favor of Indica allele. Some possibility exists that genes on one arm are insufficient to distort segregation, and that the F3 lines were fixed for one arm, not expressing segregation distortion.

Repeated R gene failure and breakdown upon pressure from new strains and favorable environment to Xoo have been reported many times, as reviewed by Khan . Planting of rice varieties with broad-spectrum disease resistance is the most sustainable strategy to protect rice from diseases and to ensure stable rice production. Results from this study suggest strongly that XM14 line, which has xa42, falls in the category of race-non-specific or broad-spectrum resistance. To prove that xa42 is truly a broad spectrum R gene, it should be tested with international Xoo races, especially those from south Asian and African countries where putative new Xoo races have been reported (Gonzalez , Mishra , Verdier ). If proven, the xa42 gene will be very useful in resistance breeding programs.