Fine mapping of the clubroot resistance gene CRb and development of a useful selectable marker in Brassica rapa.

In Chinese cabbage (Brassica rapa), the clubroot resistance (CR) gene CRb is effective against Plasmodiophora brassicae isolate No. 14, which is classified as pathotype group 3. Although markers linked to CRb have been reported, an accurate position in the genome and the gene structure are unknown. To determine the genomic location and estimate the structure of CRb, we developed 28 markers (average distance, 20.4 kb) around CRb and constructed a high-density partial map. The precise position of CRb was determined by using a population of 2,032 F2 plants generated by selfing B. rapa 'CR Shinki.' We determined that CRb is located in the 140-kb genomic region between markers KB59N07 and B1005 and found candidate resistance genes. Among other CR genes on chromosome R3, a genotype of CRa closest marker clearly matched those of CRb and Crr3 did not confer resistance to isolate No. 14. Based on the genotypes of 11 markers developed near CRb and resistance to isolate No. 14, 82 of 108 cultivars showed a strong correlation between genotypes and phenotypes. The results of this study will be useful for isolating CRb and breeding cultivars with resistance to pathotype group 3 by introducing CRb into susceptible cultivars through marker-assisted selection.

Introduction

Clubroot disease, caused by the soilborne plant pathogen Plasmodiophora brassicae and is one of the most serious diseases of Chinese cabbage (Brassica rapa) and other Brassica crops worldwide. Infected roots develop galls and cannot take up water and nutrients and plants become stunted and wilt, with subsequent losses of quality and yield. It is difficult to prevent infection through the use of agrochemicals and crop rotation because resting spores survive for several years in the soil (Tsushima , Voorrips 1995). Instead, clubroot-resistant (CR) cultivars are used to minimize losses. The CR genes from European fodder turnip (B. rapa) cultivars such as ‘Gelria R’, ‘Siloga’, ‘Debra’ and ‘Milan White’ have been introduced into Chinese cabbage (Yoshikawa 1981) and many CR cultivars have been bred. However, resistance conferred by a single gene can be overcome. In Japan, at least four pathotypes (groups 1 to 4) have been identified through the use of CR F1 cultivars of Chinese cabbage (Hatakeyama , Kuginuki ).

Eight CR loci—Crr1, Crr2, Crr3, Crr4, CRa, CRb, CRc and CRk—have been identified in B. rapa (Hirai , Matsumoto , Piao , Sakamoto , Suwabe ). We have identified Crr1, Crr2 and Crr4 from CR turnip ‘Siloga’ (Suwabe ) and a line with Crr1, Crr2 and Crr4 showed resistance to isolates of pathotype groups 1, 2 and 4, but not group 3 (unpublished). Recently, we demonstrated that CRb could confer resistance to an isolate in pathotype group 3 and could compensate for the race specificity of Crr1 and Crr2 (Kato ). CRb, which is derived from the European fodder turnip ‘Gelria R’ (Diederichsen , Hirai 2006), was originally identified on B. rapa chromosome R3 through genetic analysis using a resistant doubled haploid line derived from B. rapa ‘CR Shinki’ (Piao , 2009), and this locus functions as a strong dominant gene (Kato ). CRa, CRc and CRk were examined using six isolates by Matsumoto : CRa showed resistance to M85, which falls between groups 3 and 4, but CRc did not. CRk showed broad-spectrum resistance to most of the isolates, but heterozygous at CRk was not sufficient for resistance to M85. It was unknown whether Crr3 conferred resistance to group 3. Therefore, at this time, the usefulness of CR conferred resistance to group 3 in reported 8 CR loci were limited to CRa and CRb. In addition, CRb might be allelic or closely linked to CRa (Diederichsen ).

Although we mapped CRb between simple sequence repeat (SSR) markers KBrH059N21F and KBrH129J18R using the CR F1 cultivar ‘Akiriso’ and ‘CR Shinki’ (Kato ), there was a discrepancy between the results of Piao and ours in the position of marker TCR05 and the CRb locus. To resolve this discrepancy and develop useful markers closely linked to CRb, fine mapping of CRb is needed.

Improvement of selective markers linked to these CR loci would provide effective tools for developing cultivars with high resistance to a broad spectrum of races through the pyramiding of CR genes. Information on the type of resistance gene carried by each cultivar is also needed. To discriminate resistant cultivars carrying CRb and to introduce CRb into susceptible cultivars by the use of marker-assisted selection (MAS), we evaluated the relationship between genotype and resistance (phenotypic disease index) to isolate No. 14, using more than 100 cultivars.

In this study we describe (1) the determination of the CRb genomic region by fine mapping analysis with developed markers on the basis of recently available B. rapa A genome sequence information (Wang ); (2) the positional relationships between CRb and CRa.; and (3) the development of useful markers near CRb for introducing CRb into susceptible cultivars. We also discuss the candidate gene for CRb.

Materials and Methods

Plant materials

An F1 CR Chinese cabbage, ‘CR Shinki’ (Takii & Co. Ltd., Kyoto, Japan), was used as a donor of CRb derived from the European fodder turnip ‘Gelria R’ (Piao , 2009). Although we reported that the resistance of ‘CR Shinki’ was controlled by a single locus, since we cannot eliminate the possibility that CRa and CRb are located tandemly, it is unclear whether ‘CR Shinki’ carries CRa. An F2 population (n = 2,032) obtained by selfing ‘CR Shinki’ was used for fine mapping analysis. To examine the positional relationship between CRa and CRb, we also used another 172 F2 plants of ‘CR Shinki’ from our previous study (Kato ), the total number of F2 plants used in combined analysis was 2,204.

F2 plants of F1 CR turnip (B. rapa) ‘CR Shinano’, which carries Crr3 (Hirai , Saito ), were used to test whether Crr3 confers resistance to P. brassicae field isolate No. 14.

To evaluate the relationship between marker genotypes and resistance in Chinese cabbage accessions and to find useful selective markers for introducing CRb into susceptible lines, we tested 105 F1 CR or non-CR cultivars of Chinese cabbage and turnip and three European fodder turnips (‘Gelria R’, ‘Siloga’ and ‘Debra’) (Supplemental Table 1).

Pathogen and the CR test

Plasmodiophora brassicae field isolate No. 14, which is classified in pathotype group 3 (Hatakeyama ), was used for the CR test, which was performed as described by Suwabe . The disease index (DI) was scored according to Hatakeyama on a scale of 0 to 3: 0, no symptoms; 1, a few small, separate globular clubs on the lateral roots; 2, moderate clubbing and swelling of lateral roots and 3, absence of normal roots, presence of severe clubs on main and lateral roots. Plants with DI scores of 0 to 2 were categorized as resistant (0 is full resistance, 1 and 2 show partial resistance); those with a score of 3 were categorized as susceptible.

DI scores of each of the 108 cultivars of Chinese cabbage and turnip were calculated from the mean DI of eight seedlings. The resistance profiles of 12 F1 cultivars and two European fodder turnips examined previously by Hatakeyama were used as reference data (Table 2). The test was performed two or three times.

Table 2

Relationship between genotypes of markers closely linked to CRb and resistance to Plasmodiophora brassicae isolate No.14 in 108 cultivars

Marker	KB59N07	KB59N06	KB59N05	KB59N03	B4701	B4732	B1321	B1324	B1210	B1005	B0902	DIa	Category
Cultiver
Cultivars of Chinese cabbage
CR Shinkib	158/160	244/249	175/177	187/201	187/188	222/226	135/169	205/206	169/192	240/241	160/241	1.0 (−)	A
	H	H	H	H	H	H	H	H	H	H	H
Akiriso	S	H	H	H	H	H	H	H	H	H	H	0.9 (−)
CR Kaiou	H	H	H	H	H	H	H	H	H	H	H	1.4
CR Kisen	H	R	R	R	R	H	R	R	R	H	R	1.0 (−)
CR Ouken 65	H	H	H	H	H	H	H	H	H	H	H	1.5
CR Seiga 65	S	H	H	H	H	H	H	H	H	H	H	1.1
Haregi 60	H	H	H	H	H	S	R	R	R	H	H	0.8 (−)
Haruriso	S	H	H	H	H	S	R	R	R	H	H	0.9 (−)
Harusakari 2 gou	H	R	H	S	H	R	R	R	R	H	H	1.2
Kien 80	R	R	H	H	H	R	R	R	R	H	H	0.1 (−)
Kifuku 65	H	H	H	H	H	H	S	S	H	S	H	1.1
Kigokoro 65	H	H	H	H	H	H	R	R	R	H	H	0.9 (−)
Kigokoro 85	S	H	H	H	H	S	H	H	H	H	H	1.0 (−)
Kiraboshi	H	H	H	H	H	S	H	H	H	H	H	0.2 (−)
Kukai 65	H	H	H	H	H	H	H	H	H	H	H	0.8 (−)
Kukai 70	S	H	H	H	H	H	H	H	H	H	H	0.8 (−)
Meisyun	S	H	H	H	H	H	H	H	H	H	H	0.7 (−)
Minebuki 505	H	H	H	H	H	S	H	H	R	H	H	0.1 (−)
Moegi	S	H	H	H	H	H	H	H	H	H	H	1.3
NNH-104	S	H	H	H	H	H	H	H	H	H	H	1.3
Satobuki 613	H	H	H	H	H	H	H	H	H	H	H	0.4 (−)
Shoki	S	R	R	R	R	S	H	H	R	S	R	2.3
Shoshun	H	H	H	H	H	S	R	R	R	H	H	0.3 (−)
Super CR Shinriso	H	H	H	H	H	H	H	H	H	H	H	0.5 (−)
Yuki	S	H	H	H	H	H	H	H	H	H	H	0.9 (−)
CR Kikoma	S	H	H	H	H	S	S	S	H	S	H	1.7	C
CR Kyotakara 80	S	H	H	H	H	S	S	S	H	S	H	1.3
CR Saitaikai	S	H	H	H	H	S	H	S	H	S	H	0.4 (−)
Harebutai 65	S	H	H	H	H	S	S	S	H	S	H	2.3
Haregi 90	S	H	S	S	H	S	H	S	H	S	H	0.3 (−)
Kigokoro 80	S	H	H	H	H	S	H	S	H	S	H	0.5 (−)
Kinami 90	S	H	H	S	H	S	H	S	H	S	H	0.4 (−)
Kiryoyoshi 70	S	H	H	H	H	S	S	S	H	S	H	1.9
Kougetsu 77	S	H	H	H	H	S	S	S	H	S	H	1.4
Kougetsu 87	S	H	H	H	H	S	S	H	H	S	H	0.7 (−)
Ryutoku	S	H	H	H	H	S	S	S	H	S	H	0.4 (−)
Strong CR 75	S	H	H	H	H	S	S	S	H	S	H	0.5 (−)
W-1117	H	H	H	H	H	S	S	S	H	S	H	0.3 (−)
CR Kyotakara 70	S	H	S	S	S	S	S	S	S	S	H	0.2 (−)	D
Haruwarai	S	S	S	S	S	S	H	S	S	S	H	0.8 (−)
Kiai 65	S	H	S	S	S	S	S	S	S	S	R	0.7 (−)
Kigokoro 75	S	H	S	S	H	S	S	S	S	S	H	0.2 (−)
Kigokoro 90	S	S	S	S	S	S	H	S	H	S	S	2.1
Omoni 75	S	S	S	S	S	S	H	H	H	S	H	0.4 (−)
Super CR Akinishiki	S	H	S	S	S	S	H	S	S	S	H	0.3 (−)
Banchu Daifuku	H	H	H	H	H	S	R	R	R	H	S	3.0 (+)	E
Harukomachi	H	H	H	H	H	S	R	R	R	H	S	3.0 (+)
Hiroshimana	H	H	H	H	H	S	H	H	H	H	S	3.0 (+)
Kikunishiki	H	H	H	H	H	H	H	H	H	H	S	3.0 (+)
Kyoto 3 gou	S	H	H	R	R	H	R	R	R	H	S	3.0 (+)
Aichi Hakusai	S	S	S	S	H	S	S	S	S	S	S	3.0 (+)	B
Banki	S	S	S	S	S	S	R	S	S	S	S	3.0 (+)
Bansei Osaka Shirona	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Chikara	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Chirimen Hakusai	H	S	H	S	H	S	H	S	S	H*	S	3.0 (+)
Chukanbohon Nou 9 gou	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Gokui	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Hankekkyu Santosai	S	H	S	S	S	S	S	S	S	S	S	3.0 (+)
Harukisaku	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Harumakigokuwase	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Harusakari	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kaga Hakusai	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kanjiro	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kanmidori	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)	B
Kashin Hakusai	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kekkyu Chiifu	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kiko 65	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kikumusume 65	S	S	S	S	H	S	S	S	S	S	S	3.0 (+)
Kikumusume 70	S	S	S	S	S	S	S	S	S	H	S	3.0 (+)
Kikumusume 80	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kimikomachi	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Kimusan 75	S	S	S	S	S	S	H	S	S	S	S	3.0 (+)
Kurimu 2 gou	S	S	S	S	S	S	H	S	S	S	S	3.0 (+)
Matushima Shin 2 gou	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Meikyo	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Mumbichi	S	H	S	S	S	S	S	S	S	S	S	3.0 (+)
Muso	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Nagasaki Hakusai	S	S	S	S	H	S	S	S	S	H	S	3.0 (+)
Nakate Osaka Shirona	S	S	H	H	H	S	S	S	H	S	S	3.0 (+)
Natsuryokutosyosai	S	H	H	H	H	S	S	S	H	S	S	3.0 (+)
Okiniiri	H	S	S	S	H	S	S	S	S	S	S	3.0 (+)
Orenjikuin	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Osome	S	H	H	H	H	S	S	S	H	S	S	3.0 (+)
Puchihiri	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Saiki	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Saisei	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Shigatsu Shirona	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Shimoshirazuna	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Shimoyamachitose	S	H	H	H	H	S	S	S	H	S	S	3.0 (+)
Shin Hayabusa	S	S	S	S	H*	S	S	S	S	S	S	3.0 (+)
Shinriso	S	S	S	S	S	S	S	S	S	H	S	3.0 (+)
Shinseiki	S	S	S	S	S	S	S	S	S	H	S	3.0 (+)
Shunju	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Super CR Hiroki	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Taibyo Hoshun 60	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Tsuyoki	S	S	S	S	S	S	H	S	S	S	S	3.0 (+)
W-1116	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Cultivars of Japanese turnip
CR Takamaru	H	H	H	H	H	S	R	R	R	R	H	0.1 (−)	A
CR Yukimine	H	H	H	H	H	H	R	R	R	R	R	1.0 (−)
CR Omasa	S	S	S	S	S	H	S	S	S	H	S	3.0 (+)	B
Hakuba	S	S	S	S	S	S	S	S	–	H	S	3.0 (+)
Hida Benikabu	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Iwate Rokunohe Akachokabu	S	S	S	S	S	S	S	S	S	S	S	3.0 (+)
Natsumaki 13 gou	S	S	S	S	S	S	H	S	S	H	S	3.0 (+)
Syogoin ohmarukabu	H*	S	S	S	H	S	S	S	–	S	S	3.0 (+)
CR European fodder turnips
Gelria R	R	H	R	H	H	H	R	R	R	H	H	0.1 (−)	A
Siloga	H	H	H	H	H	S	H	H	R	S	H	1.1
Debra	H*	S	S	S	H*	S	H	H*	H*	S	H*	2.2	C
PIC	0.43	0.82	0.54	0.71	0.38	0.63	0.33	0.35	0.31	0.28	0.51
Diagnostic accuracy (%)	66.7	87.0	82.4	81.5	78.7	69.4	77.8	77.8	86.8	65.7	99.1

+, susceptible (DI = 3.0); −, resistant (DI ≤ 1.0). Numbers in italics were reported by Hatakeyama .

Dual numbers indicate the genotypes of each marker in ‘CR Shinki.’ Underlining indicates resistance alleles.

Segregation in a cultivar and both resistant and susceptible genotypes detected.

−, No amplification detectable.

R, Homozygous for the resistant genotype; S, absence of the resistance genotype; H, heterozygous (resistant and other genotypes); PIC, polymorphic information content.

Development of genetic markers linked to CRb

DNA markers for fine mapping of CRb were designed on the basis of genome sequencing information from the B. rapa Genome Sequencing Project (BrGSP: http://www.brassica-rapa.org/BRGP/index.jsp; Mun ) and Brassica database (BRAD: http://brassicadb.org/brad/; Wang ). We previously demonstrated that SSR markers, KBrH059N21F, KBrH129J18R and KBrB091M11R developed from bacterial artificial chromosome (BAC) end sequences in BrGSP were located around CRb (Kato ). We developed more SSR markers based on four full BAC sequences (KBrH059N21, KBrH129J18, KBrB091M11 and KBrH069E08). Although CRb was located between KBrH059N21 and KBrH129J18 (Kato , Fig. 1A), these two BAC clones were discontinuous, and genetic information in this region was not available. To develop a marker in this region, we used the reference B. rapa genome (line Chiifu-401) sequence in BRAD. SSR primers were designed in the read2Marker program (Fukuoka ). These SSR markers and indel markers based on KBr BAC sequences and the genomic sequence information of BRAD were named KB_N and B_N markers, respectively (Supplemental Table 2). PCR amplification was carried out in 10-μl reaction mixtures containing 10 ng genomic DNA, 4 pmol of each primer and 2× Quick Taq HS Dye Mix (Toyobo, Osaka, Japan) in a GeneAmp PCR System 9700 (Applied Biosystems, Foster City, CA, USA). The reaction was performed with the following parameters: 1 cycle of 94°C for 2 min; 35 cycles of 94°C for 30 s, 55°C for 30 s and 68°C for 30 s and final 68°C for 5 min. Polymorphisms were detected as described (Hatakeyama ). To carry out fine mapping of CRb in ‘CR Shinki’, we searched for polymorphisms between the selected resistant and susceptible F2 plants as described (Kato ).

Fig. 1

A partial linkage map of Brassica rapa in the CRb region and a corresponding physical map on R3. (A) Linkage map containing the CRb locus in our previous study (Kato ). Marker names and genetic distances (cM) are indicated on the right and left, respectively. (B) The positions of markers developed between TCR05 and KBrB091M11R in this study. Marker names and positions (kb) are shown at the left and right, respectively. Positions of markers were determined on the basis of the B. rapa genome sequence of chromosome R3 from the Brassica database (BRAD: http://brassicadb.org/brad/). CRa-linked SC2930 (Matsumoto ) is shown in bold. All developed marker positions are shown in Supplemental Table 2.

We determined the genotype of CRa closest marker, SC2930 (Matsumoto ), and CRb near marker, KBrH129J18R (Kato ) and CRb phenotype of 172 F2 plants and the recombinants selected from 2,032 F2 plants for comparing the position between CRb and CRa. We also used OPC11-2S (Saito ) for Crr3 analysis.

Fine mapping of CRb

We fine-mapped CRb by comparing phenotypes and genotypes of markers in recombinants. We screened 2,032 F2 plants with two markers, TCR05 and KBrB091M11R (Fig. 1A) and obtained F3 seeds by self-pollination of F2 recombinants. We tested 40 to 60 F3 plants of each F2 line for CR and determined the genotype of the markers around CRb. After removing heterozygotes and nonrecombinants, we calculated the mean DI for each F3 line from the results of 9 to 15 recombinants homozygous for the two markers in each F3 line. To confirm that CRb is located between TCR05 and KBrB091M11R, we tested for CR in each of the three non-recombinant F3 lines. Analysis of the data obtained from these markers and the CR test allowed us to narrow down the position of CRb.

To estimate candidate genes for CRb, we inspected protein homologies within the chromosomal regions determined by fine mapping analysis using a BLASTX search against protein sequences in GenBank (http://www.ncbi.nlm.nih.gov/genbank).

Evaluation of the usefulness of the markers for MAS

We determined the genotypes of 11 markers around CRb and compared the relationship between marker genotypes and resistance in 108 cultivars. CRb confers resistance to P. brassicae field isolate No. 14 by a single dominant gene (Kato ). The diagnostic accuracy for MAS was calculated as Vidal :

where TP (true positives) is the number of resistant cultivars with a resistance allele (either homozygous [“R”] or heterozygous [“H”]; Table 2) and TN (true negatives) is the number of susceptible cultivars with no resistance allele (homozygous [“S”]; Table 2).

The polymorphic information content (PIC) values of 11 markers for the 108 cultivars were calculated as described by Anderson .

Results

Development of DNA markers based on genomic information of Brassica rapa

We designed 149 primer pairs, 54 from BrGSP and 95 from BRAD. Of these 149, 46 (31%) showed polymorphism and all amplified PCR-based co-dominant markers; however, 13 were excluded because their marker distances were too close to another marker. The remaining 33 markers comprised 26 SSR and 7 indel markers between TCR05 and KBrB091M11R (Fig. 1 and Supplemental Table 2). Fifteen of these markers were based on BAC sequences from KBrH069E01 (2 of KB69_N), KBrH059N21 (5 of KB59_N), KBrH129J18 (5 of KB29_N) and KBrB091M11 (3 of KB91_N). The other 18 markers (B_N) were based on the 300-kb region (from 24.3 to 24.6 Mb on R3 in BRAD) between BACs KBrH059N21 and KBrH129J18. Forty-two predicted genes (open reading frames: ORFs) existed in the region; 4 (B4701, B1005, B0902 and BGA01) of the 18 markers were located in ORFs, and the other 14 in the non-coding region. We fully developed 28 markers within the genomic region containing CRb between KBrH059N21F and KBrH129J18R (Fig. 1); the average distance of these marker loci was 20.4 kb.

We used 2,032 F2 plants to more precisely localize the markers with respect to CRb. To more precisely delimit the CRb locus, we genotyped the F2 plants first with the TCR05 and KBrB091M11R markers, and identified 92 F2 plants recombinant between these two markers. We then determined the genotypes of these plants with 37 markers including developed markers between TCR05 and KBrB091M11R (Fig. 2). The 92 recombinants were divided into three groups: group 1, 35 plants with recombination between KBrH059N21F and KBrH129J18R; group 2, 37 plants with recombination between TCR05 and KBrH059N21F and group 3, 20 plants with recombination between KBrH129J18R and KBrB091M11R. We obtained F3 seeds from 22 F2 recombinant plants from group 1, 9 from group 2 and 4 from group 3 and used them for the CR test. CR scores of the F3 families segregated into two classes of DI scores, <0.6 (resistant) and >2.8 (susceptible) (Fig. 2). A comparison of phenotypic DIs and genotypes in the F3 plants showed that CRb was proximal to TCR05 and that the sequences conferring resistance are located in the ca. 140-kb region between KB59N07 (based on recombinant lines 1417, 1866 and 2002) and B1005 (based on recombinant line 1118). This region was 35 kb (KB59N07) to 175 kb (B1005) proximal to KBrH059N21F (2420.1–2434.2 kb on R3 based on information from BRAD). Phenotypes of non-recombinant F3 plants were completely matched to the allele type of TCR05 and KBrB091M11R. These were either fully resistant (mean DI = 0.0) or fully susceptible (mean DI = 3.0) in each of the three F3 lines (n = 16 for each; data not shown).

Fig. 2

Graphical genotype of selected recombinants and their clubroot disease indexes (DI) in the F3 families. Line names and DIs are shown on the left and right, respectively. Marker names are indicated at the top. The DI was calculated from the average of 9 to 15 plants in each line. Allelic identities are shown in white for resistance alleles and black for susceptibility alleles. Potential chromosomal regions spanning the CRb gene are highlighted above in.gray R, resistant; S, susceptible.

The protein homologies within the 140-kb genomic region between KB59N07 and B1005 in each line were inspected by using a BLASTX search against the NCBI database. Fourteen sequences contained coding regions for proteins with putative functions (Table 1), including the Rho-binding family protein, B-glucosidase 47, the MATE family protein and the Toll-interleukin-1 receptor/nucleotide-binding site/leucine-rich repeat (TIR-NBS-LRR) class disease-resistance protein.

Table 1

Putative function of candidate genes in the CRb region, deduced by BLASTX sequence search of the NCBI database

Markera	Protein homology	E-value
KB59N07
	Nitrate transporter	0
	Pentatricopeptide repeat-containing protein	3e−154
	DNA-directed RNA polymerase	8e−179
KB59N06
	Rho-binding family protein	4e−17
KB59N05
	Homeobox-leucine zipper protein MERISTEM L1	0
	β-glucosidase 47	1e−146
	β-glucosidase 47	1e−77
	Tobamovirus multiplication protein 1	2e−14
	Peptide methionine sulfoxide reductase	9e−83
	MATE efflux family protein	3e−167
KB59N03
	MATE efflux family protein	9e−71
B4701	MATE efflux family protein	6e−176
B4732
	TIR-NBS-LRR class disease resistance protein	1e−133
B1321
B1324
B1210
B1005	TIR-NBS-LRR class disease resistance protein	3e−149

Ten markers identified in the genomic region by fine mapping of CRb (see Fig. 2). B4701 and B1005 markers were positioned in the coding region; other markers were in the flanking region.

Relationship between CRb and other CR loci

To clarify the positional relationships between CRa and CRb, we compared the genotype of SC2930 and KBrH129J18R in the F2 population. SC2930 was located at 24,780–24,781 kb on R3 according to genomic sequence information from BRAD and was close to KBrH129J18R (Fig. 1 and Supplemental Table 2). The genotypes of SC2930 and KBrH129J18R were consistent in all 172 F2 plants and 92 recombinants selected from 2,032 F2 plants of ‘CR Shinki’, and no recombinants were detected between the two markers (data not shown).

The F2 plants of ‘CR Shinano’, which has a Crr3 resistance gene, were all susceptible to P. brassicae isolate No. 14 (Fig. 3). The genotypes of these plants were segregated with the Crr3-linked marker OPC11-2S (Saito ), whereas genotypes of the five markers around CRb (KB59N05, KB59N03, B4701, B4732 and B0902) showed a susceptible allele pattern (Fig. 3, only B0902 is shown). These results indicate that Crr3 is not effective against isolate No. 14.

Fig. 3

Relationships between resistance to pathotype group 3 and genotypes of Crr3. Eleven F2 plants derived from selfing ‘CR Shinano’ and having Crr3 (Saito et al. 2004) were inoculated and genotyped with Crr3 and CRb markers. The arrowheads on the left indicate resistance (R) or susceptibility (S) alleles. Primers used in this analysis are listed on the left. The last three columns indicate the control F3 plants of ‘CR Shinki’: RR, homozygous resistant; rr, homozygous susceptible; and Rr, heterozygous. H, heterozygous; M, marker.

Usefulness of the new markers in MAS

To evaluate the usefulness of the new CRb-linked markers, we examined the phenotypes and genotypic patterns of 11 markers of 108 cultivars. Fifty-eight (54%) of these cultivars, including some CR cultivars, such as ‘Kiko 65’, ‘Shinseiki’ and ‘CR Omasa’, showed susceptibility (DI = 3.0), while another 50 (46%) were resistant (DI ≤ 1.0) or showed partial resistance (DI = 1.1–2.3) to P. brassicae isolate No. 14 (Table 2). On the basis of the relationship between resistance to No. 14 and the genotypic pattern of the 11 markers around CRb, the 108 cultivars were roughly classified into five categories: A, B, C, D and E (Table 2). Category A contained 29 cultivars that were resistant to No. 14 and were heterozygous or homozygous for the resistant genotype at most markers. Category B contained 53 cultivars that were susceptible to No. 14 and were homozygous for the susceptible genotype at most of the markers. Cultivars of Category C and D were resistant to No. 14 but were homozygous for susceptibility at half of the markers (C) or most of the markers (D). Category E contained 5 cultivars that were susceptible, although they were heterozygous or homozygous for resistance at most markers. In categories A and B, the genotypes of the markers are likely to be consistent with the resistance and susceptibility responses, respectively, to No. 14.

Of the markers tested, PIC values ranged from 0.28 to 0.82, with an average of 0.49 (Supplemental Table 2), and diagnostic accuracies ranged from 65.7% to 99.1%, with an average of 79.4% (Table 2). The genotype of marker B0902 was coincident with the CR phenotype, except in ‘Kigokoro 90.’ Most cultivars with the resistance genotype showed resistance or partial resistance, while cultivars that were homozygous for the susceptibility genotype had DI = 3.0.

Discussion

We drew on the results of our previous study (Kato ) to develop high-density B_N markers around CRb, using the whole-genome sequence information of B. rapa in BRAD. Using these developed markers, we delimited the candidate region of CRb within the 140-kb region between KB59N07 and B1005 by fine mapping analysis (Fig. 2). CRb was previously reported to be distal to TCR05 on R3 (Piao ), but our previous coarse mapping data placed CRb proximal to TCR05 (Kato ). This study supported our previous data and clarified the accurate position of CRb.

Fourteen functional proteins were predicted in the 140-kb region. The Rho family proteins are involved in mediating the plant defense response (Agrawal ). The Rho-binding family includes ROP-binding kinase 1, which is involved in resistance to barley powdery mildew fungus (Huesmann ). Therefore, sequence similarity places the Rho-binding protein between markers KB59N05 and KB59N06 might be involved in resistance to clubroot. We found two TIR-NBS-LRR class disease-resistance proteins in the candidate interval (Table 1). Many plants have major classes of resistance (R) genes, such as the NBS-LRR structure class (Eitas ). Crr1, Crr2 and CRb have similar origins in the ancestral genome, which has disease-resistance gene clusters—as in chromosome 4 of Arabidopsis thaliana (Piao , Suwabe ). Therefore, one of the two sequences encoding TIR-NBS-LRR proteins found in the 140-kb region could also be a candidate gene for CRb.

We then examined the relationship among three CR genes, CRa, CRb and Crr3, located on chromosome R3. In segregation analysis using total 2,204 F2 plants of ‘CR Shinki’, no recombinants between SC2930 and KBrH129J18R were detected and these markers cosegregated. The physical distance between two markers is 23 kb in line Chiifu-401 and this segregation analysis indicated the distance in CRb region might be conserved. The position of SC2930 which is 0.6 cM proximal to CRa (Matsumoto ) and KBrH129J18R which is 0.4 cM proximal to CRb (Fig. 1A, Kato ) were reported. Although two findings are based on independent experiment, these linkage distances were very similar and it is speculated the position of CRa and CRb is the same or extremely near. CRa showed pathotype specific resistance to M85 (Matsumoto , 2012). It is likely that the pathogenicity of M85 is similar to that of No. 14 because both isolates showed similar pathogenic responses against our differential hosts ‘Super CR Hiroki’ and ‘Ryutoku’ (Hatakeyama , Matsumoto ). Further, both ‘CR Shinki’ carrying CRb (Piao ) and the doubled haploid line T136-8 carrying CRa (Matsumoto , 2012) showed similar resistance profiles to pathotype groups 3 and 4 but were susceptible to groups 1 and 2 (Hatakeyama , Matsumoto ). These results suggest that the CRb and CRa genes are identical or are clustered together.

Crr3 and CRk were identified on R3, but CRa was located over 30 cM from them (Sakamoto ). Our study confirmed through a CR test that Crr3 does not confer resistance to P. brassicae isolates No. 14, which is classified as pathotype group 3. These results indicate that Crr3 does not compensate for the race specificity of Crr1 and Crr2. CRk could not be examined because ‘CR Kanko’ carrying CRk (Sakamoto ) is no longer marketed.

The relationship between the genotypes of 11 markers around CRb and the resistance or susceptibility to isolate No. 14 identified the cultivars in Category A as resistant and those in Category B as susceptible to No. 14, indicating that the cultivars in Category A probably possessed CRb and those in Category B did not. Our results suggest that our new markers might select cultivars carrying CRb (Category A) and be useful when introducing CRb into susceptible cultivars (Category B). We consider that marker B0902 would be particularly useful, because it has high polymorphism (PIC = 0.51) and it is easily detectable by agarose gel electrophoresis (Fig. 3).

There were discrepancies between genotypes and phenotypes in other categories. Category D cultivars probably did not possess CRb but were resistant to isolate No. 14, suggesting that other CR loci might confer resistance to No. 14. CRa and CRk, which are effective to M85, are possible candidate CR loci. Since CRa showed pathotype-specific resistance and CRk showed broad-spectrum resistance (Matsumoto ), CR test of the cultivars of Category D using isolates of other pathotypes except for group 3 would clarify which CR loci confer resistance in these cultivars. In contrast, Category E cultivars were fully susceptible, although they probably had CRb. These cultivars might have nonfunctional CRb alleles, and the markers closely linked to CRb did not reflect this mutation. Category C cultivars were resistant but their genotypes varied and therefore it is unclear whether CRb or other CR loci conferred resistance. However, the genotypes of some markers located between KB59N06 and B4701 and B1210 were heterozygous in most of the cultivars in Category C, suggesting that CRb is located in these regions. Cloning of the CR genes would help to clarify the identities and copy number of CR genes in the present CR cultivars.

Clubroot resistant genes Crr1, Crr2 and Crr4 did not confer resistance to isolate No. 14 (unpublished data). We also demonstrated that Crr3 is not related to resistance/susceptibility and that CRb has an important role in conferring resistance. Our profiling analysis indicated that most of the susceptible cultivars did not share the resistance genotypes of the markers around CRb. In a previous study (Kato ) we showed that ‘CR Shinki’ and ‘Akiriso’ carried CRb as a single gene, and consequently we can easily introduce CRb from these cultivars into susceptible cultivars by using the marker information from this study.

In conclusion, we demonstrated by fine mapping analysis that CRb is localized in a genomic region of about 140 kb between KB59N07 and B1005, where we found disease-related genes. We also demonstrated positional relationships between CRa and CRb. A set of 11 markers near CRb, in particular, B0902, was able to discriminate resistance in 108 cultivars. Our methodology in developing markers and our results on the genetic information of CRb are important contributions to breeding new CR cultivars by MAS and will be useful for isolating the CRb gene by positional cloning.