Sources of resistance to Plasmodiophora brassicae (clubroot) pathotypes virulent on canola

Abstract

A collection of 955 Brassica accessions including B. rapa (718), B. napus (94), B. juncea (93), B. oleracea (30), B. carinata (12) and B. nigra (8) was screened against Plasmodiophora brassicae pathotype 3 (1 × 10⁶ resting spores cc⁻¹ growth medium), the predominant strain of the pathogen on canola in western Canada. A total of 35 accessions (mostly B. rapa) showed at least 50% reduced clubroot severity relative to a susceptible control, with 15 showing complete resistance (clubroot-free). Ten resistant accessions representing Brassica A-, B- and C-genome species were tested further using a 10-fold higher pathogen inoculum dose (1 × 10⁷ resting spores cc⁻¹ growth medium) and by testing them against the five pathotypes (2, 3, 5, 6 and 8) of P. brassicae found in Canada. One B. nigra, two B. oleracea and four B. rapa (oriental vegetable) accessions maintained a high level of resistance under the higher pathogen inoculum pressure, while one B. nigra and two B. rapa (turnip) accessions showed moderate resistance. Most of the selected clubroot-resistant accessions showed consistent resistance to each of the five P. brassicae pathotypes found in Canada, except for one B. nigra and two turnip accessions, which varied slightly against different pathotypes. Several promising sources of clubroot resistance were identified in this study that can be used to develop new canola germplasm with a diverse clubroot resistance background for potentially more durable clubroot resistance.

Résumé

Une collection de 955 accessions de Brassica, comprenant B. rapa (718), B. napus (94), B. juncea (93), B. oleracea (30), B. carinata (12) et B. nigra (8), a été criblée vis-à-vis de Plasmodiophora brassicae pathotype 3 (1 × 10⁶ spores de repos/cc−1 de milieu de culture), la souche de l’agent pathogène la prédominante du canola dans l’Ouest canadien. En total, 35 accessions (B. rapa, principalement) ont montré une réduction de la sévérité de la hernie par de l’ordre de 50 % en comparaison à un témoin susceptible, dont 15 ont montré une résistance complète (sans hernie). Dix accessions résistantes représentant des espèces de Brassica à génomes A, B et C ont de plus été testées en utilisant une dose d’inoculum 10 fois plus supérieure (1 × 10⁷ spores de repos/cc−1 de milieu de culture) et ont été testés vis-à-vis de 5 pathotypes (2, 3, 5, 6 et 8) de P. brassicae trouvés au Canada. Une accession de B. nigra, deux de B. oleracea et quatre de B. rapa (végétale oriental) ont maintenu un haut degré de résistance sous la pression de la dose la plus élevée d’inoculum, tandis qu’une accession de B. nigra et deux de B. rapa (navet) ont montré une résistance modérée. La plupart des accessions résistantes sélectionnées ont montré une résistance continue à chacun des cinq pathotypes de P. brassicae trouvés au Canada, à l’exception d’une accession de B. nigra et de deux de navet dont la résistance variait légèrement par rapport au différents pathotypes. Plusieurs sources prometteuses de résistance à la hernie identifiées au cours de cette etude pourraient être utilisées pour développer un nouveau germoplasme de canola ayant différent background de résistance à la hernie pour une résistance plus durable à cette maladie.

Keywords:

• black mustard
• cabbage
• Chinese cabbage
• crucifer
• pak choy
• resistance spectrum

Mots clés:

• choux
• Crucifères
• moutarde noire
• pak-choï
• pé-tsai
• spectre de résistance

Introduction

Clubroot, caused by the plasmodiophorid pathogen Plasmodiophora brassicae Woronin, is a serious disease of Brassica crops worldwide (Dixon 2009) and a threat to canola (Brassica napus L.) production in western Canada (Howard et al. 2010; Hwang et al. 2012). Pathotypes 2, 3, 5 and 8 of P. brassicae (Williams 1966) have been found in western Canada, with pathotype 3 being the most prevalent (Xue et al. 2008; Cao et al. 2009) and virulent (Strelkov et al. 2006, 2007) on canola. Pathotypes 2 and 6 are more common in eastern Canada (Cao et al. 2009) on cruciferous vegetable crops (Ayers & Lelacheur 1972; Chiang & Crête 1985, 1989) as well as on canola (Vigier et al. 1989; Pageau et al. 2006).

Cultivar resistance is a cornerstone for the management of clubroot (Hirai 2006; Diederichsen et al. 2009). Resistant Chinese cabbage (B. rapa subsp. pekinensis) and oilseed rape (B. napus) have been developed in Japan (Hirai 2006; Saito et al. 2006; Sakamoto et al. 2008; Kamei et al. 2010) and Europe (Gowers 1982; Diederichsen et al. 2006), respectively. Almost all clubroot-resistant (CR) Brassica crops have single gene-based resistance that is often race- or pathotype-specific (Diederichsen et al. 2009). Many clubroot resistance genes can be traced back to the European fodder turnips in which they were originally identified (Crute et al. 1980; Piao et al. 2009). All canola cultivars in western Canada were highly susceptible to clubroot (Strelkov et al. 2006) until 2009, when the first resistant cultivar was introduced. Since then, several additional CR cultivars have been released for commercial production, but all of them carry a single clubroot resistance gene. Historically, single gene-based resistance does not provide durable control of clubroot, as observed for Chinese cabbage in Japan (Hatakeyama et al. 2006) and winter oilseed rape in Europe (Oxley 2007). LeBoldus et al. (2012) exposed one of the Canadian CR canola cultivars to a pathotype 3 population of P. brassicae under controlled conditions and observed a substantial erosion of resistance after only two repeated cycles. Therefore, it is prudent to identify diverse sources of clubroot resistance and develop canola germplasm with resistance based on complementary modes of action. These efforts may help broaden the basis of resistance and enhance the long-term effectiveness of clubroot management.

Clubroot resistance was first documented in the 1930s, but efforts to screen for resistance were sporadic until the 1960s, when highly resistant European turnips were identified (Wit & Van De Weg 1964; Tjallingii 1965). So far, clubroot resistance has been reported more frequently in certain crop types, including turnip (B. rapa), rutabaga (B. napus) and white cabbage or kale types of B. oleracea (Diederichsen et al. 2009). Some of the most resistant germplasm has been found in B. rapa (Hirai 2006), where resistance to multiple P. brassicae pathotypes has been observed more often than in other Brassica species (Toxopeus et al. 1986). Only some of the pathogen isolates from Europe or Japan have been able to overcome the resistance in B. rapa (turnip), whereas more isolates were able to cause clubroot on the B. napus and B. oleracea hosts in the European Clubroot Differential (ECD) set (Toxopeus et al. 1986). The turnips ‘Siloga’, ‘Gelria’, ‘Milan White’ and ‘Debra’ have also been used as parental stocks to breed other CR cultivars (Diederichsen et al. 2009). Brassica oleracea is another potentially important source of clubroot resistance and has been evaluated extensively. Crisp et al. (1989) assessed nearly 1000 B. oleracea accessions and confirmed the resistance in several European kales and cabbage. Carlsson et al. (2004) also found CR germplasm while screening a collection of 52 landraces, old cultivars and wild accessions of B. oleracea and closely related Brassica species. Clubroot resistance has also been introduced into B. oleracea from rutabaga (Chiang et al. 1977) via interspecific hybridization. Crute et al. (1983) suggested that CR B. oleracea was more likely to be overcome by P. brassicae isolates that might not be virulent on the ECD B. rapa and B. napus hosts, possibly due to different resistance mechanisms in B. oleracea. In Canada, clubroot resistance has previously been sought for the pathotypes common in the central or eastern provinces. Chiang and Crête (1972) assessed 334 Brassica accessions against pathotypes 2 and 6 collected in eastern Canada and reported that several cabbages were highly resistant. Similarly, Vigier et al. (1989) tested 31 spring canola cultivars and breeding lines against pathotypes 2 and 6 and identified several that were moderately resistant. The Canadian cabbage ‘Acadie’ was resistant to pathotypes 1 and 6, while ‘Richelain’ was resistant to pathotypes 2, 6 and 7 (Chiang & Crête 1985, 1989). Hasan et al. (2012) recently evaluated 275 Brassica accessions against the five P. brassicae pathotypes found in Canada and identified resistance to all five pathotypes in the diploid species B. rapa, B. nigra and B. oleracea, and in the amphidiploid B. napus. However, the study included a relatively small number of B. rapa accessions (36) even though this A-genome group of Brassica is considered the most important source for clubroot resistance (Toxopeus et al. 1986; Hirai 2006).

The objectives of this study were to: (1) screen a large collection of Brassica germplasm, especially B. rapa, for resistance to P. brassicae pathotype 3; (2) evaluate selected CR accessions using larger plant populations and higher pathogen inoculum pressure to determine the effectiveness and consistency of resistance; and (3) assess the resistance spectrum of these selected CR accessions against all P. brassicae pathotypes found in Canada.

Materials and methods

Sources of Brassica germplasm

A collection of 955 Brassica accessions including cultivars, landraces, breeding lines and other selections were obtained from nine sources in Canada, China, Germany, the UK and the USA. Species included B. carinata (12), B. juncea (93), B. napus (94), B. nigra (8), B. oleracea (30) and B. rapa (718). Most of the B. rapa accessions were obtained from Plant Gene Resources of Canada (PGRC) at the Saskatoon Research Centre (SRC), Agriculture and Agri-Food Canada (AAFC), and all the B. carinata accessions were obtained from the Brassica breeding programme at AAFC-SRC. Viterra Inc. (Saskatoon, SK, Canada) provided most of the B. juncea accessions, and the B. napus accessions were provided by PGRC, the Anhui Academy of Agricultural Science, China and the Sichuan Academy of Agriculture Science, China. Several commercial B. rapa (Chinese cabbage) and B. oleracea (cabbage) cultivars with reported resistance to clubroot were received from Bejo Seeds Inc. (Oceano, CA, USA) and Syngenta Seeds UK Ltd. (Fulbourn, Cambridge, UK). The Leibniz Institute of Plant Genetics and Crop Plant Research (IPK, OT Gatersleben, Germany) and the Genetic Resources Unit, Warwick HRI (University of Warwick, UK) also provided a few accessions of B. nigra and B. napus, respectively. In early screening trials, the clubroot-susceptible (CS) canola ‘Fortune RR’ was used as an inoculated control. In later trials, in which selected accessions were tested for resistance against a range of pathotypes, B. rapa ‘Granaat’ (Chinese cabbage) was used as a universally susceptible control (ECD-05). Throughout the study, non-inoculated plants did not produce clubroot symptoms, so they were not routinely included in the trials.

Pathogen inoculum

Initial screening of germplasm accessions (Trial 1) and assessing the performance of selected accessions under higher pathogen inoculum pressure (Trial 2) were conducted using resting spores derived from a field population of P. brassicae. Clubroot galls were collected from a severely diseased canola field near Leduc, Alberta. The pathogen population in this field was predominantly pathotype 3 (Williams 1966), which is highly virulent on canola (Strelkov et al. 2007; Xue et al. 2008). Pathogen inoculum was increased in growth cabinets programmed to provide 18/23°C (night/day) and a 14-h photoperiod (512 µmol m² s⁻¹). ‘Fortune RR’ was inoculated by soil drenching using a P. brassicae resting spore suspension (1 × 10⁷ spores mL⁻¹) at seeding. Approximately 3 g of galls were immersed in 50 mL distilled water for 2 h to soften the tissue and subsequently homogenized in a Waring blender at high speed for 1 min. The resulting suspension was filtered through eight layers of cheesecloth, and the concentration of resting spores was estimated with a haemocytometer. The resting spore suspension was pipetted onto the surface of a growth medium in a root trainer called a ‘conetainer’ (3.5-cm diam × 20-cm tall; Stuewe and Sons, Corvalis, OR). Watering helped distribute the resting spores throughout the growth medium. About 5–6 weeks after inoculation, galls from diseased plants were harvested, dried at room temperature for 2 weeks, and buried in Sunshine #3 potting mix (SunGro Horticulture, Vancouver, BC) for about 1 month to allow maximum conversion of plasmodia into resting spores. The potting mix was kept moist by weekly watering to near saturation. Processed galls were dried at room temperature and stored at −20°C until use.

Single-spore isolates of P. brassicae pathotypes 2, 3, 5, 6 and 8, also classified based on the Williams (1966) differentials, were used to evaluate the resistance spectrum of selected accessions (Trial 3). These isolates included SACAN-ss3 (pathotype 2), SACAN-ss1 (pathotype 3), ORCA-ss4 (pathotype 5), AbotJE-ss1 (pathotype 6) and ORCA-ss2 (pathotype 8) (Xue et al. 2008). Galls caused by single-spore inoculum of these P. brassicae pathotypes on ‘Granaat’ were stored at −20°C until use.

Trial 1. Screening for resistance to clubroot caused by P. brassicae pathotype 3

Resting spores derived from a predominantly pathotype 3 field population of P. brassicae were used as inoculum for the initial assessment of the 955 Brassica accessions. Conetainers filled with Sunshine #3 potting mix (pH 5.8–6.3) were inoculated by pipetting 5 mL of a resting spore suspension (1 × 10⁷ spores mL⁻¹) into each conetainer to obtain about 1 × 10⁶ spores cc⁻¹ growth medium. Immediately after adding the pathogen inoculum, two seeds were sown into each conetainer at about 1-cm depth. Seeded conetainers were placed on a holding rack in a growth cabinet programmed as described above for 5 weeks. The growth medium was saturated during the first week post-seeding by daily watering and maintained at high moisture levels by watering every other day after the first week. Each seeded conetainer was thinned to a single plant at about the 1-true-leaf stage.

Space limitations and the large number of entries precluded simultaneous testing of all the accessions. Instead, accessions were tested in batches of 13, with 7 plants (replicates) per accession. A CS control (also 7 plants) was included in each batch. Each test was arranged as a completely randomized design (CRD). Five weeks after inoculation, plants were assessed individually for clubroot severity using a 0–3 scale (Fig. 1) adopted from Kuginuki et al. (1999) and Strelkov et al. (2006), where 0 = no galling, 1 = small galls on less than 1/3 of roots, 2 = small to medium-size galls on 1/3–2/3 of roots and 3 = severe galling with medium- to large-size galls on more than 2/3 of roots. Mean clubroot severity for each accession was compared with that of the CS control included in the same test. A test was deemed valid only when all CS control plants were diseased, with an average clubroot severity rating exceeding level 2 on the severity scale. This initial testing was done once to select highly resistant accessions for further assessment; a total of 75 batches of testing were carried out between 2009 and 2011.

Fig. 1. (Colour online) Left to right: Pictorial key (0–3) used for clubroot severity rating, 4–6 weeks after inoculation with a resting spore suspension of Plasmodiophora brassicae.

Trial 2. Performance of selected CR accessions under higher pathogen inoculum pressure

Two B. nigra, two B. oleracea and six B. rapa accessions with moderately high to very high resistance ratings in Trial 1 were selected for further assessment using larger plant populations and a higher inoculum concentration to assess variability in clubroot resistance. A protocol similar to that described for Trial 1 was used, but the pathogen inoculum level was increased by a factor of 10, reaching 1 × 10⁷ resting spores cc⁻¹ growth medium, to minimize the chance for disease escape. Twenty-one to 90 plants per accession, depending on the consistency of clubroot resistance observed in Trial 1, were inoculated. Fewer plants were inoculated for accessions that had a complete resistance reaction in Trial 1 (little variation among individuals); each accession was evaluated in 1–3 tests. ‘Fortune RR’ was included in each test as a susceptible control; it was previously determined to be 100% diseased with a severity rating of 3. Inoculated plants were placed in a growth cabinet programmed as described above for 6 weeks and assessed individually for clubroot severity using the 0–3 scale. The data were expressed as the number of plants in each clubroot severity category (resistance distribution).

Trial 3. Resistance spectrum of selected CR accessions against the five pathotypes of P. brassicae found in Canada

Seeds of the 10 CR accessions selected for Trial 2 were pre-germinated on moistened filter paper in Petri dishes, and 1-week-old seedlings were inoculated with P. brassicae by dipping the entire root system in a resting spore suspension for 10 s prior to being transplanted individually into Sunshine LA4 potting mix in 6-cm-diam plastic pots. Resting spores were extracted from frozen galls of each of the five single-spore P. brassicae isolates (pathotypes 2, 3, 5, 6 and 8) described above using the protocol described by Strelkov et al. (2006); resting spore suspensions were adjusted to 1 × 10⁷ spores mL⁻¹ using a haemocytometer. The inoculated seedlings were placed in a greenhouse (20 ± 2°C) for 6 weeks to allow disease symptoms to develop. The growth medium was saturated with tap water during the first week after inoculation and kept moist by regular watering. ‘Granaat’ was used as the CS control in these trials.

Resistance/susceptibility to each pathotype was determined individually for each accession. Two repetitions of the experiment were conducted, one at the University of Alberta, Edmonton and the other at the Crop Diversification Centre North, Alberta Agriculture and Rural Development, Edmonton. The experiment was a randomized complete block design (RCBD) with three blocks in each repetition. Within a block, 12 plants of each accession and the CS control were inoculated. Approximately 6 weeks after inoculation, each plant was uprooted and scored for clubroot severity using the 0–3 scale described above. For statistical analysis, a disease severity index (DSI) was calculated using the following formula modified from Horiuchi and Hori (1980) by Strelkov et al. (2006):

Data analysis

Mean clubroot severity based on seven replicates (plants) was calculated for each accession and the CS controls in Trial 1. Per cent clubroot reduction was calculated using the following formula and by comparing each accession to the CS control included in the same test:

Per cent clubroot reduction was a relative value against the CS control, and was therefore used to rank the resistance of multiple accessions within a species. The data for selected accessions evaluated under higher pathogen inoculum pressure and using larger populations (Trial 2) were organized as the total number of plants falling into each clubroot severity category (0–3) to characterize the consistency or variation in an accession.

The DSI data from the study of resistance spectrum with selected accessions (Trial 3) were analysed using the Statistical Analysis System (SAS Institute, version 9.1, Cary, NC). A square-root transformation was used to normalize the distribution of percentage data. Data from repeated trials were homogeneous based on Bartlett’s Test, and were therefore pooled prior to analysis. Analysis of variance (ANOVA) was performed initially using PROC ANOVA, and DSI means associated with each accession were separated using Fisher’s Protected LSD_0.05 when the ANOVA indicated significance (P ≤ 0.05). Means presented in the Results are based on non-transformed data.

Results

Trial 1. Screening for resistance to P. brassicae pathotype 3

Of the 955 accessions screened, the majority (631) were equally or more susceptible to P. brassicae pathotype 3 relative to the susceptible control ‘Fortune RR’ (Table 1). Fifty accessions showed low to moderate resistance, with 25–49% less clubroot severity than the control. Thirty-six accessions showed ≥50% clubroot reduction, of which 15 were completely resistant with no visible symptoms 5 weeks after inoculation. Highly resistant accessions (>75% clubroot reduction) were found in B. rapa (17), B. nigra (4) and B. oleracea (2), but not in any of the amphidiploid species tested.

Table 1. Brassica accessions falling into different resistance categories based on the reduction of clubroot severity relative to a susceptible control inoculated with Plasmodiophora brassicae pathotype 3 during resistance screening (Trial 1).^a

	Reduction of clubroot severity (%)
Taxon	0	1–24	25–49	50–74	75–100	Subtotal
B. carinata	9	3	0	0	0	12
B. juncea	53	30	10	0	0	93
B. napus	70	20	4	0	0	94
B. nigra	4	0	0	0	4	8
B. oleracea	22	2	1	3	2	30
B. rapa	473	183	35	10	17	718

^a The total number of accessions evaluated was 955.

The largest group of accessions in this screening was B. rapa, with a total of 718 entries. Although only 26 of the B. rapa accessions showed >50% clubroot reduction (Table 2), a number of highly resistant accessions were identified in this group, especially in the subtaxa subsp. chinensis (pak choy), subsp. pekinensis (Chinese cabbage) and subsp. rapa (fodder and vegetable turnips). Ten accessions demonstrated complete resistance to pathotype 3, including Polish canola (‘96-6991’ and ‘96-6992’), Chinese cabbage (‘Bejo 2833’, ‘Bilko’, ‘Emiko’ and ‘Jazz Napa Cabbage’), pak choy (‘Flower Nabana’) and turnip (‘Taronda’, ‘Vedette’ and ‘Vollenda’). Some Polish canola and turnip accessions also showed moderate to high levels of resistance (50–93% clubroot reduction); a few accessions had consistently smaller galls than those observed on the CS control, but most segregated for resistance and susceptibility.

Table 2. Brassica accessions with >30% clubroot severity reduction relative to a susceptible control^a after inoculation with Plasmodiophora brassicae pathotype 3.

(Plant species) Cultivar or accession	Source (seed provider)^b	Clubroot reduction (%)
(Brassica rapa)
96-6991	SRC	100
96-6992	SRC	100
Bejo 2833	Bejo Seeds	100
Bilko	Bejo Seeds	100
Emiko	Bejo Seeds	100
Flower Nabana	Viterra	100
Jazz Napa Cabbage	Viterra	100
Taronda	PGRC	100
Vedette	PGRC	100
Vollenda	PGRC	100
96-6990	SRC	93
Siloga	PGRC	92
CN 111222	PGRC	82
CN 111189	PGRC	81
Nozawana	Viterra	81
96-6993	SRC	77
T38	SRC	77
T35	SRC	70
CN 111014	PGRC	61
T49	SRC	57
Purple Top White Globe	Viterra	57
CN 111004	PGRC	56
96-6994	SRC	56
Y.S.M.	Viterra	54
L-141P2	Viterra	54
Purple Top	Viterra	52
CN 110971	PGRC	50
22398-6 Sylvestris	SRC	48
96-6989	SRC	48
Feng Quing Choi	Viterra	48
CN 111214	PGRC	46
Shuka, Autumn Torch	Viterra	44
CN 111108	PGRC	44
CN 101909	PGRC	40
CN 111212	PGRC	39
CN 110998	PGRC	38
CN 110984	PGRC	38
CN 101916	PGRC	37
CN 111220	PGRC	36
CN 111225	PGRC	36
CN 111227	PGRC	36
CN 111237	PGRC	35
CN 111253	PGRC	35
Unknown	Viterra	35
Unknown	Viterra	35
Goldvital	PGRC	33
Komatsuna Red	Viterra	33
CN 101915	PGRC	32
CN 101934	PGRC	31
CN 111257	PGRC	30
CN 111286	PGRC	30
(Brassica napus)
Askegarde	Viterra	100^c
SW 02763	Viterra	78^c
Wilhelmsburger	HRI	44
(Brassica juncea)
Shi-Li-Hon	Viterra	48
Big Stem	Viterra	47
T.M. 12	Viterra	44
Red-Giant	Viterra	40
Bau-sin	Viterra	38
Garnet Giant	Viterra	36
San-Ho Gianty	Viterra	33
Chirimen Hakarashi	Viterra	32
(Brassica oleracea)
Tekila	Syngenta	100
Kilaherb	Syngenta	100
Pee Wee	PGRC	65
Gruner Angeliter	PGRC	59
CN 35413	PGRC	56
Primevert	PGRC	46
(Brassica nigra)
BRA 192/78	Viterra	100
CR 2120	IPK	100
CR 2716	IPK	100
PI 219576	Viterra	88

^a The resistance was based on the comparison of average clubroot severity of each entry (n = 7) with that of a susceptible canola cultivar (‘Fortune RR’) used in each trial. Under this condition, the commercial clubroot-resistant canola ‘45H29’ (Pioneer Hi-Bred Canada) was generally free of clubroot.

^b Sources of germplasm:

• PGRC, Plant Gene Resources of Canada, Agriculture and Agri-Food Canada, 107 Science Place, Saskatoon, Saskatchewan S7N 0X2, Canada;

• SRC, Saskatoon Research Centre, Agriculture and Agri-Food Canada, 107 Science Place, Saskatoon, Saskatchewan S7N 0X2, Canada;

• Bejo Seeds, Bejo Seeds Inc., 1972 Silver Spur Place, Oceano, California, USA 9344;

• HRI, Genetic Resources Unit, Warwick HRI, University of Warwick, Wellesbourne, Warwick CV35 9EF, UK;

• IPK, Leibniz Institute of Plant Genetics and Crop Plant Research, OT Gatersleben, Corrensstrasse 3, D-06466 Stadt Seeland, Germany;

• Syngenta, Syngenta Seeds UK Ltd.; CPC4, Capital Park, Fulbourn, Cambridge CB21 5XE, UK;

• Viterra, Viterra Inc., Research and Development, 201–407 Downey Road, Saskatoon, Saskatchewan S7N 4L8, Canada.

^c Against P. brassicae pathotype 2 only; not resistant to pathotype 3.

Four of the B. napus accessions showed low to moderate resistance (25–44% clubroot reduction) to pathotype 3, but the remainder were highly susceptible (data not shown). Ten B. juncea accessions also showed low to moderate resistance (28–48% clubroot reduction), with noticeably smaller galls than those on the CS control. However, none of these accessions had >50% clubroot reduction (Table 2) and most were highly susceptible (data not shown). Only limited numbers of B. carinata accessions were assessed in this study and all were susceptible or highly susceptible to pathotype 3 (data not shown). The majority of B. oleracea accessions were highly susceptible, but several showed moderate to high levels of resistance to pathotype 3 (Table 2). Most of the CR B. oleracea are cabbages, although ‘Gruner Angeliter’ is kale. Two commercial cabbages, ‘Kilaherb’ and ‘Tekila’, were highly resistant to pathotype 3, while others segregated for resistance and susceptibility. Four of the eight B. nigra accessions showed a high level of resistance to pathotype 3 (Table 2); ‘BRA 192/78’, ‘CR 2120’ and ‘CR 2716’ exhibited immunity, and two ‘PI 219576’ plants had only small galls while the remainder were clubroot-free.

Trial 2. Performance of selected CR accessions under higher pathogen inoculum pressure

Under 10-fold higher pathogen inoculum pressure, the 10 selected resistant accessions including B. rapa, B. oleracea and B. nigra originating from different parts of the world showed variable resistance patterns against P. brassicae pathotype 3 (Table 3). ‘Flower Nabana’, ‘Jazz Napa Cabbage’ and ‘Tekila’ were completely resistant with no evidence of disease, whereas ‘PI 219576’, ‘Purple Top’ fodder turnip and ‘Purple Top White Globe’ vegetable turnip segregated in an almost 1 : 1 ratio (resistant : susceptible). ‘PI 219576’ was highly resistant (although not immune) in Trial 1, but the two turnip accessions were only moderately resistant in the same trial. The remainder of the selected CR accessions appeared immune, but there were incidental plants that showed slight to severe (scale 1–3) clubroot symptoms.

Table 3. Clubroot resistance of individual plants of selected Brassica accessions inoculated with Plasmodiophora brassicae pathotype 3 (Trial 2).^a

Taxon	Type of plant	Cultivar or accession	Origin (country)^b	Trial 1 (% clubroot reduction)	Trial 2 (0–1–2–3)^c
B. nigra	Black mustard	BRA 192/78	Greece	100	51–0–0–2
B. nigra	Black mustard	PI 219576	Puerto Rico	88	21–0–0–20
B. oleracea var. capitata	Cabbage	Kilaherb	Unknown	100	54–1–0–1
B. oleracea var. capitata	Cabbage	Tekila	The Netherlands	100	54–0–0–0
B. rapa subsp. chinensis	Pak choy	Flower Nabana	USA	100	21–0–0–0
B. rapa subsp. pekinensis	Chinese cabbage	Bejo 2833	USA	100	30–2–5–5
B. rapa subsp. pekinensis	Chinese cabbage	Emiko	USA	100	54–0–0–1
B. rapa subsp. pekinensis	Chinese cabbage	Jazz Napa Cabbage	Japan	100	21–0–0–0
B. rapa subsp. rapa	Turnip (forage)	Purple Top	USA	52	42–6–4–40
B. rapa subsp. rapa	Turnip (vegetable)	Purple Top White Globe	USA	57	37–6–4–28

^a The concentration of pathogen inoculum was 1 × 10⁷ resting spores cc⁻¹ growth medium (10-fold higher than in Trial 1).

^b Country the accession was from, but the exact geographic origin of accession was unknown. The cabbage cultivars ‘Kilaherb’ and ‘Tekila’ were from Syngenta Seeds UK Ltd.

^c 0–1–2–3: The number of plants in each disease severity category (0–3).

Trial 3. Resistance spectrum of selected CR accessions

The 10 CR accessions evaluated in Trial 2 were also tested against the five P. brassicae pathotypes (2, 3, 5, 6 and 8) found in Canada. Throughout the study, the universally susceptible control ‘Granaat’ (ECD-05) developed severe clubroot symptoms (DSI ≥ 97.0%) after inoculation with each of the pathotypes (Table 4). ‘BRA 192/78’, ‘Kilaherb’, ‘Tekila’, ‘Flower Nabana’ and ‘Jazz Napa Cabbage’ were immune or highly resistant to each of these pathotypes, consistently reducing DSI by >90% relative to the CS control. For each of these accessions, there was little variation in resistance to the different pathotypes (P < 0.05, Fisher’s Protected LSD). ‘Emiko’ and ‘Bejo 2833’ appeared slightly less resistant than the above-mentioned accessions, but their reaction to the different pathotypes did not differ (P < 0.05, Fisher’s Protected LSD). ‘PI 219576’, ‘Purple Top’ and ‘Purple Top White Globe’, however, were only moderately resistant to these pathotypes, reducing DSI by 44–65% (Table 4). These accessions did show slight variation in reaction to the different pathotypes; ‘PI 219576’ was slightly less resistant to pathotype 3 than to pathotypes 2 or 5, and ‘Purple Top White Globe’ was slightly less resistant to pathotype 3 than to pathotypes 2 or 8 (P < 0.05, Fisher’s Protected LSD). In contrast, ‘Purple Top’ was more resistant to pathotypes 2 and 3 than to pathotype 8 (P < 0.05, Fisher’s Protected LSD). There was no clear pattern of differential interaction against a specific pathotype between different Brassica species.

Table 4. Disease severity index (%) on selected Brassica accessions in response to inoculation with single-spore isolates of the five Plasmodiophora brassicae pathotypes identified in Canada (Trial 3).

Type of plant	Cultivar or accession	P-type 2^a	P-type 3	P-type 5	P-type 6	P-type 8
Chinese cabbage	Granaat^b	97.0 e	98.6 f	98.9 d	97.7 e	97.7 e
Black mustard	BRA 192/78	0 a	0 a	0 a	1.4 a	0 a
Black mustard	PI 219576	40.7 d	51.9 e	38.5 c	45.8 d	44.0 d
Cabbage	Kilaherb	0 a	0 a	0 a	0 a	2.3 ab
Cabbage	Tekila	0 a	1.4 ab	0.9 a	3.3 ab	0 a
Chinese cabbage	Bejo 2833	3.2 b	15.7 cd	17.6 b	16.2 bc	10.2 bc
Chinese cabbage	Emiko	16.7 c	18.1 d	25.0 b	25.5 c	19.0 c
Pak choy	Flower Nabana	5.7 b	5.1 bc	1.4 a	1.9 a	4.2 ab
Chinese cabbage	Jazz Napa Cabbage	0.9 a	0.9 ab	2.3 a	0.9 a	2.3 ab
Turnip (forage)	Purple Top	34.7 d	43.1 e	45.8 c	55.1 d	55.6 d
Turnip (vegetable)	Purple Top White Globe	43.1 d	53.7 e	49.1 c	47.7 d	39.8 d

^a Means (n = 6) within a column followed by the same letter(s) do not differ (P < 0.05, Fisher’s Protected LSD).

^b Universally susceptible control.

Discussion

Previous reports of screening for clubroot resistance focused primarily on pathotypes prevalent on other continents (Tjallingii 1965; Toxopeus & Janssen 1975; Dias et al. 1993; Voorrips & Visser 1993; Kopecký et al. 2012) or on vegetable crops in North America (Ayers & Lelacheur 1972; Chiang & Crête 1985; Miller & Williams 1986). One recent study targeted pathotypes on canola in western Canada (Hasan et al. 2012) and assessed genotypes from the Brassica A, B and C genomes and their corresponding amphidiploids. The number of accessions evaluated in each group ranged from 24–77, with most coming from B. nigra (B genome). The current study focused on B. rapa because many of the highly resistant genotypes reported previously were in this species (Hirai 2006) and there was a large collection of B. rapa accessions accessible from Plant Gene Resources of Canada, Agriculture and Agri-Food Canada. Overall, only a small number of the 955 Brassica accessions tested in the current study were highly resistant to pathotype 3, but these CR accessions represented A-, B- and C-genome and their amphidiploid species. Genetic variation in P. brassicae was reported first by Honig (1931) based on the interaction of three crucifer species against P. brassicae isolates of different origin. More studies (Williams 1966; Buczacki et al. 1975; Somé et al. 1996) were conducted later to elucidate physiological specialization in P. brassicae, but the races of P. brassicae are yet to be clearly defined. In the current study, five P. brassicae pathotypes found in Canada and characterized using the Williams (1966) differential set by Strelkov et al. (2007) were used.

Most of the highly resistant accessions were found in B. rapa in the current study, with many of them being turnips. A few accessions of oriental crucifer vegetables like pak choy and Chinese cabbage also showed strong resistance, but the origin of clubroot resistance in these accessions is not known. In previous studies, highly resistant B. rapa also showed less variability to different P. brassicae pathotypes than other CR Brassica species (Toxopeus et al. 1986). European fodder turnip was particularly resistant to clubroot (Diederichsen et al. 2009) and some believed that clubroot resistance genes in B. rapa might exist only in European turnips (Crute et al. 1980; Yoshikawa 1993). Yoshikawa (1993) questioned whether some or all of the clubroot resistance genes in oriental vegetable cultivars were from turnip. In our study, many turnip accessions were only moderately resistant to pathotype 3, while the CR oriental vegetables were nearly immune. Further study of ‘Purple Top’ (fodder) and ‘Purple Top White Globe’ (vegetable) showed that the lower resistance of these two turnips was more due to their genetic heterogeneity, with clear segregation for resistance and susceptibility when tested in larger populations (Table 3). As a typical cross-pollinated species, most B. rapa populations are expected to be genetically heterogeneous. The CR oriental vegetable accessions tested are all hybrid cultivars and despite their high genetic heterogeneity, their parental lines might be relatively homogeneous and the clubroot resistance genes they carry are likely dominant since they resulted in uniform resistance. The origin and similarity of clubroot resistance genes in these B. rapa accessions are still being determined by molecular mapping (Yu et al. 2013); nevertheless, several of them are good sources of resistance to the predominant P. brassicae pathotype in western Canada. The turnips ‘Siloga’, ‘Taronda’, ‘Vedette’ and ‘Vollenda’ were also highly resistant to pathotype 3. Detailed assessments were not carried out on these accessions due to poor seed viability, but Hasan et al. (2012) reported that all five of the turnip accessions they tested were highly resistant to the five P. brassicae pathotypes found in Canada. It appears that turnips carry useful clubroot resistance genes that are effective against all the P. brassicae pathotypes found in Canada.

Most of the B. napus accessions showed little resistance to pathotype 3. Even the rutabaga ‘Wilhelmsburger’, which was resistant to many P. brassicae isolates from France (Somé et al. 1996), showed only moderate resistance (Table 2). Similar results were also observed with the European rutabaga ‘Askegarde’, as well as the spring canola line ‘SW 02763’, which has clubroot resistance genes from ‘Askegarde’. These two accessions were susceptible to pathotype 3, but highly resistant to pathotype 2 (Table 2). It appears that clubroot resistance genes in rutabaga, including ‘Askegarde’ and ‘Wilhelmsburger’, are pathotype-specific and ineffective or only moderately effective against pathotype 3 in western Canada. Strelkov et al. (2006) reported pathotypes 3 and 5 were more virulent on canola and 48 canola cultivars evaluated were all highly susceptible to pathotype 3. Hasan et al. (2012) reported that only three of the 36 B. napus accessions tested were resistant to pathotype 3 and one was resistant to pathotype 5. Canola cultivars resistant to pathotypes 3 and 5 have been available in western Canada since 2009 (LeBoldus et al. 2012), and some of them might have received strong clubroot resistance genes from other sources, especially B. rapa, as evidenced by the development of the rapeseed cultivar ‘Mendel’ (Diederichsen et al. 2006).

Clubroot resistance was also identified among the B. nigra and B. oleracea accessions tested. There was a higher proportion of B. nigra accessions with a high level of resistance relative to the other species tested, although the sample size was small (eight accessions). Clubroot resistance in B. nigra was reported in the 1980s and transferred into oilseed rape (B. napus) through asymmetric somatic hybridization (Sacristán et al. 1989; Gerdemann-Knörck et al. 1995). In a recent study by Hasan et al. (2012), about 86% of the 77 B. nigra accessions were resistant to all P. brassicae pathotypes found in Canada. Three of the eight B. nigra accessions in our study showed complete resistance to pathotype 3, and ‘BRA 192/78’ also was resistant to all pathotypes found in Canada. ‘PI 219576’ was noticeably heterogeneous, with clear segregation where plants were either disease-free or had severe clubroot symptoms (Table 3). This segregation pattern was generally consistent for each of the pathotypes found in Canada (Table 4), with disease-free plants obtained in each case. These results highlight the additional sources of clubroot resistance that are effective against all five P. brassicae pathotypes found in Canada.

Previously, Crisp et al. (1989) evaluated approximately 1000 B. oleracea accessions and confirmed the existence of clubroot resistance in several European kales and cabbages, with a new source of clubroot resistance identified in the cabbage ‘Eire’. Based on the test of ECD hosts, Crute et al. (1983) suggested that CR B. oleracea would more likely be overcome by the P. brassicae isolates that were not virulent on the B. rapa and B. napus hosts used in the ECD set. The Brassica C-genome group is thought to carry clubroot resistance genes less frequently than the other species (Piao et al. 2009); alternatively, the genes in B. oleracea might confer a lower level of resistance (Diederichsen et al. 2009). Clubroot resistance genes have also been introduced into B. oleracea from B. napus (rutabaga) by interspecific hybridization (Chiang et al. 1977), but it is difficult to determine the origin and similarity of these genes in B. oleracea because of generally poor pedigree records for most of the original sources of resistance. Several CR cabbages developed in Canada were resistant to pathotypes 1, 2, 6 and 7 (Chiang et al. 1977; Chiang & Crête 1985, 1989), but their resistance to pathotypes 3 and 5, which are more relevant to clubroot on canola in western Canada, was not reported. Of the 48 B. oleracea accessions tested by Hasan et al. (2012), five showed resistance to pathotype 3 and three showed resistance to pathotype 5. In our study, five of the 30 B. oleracea accessions tested were resistant to pathotype 3, with the cabbages ‘Kilaherb’ and ‘Tekila’ being highly resistant to all five pathotypes found in Canada. These findings highlight potentially highly effective sources of clubroot resistance from C-genome species against pathotypes 3 and 5 on canola.

No highly resistant accessions of the amphidiploid species B. juncea or B. carinata were identified even though a relatively large number of B. juncea accessions was screened. Several accessions showed moderate resistance, but root galls (scale 1–2) were always present on these plants and plant growth was often negatively affected (data not shown). These results are consistent with previous observations where clubroot resistance was not reported in B. juncea or B. carinata (Diederichsen et al. 2009). Hasan et al. (2012) reported that none of the 48 B. juncea or 24 B. carinata accessions they tested carried resistance to any of the pathotypes found in Canada. They postulated that if CR B. nigra had been involved in the evolution of B. juncea or B. carinata in nature, it would be possible that the B-genome resistance in these two amphidiploids is hypostatic. It is difficult to explain the lack of resistance in B. juncea and B. carinata since clubroot resistance genes are common in the B-genome species (Hasan et al. 2012). We phenotyped a small F₁ population (30 plants) derived from a cross between a CS doubled haploid B. carinata line and CR ‘BRA 192/78’ (B. nigra) and observed an almost 1 : 1 segregation ratio (resistant : susceptible) for reaction to pathotype 3 (data not shown). This may indicate that resistance to pathotype 3 in the heterozygous ‘BRA 192/78’ is controlled by a single dominant gene.

This is one of the largest studies to date attempting to identify diverse sources of clubroot resistance for the development of CR canola germplasms, thereby enhancing the effectiveness and durability of clubroot resistance in canola crops in western Canada. The screening of the B. rapa group is extensive, and included a variety of crop types with a large number of CR accessions identified. The 10 selected CR accessions, representing Brassica A-, B- and C-genome species, appear promising with highly resistant plants obtained from each source that are effective against all five pathotypes of P. brassicae found in Canada. Further studies to characterize clubroot resistance genes in the selected accessions and to incorporate complementary clubroot resistance genes into canola breeding lines have been planned.

Acknowledgements

We thank Michelle Francisco, Alana MacDonald, Robert Laprairie, Lucas Hart, Jocelyn Reeve, Jeffrey Albert, Danielle Head, Terrie Tran, Rizza Rayes, Christie McGregor, Chantal del Carmen, Nicolas Dimopoulos and Laura Kessler for technical support. Dr M.R. McDonald, University of Guelph, provided Syngenta cabbage hybrid seed ‘Kilaherb’ and ‘Tekila’. Financial support from the Saskatchewan Agriculture Development Fund (Project #20090359) is acknowledged.