Reaction of clubroot-resistant genotypes of Brassica rapa, Brassica napus and Brassica oleracea to Polish Plasmodiophora brassicae pathotypes in laboratory tests

The Brassica genotypes selected for the experiments were previously found to be resistant to various Plasmodiophora brassicae pathotypes (Pb). Their interaction with pathotypes Pb2, Pb3 and Pb9 isolated in Poland was studied, using macroscopic observation for the presence of root galls, microscopic observations of P. brassicae plasmodia in the root hairs and quantitative PCR for determination of the pathogen’s quantity in plant roots and growing media. Of the Brassica genotypes studied, only B. rapa var. capitata line AABBcc was fully resistant to all the Polish pathotypes of P. brassicae. Some of the other “clubroot-resistant” genotypes tested were resistant to selected pathotypes, e.g. Brassica napus var. rapifera ‘Wilhelmsburger’ to Pb 2 and Pb3, Brassica oleracea var. capitata ‘Kilaton F1’ to Pb2 and Brassica rapa subsp. pekinensis ‘Bilko F1’ to Pb3, but were susceptible to others. B. oleracea var. capitata ‘Bindsachsener’, B. oleracea var. acephala subvar. lacinata ‘Verheul’ and B. napus var. napus ‘Mendel F1’ were moderately to highly susceptible to all Polish P. brassicae pathotypes. These results show that the classification of virulence of P. brassicae pathotypes selected in various areas differs significantly toward the same Brasssica genotypes and puts in question the practical value of pathotype classification determined with differential sets for farmers and plant breeders. Our results showed that B. rapa var. capitata AABBcc line, B. napus var. rapifera ‘Wilhelmsburger’, B. oleracea var. capitata ‘Kilaton F1’, B. rapa subsp. pekinensis ‘Bilko F1’ could be considered in Brassica breeding programmes as a source of resistance to Polish P. brassicae isolates.


Introduction
Clubroot is a very dangerous disease affecting most of Brassica species and many other cruciferous plants. Its causative agent is the obligate parasite Plasmodiophora brassicae Wor. (Dixon 2009). The pathogen has a complex life cycle comprising three stages: 1) root hair infection by resting spores present in soil, 2) cortical infection, and 3) formation of resting spores and their release to the soil from decaying root galls (Kageyama and Asano 2009;McDonald et al. 2014). The primary phase of the life cycle of P. brassicae (root hair infection) occurs in both susceptible and resistant cultivars of host species, and even in non-host plants (Deora et al. 2013). However, the infection rate and further development of the pathogen in root hairs differ among Brassica species, with development occurring quicker in susceptible lines and slower in the resistant ones. Characteristic disease symptoms appear on roots in the form of irregular galls, initially small, light yellow and fragile and then dark and decaying. Diseased tissues lose their ability to uptake water and nutrients from the soil, which results in wilting and stunting of the above ground part of plants (Strelkov et al. 2006). Decaying diseased roots release resting spores into the soil, where they can remain viable for at least 20 years (Dixon 2009).
The disease causes severe losses on cruciferous vegetable and rapeseed plantations in Poland and worldwide (Robak 1991;Wallenhammar 1996;Dixon 2014). Depending on soil conditions, infestation rate, pathogenicity of P. brassicae local strains and susceptibility of a crop's variety, yield losses can range from 30% to even 100% (Wallenhammar 1996). In control of the disease, the most important are preventive treatments including usage of pathogen-free fields, early pathogen detection, non-host crop rotation, use of the bait, decoy or catch crops, maintaining the appropriate soil pH and soil drenches with fungicides (Murakami et al. 2000;Peng et al. 2014;Hwang et al. 2015). However, these methods would not reduce clubroot severity in heavily infested fields and they are not cost effective. Besides, soil drenches with fungicides have a very negative impact on the environment. Thus, the use of resistant varieties is considered as the most efficient method for control of the disease (Rahman et al. 2014). However, the choice of resistant cultivars is limited and the varietal resistance can be easily broken due to high pathogenic variability of P. brassicae (Diederichsen et al. 2014). Therefore, resistance to clubroot is one of the mayor tasks in Brassica breeding programmes.
P. brassicae consists of numerous physiological pathotypes (races), differing in their ability to infect specific host genotypes (Robak 1991). Several sets of plant hosts have been proposed to determine these pathotypes (Williams 1966;Buczacki et al. 1975;Strelkov and Hwang 2014). The most commonly used is European Clubroot Differential Set (ECD), developed by Buczacki et al. (1975), which consist of three subsets: B. rapa (five hosts), B. napus (five hosts) and B. oleracea subset (five hosts) and allows to select different pathotypes of P. brassicae. The differential set proposed by Williams (1966) uses four differential genotypes: B. napus (two host) and B. oleracea (two host), which can differentiate sixteen pathotypes. The pathogenicity of P. brassicae pathotypes is highly variable within and between field populations. According to Robak (1991), in Poland the most common are pathotypes 2, 3 and 9.
In clubroot resistance breeding programs it is of a great importance to evaluate the available sources of resistance. There are several resistant Brassica species and cultivars harbouring resistance genes, but their resistance may vary depending on a pathotype of the pathogen. The aim of this study was: 1) to evaluate the degree of resistance of several Brassica genotypes carrying various resistance genes to Polish isolates of P. brassicae by the use of micro-and macroscopic observation of the disease symptoms and molecular tests of the presence of spores in roots of the tested plants; and 2) to determine whether cultivation of selected plants can lead to the reduction of the pathogen's quantity in the soil.

Plant material
The experiments were done on five genotypes of B. oleracea, B. napus and B. rapa, selected from European Clubroot Differential Set, and on three commercial cultivars. These genotypes have been found to be highly resistant to several P. brassicae pathotypes (Kuginuki et al. 1999;Diederichsen et al. 2006;Strelkov et al. 2006;Cebula et al. 2010;Cao et al. 2014) and are recommended as a source of resistance traits in breeding (Hasan et al. 2012). The Chinese cabbage (B. rapa ssp. pekinensis) cv. Granaat, extremely susceptible to all known pathotypes of clubroot (Kopecký et al. 2012), was used as a negative control for resistance to clubroot. The list of the genotypes used in the study is presented in Table  Source of the P. brassicae isolates Infected root samples, with visible galls, were collected from the commercial plantations of Brassica oleracea var. capitata cv. Kamienna Głowa located in various parts of Poland. The samples were washed, cleaned, air dried and catalogued. The pathotypes of the isolates were determined as P. brassicae (Pb) 2, 3 and 9 based on the reaction of the four genotypes included in differential set of Williams (1966) to the pathogen (Table 2).

Inoculum preparation
Spore suspension was prepared as described by Voorrips (1992). The clubbed roots were homogenized with distilled water (1:3, w:v) at high speed in a mechanical blender. The suspension was filtered four times through 5 layers of cheesecloth. The spore concentration in the filtrate was quantified in haemocytometer and was adjusted to approximately 10 8 spores per mL through dilution with sterile distilled water. Obtained spore suspension was stored at 4°C until use. To propagate these pathotypes for further studies, seeds of the susceptible B. rapa cv. Granaat were sown in growing medium inoculated with resting spores (10 7 spores per mL of the medium) of each P. brassicae pathotype and grown in a greenhouse under 14-h light/10-h dark photoperiod at 22°C. After 5 weeks of growth in the inoculated medium, the plants were dug out, cleaned and then the galls were separated and stored at −20°C until use.
Glasshouse bioassay of P. brassicae resistance -Macroscopic observation The growing medium used in the tests was prepared by mixing the peat substrate (Klasmann) at pH 5.2 with the low peat (pH 5.9) in 1:1 proportion (v:v), and then sterilizing at 100°C for 20 min. After 2 weeks, the resting spore suspension of P. brassica pathotypes Pb2, Pb3 and Pb9 was added to growing medium in quantity to obtain final spore concentration of about 10 7 spores per mL and mixed thoroughly. The experiments were carried out in plastic trays filled with 2 L of inoculated growing media. The trays filled with noninoculated medium served as a control. 40 seeds from each genotype were seeded into the trays in three replicates. Plants were grown in a glasshouse under natural light, photoperiod 14-h/10-h and temperature 22°C. The growing medium moisture was maintained by regular watering. 10, 20 and 35 days after sowing the sample consisted of 10 randomly selected plants was uptaken, the roots were washed under the running water and blotted on paper towel. The occurrence of root galls was assessed individually on each plant on a 0-3 scale according to Buczacki et al. (1975) and Kuginuki et al. (1999), where: 0no symptoms visible on roots, 1single galls on lateral roots, 2numerous galls on lateral and/or main roots, 3distortion of the entire root system, uniform galls on the main root. The trials were conducted as a complete randomized block design with three replications per a treatment. For where: n is the number of plants in each class; N is the total number of plants; 0, 1, 2 and 3 are the symptoms severity indices. All roots were then packed in plastic bags and stored at −70°for further qPCR analysis. The bioassays were conducted twice. qPCR analyses of P. brassicae in plant roots and growing medium The research was carried out on the roots obtained from plants grown for 10, 20 and 35 days, as described above. In the case of plants growing in non-inoculated growing medium (control), the samples were collected only once, after 35 days. The samples of growing media were collected twice, before sowing and after the experiment was completed. Before DNA extraction, root samples were thoroughly washed with distilled water and blotted on paper towel. Both the roots and growing media were pulverized in liquid nitrogen with mortar and pestle. DNA extraction from plant material was made using DNeasy Plant Mini Kit (Qiagen) and from soil samples using NucleoSpin Soil Kit (Macherey Nagel), according to manufacturer's instructions.
Real-time PCRs were performed using PbF and PbR primers and fluorescently labelled PbP probe specific for P. brassicae (Wallenhammar et al. 2012) (Table 3). All qPCR amplifications were conducted using a LightCycler 480 Probe Master (Roche) in 25 μL volume containing 5 μL template DNA, 0.45 μL primer F, 0.15 μL primer R, 0.5 μL probe, 12.5 μL Master Mix and 6.4 μL H 2 0. DNA extracts were diluted five-fold before analysis and were analysed in triplicate. All reactions were run at 95°C for 10 min followed by 45 cycles at 95°C for 15 s and finally 59°C for 1 min. The standard curve for calculating the number of P. brassicae genome copies in analysed samples was constructed using plasmid carrying the P. brassicae target sequence as a template in qPCR, prepared according to Wallenhammar et al. (2012). A 10-fold dilution series of the plasmid standard were analysed in triplicate using the real-time PCR assay and Ct values were then plotted against the amount of plasmid DNA to create a standard curve. The pathogen is present in plants and soil in various development forms. To simplify the further calculations, the results of quantitative determination of its DNA copy numbers were presented as equivalents of resting spores, as described by Wallenhammar et al. (2012).
Microscopic observation of P. brassicae plasmodia in root hair The plants were grown in Phytotoxkit test plates (Tigret sp. z o. o.) filled with growing medium inoculated with P. brassicae resting spores at density of 10 7 per mL. Ten seeds of each plant genotype studied were sown per one plate in three replications and germinated in growing chamber (Sanyo) with relative humidity of 60%, at 22°C in the light for 14 h and at 20°C in the dark for 10 h. Microscopic observations of plants roots were conducted using Olympus BX 41 microscope from 4th to 20th day after sowing. The roots were carefully extracted from the growing medium, washed several times in sterilized water and cut into 1 cm fragments. Three fragments of each root were examined, beginning from the fragments situated nearest of the seedling's stem. The number of plasmodia containing mature, spherical zoosporangia was estimated in 1 cm fragments of roots. Average number and maximum number of plasmodia was estimated in 18 pieces of root hairs for each genotype. The experiment was conducted twice for each P. brassicae pathotype and each plant genotype.

Data analysis
The data were analysed by standard statistical procedure. Significance of differences between means was established by one-way analysis of variance and Newman-Keuls test at P < 0.05.

Glasshouse resistance bioassay -Macroscopic observations
The first galls on roots appeared 20 days after sowing on the susceptible Chinese cabbage 'Granaat' growing in media inoculated with P. brassicae pathotypes 2 and 9 and on resistant swede 'Wilhelmsburger' and kale 'Verheul' growing in media inoculated with Pb9 and on Chinese cabbage 'Bilko F 1 ' growing in media inoculated with Pb3 and Pb9 ( qPCR analyses of P. brassicae in plant roots and growing medium The pathogen was present in the roots of all the tested genotypes grown in media inoculated with pathotypes 2, 3 and 9 on all sampling times (  (Table 5). The level of spores decreased between 10th and 35th day after sowing in the roots of turnip line AABBcc, kale 'Verheul' and headed cabbage 'Kilaton F 1 ', but increased in the roots of remaining genotypes. More complicated was the time course of root infection in the case of Pb9. Analysis of the roots collected after 20 days from sowing showed a decrease of spore numbers in reference to quantity of spores in the roots collected at the first sampling date, but after additional 15 days of growth the quantity of spores increased, in some genotypes even exceeding the quantity detected in the roots after 10 days of growing (Table 5). At the beginning of the experiment, in all the tested trays there was about 10 7 resting spores in 1 g of growing medium (Table 6). After 5 weeks of growth of susceptible Chinese cabbage 'Granaat' the highest number of resting spores was observed in growing media infested with all the pathotypes tested. Cultivation of the other tested genotypes contributed to the increase of the number of spores only in the growing media inoculated with pathotypes 2 and 9, which was especially visible in case of turnip line AABBcc and headed cabbage 'Bindsachsener' grown in the medium infected by Pb2 as well as in case of swede 'Wilhelmsburger' and rapeseed 'Mendel F 1 ' grown in medium infected by Pb9. It was noted that the quantity of spores of Pb3 decreased after 35 days of growth of all the genotypes in the infested media. The largest decrease was observed in Pb3 inoculated medium where rapeseed 'Mendel F 1 ' was cultivated.
Microscopic observations of P. brassicae plasmodia in root hairs of the genotypes tested The first plasmodia in hair roots appeared within 4-10 days from sowing the seeds (Fig. 2). The average number of plasmodia in the observed fragments of the roots depended on the plant genotype and on the P. brassicae pathotype. For the Pb2, the highest average and maximum number of plasmodia was observed in root hairs of turnip line AABBcc and headed cabbage 'Kilaton F 1 '. The lowest average number of plasmodia was observed in root hairs of kale 'Verheul' and the lowest maximum number of plasmodia was recorded in swede 'Wilhelmsburger'. In the case of Pb3 the highest average and maximum number of plasmodia was observed in root hairs of kale 'Verheul'. Also, high maximum number of plasmodia in root hairs was observed in Chinese cabbage 'Bilko F 1 '. The lowest number of plasmodia was observed in case of swede 'Wilhelmsburger'. For Pb9, the highest average and maximum number of plasmodia was observed in root hairs of rapeseed 'Mendel F 1 '. Similarly, high average and maximum numbers of plasmodia per 1 cm of root were observed in turnip line AABBcc (Table 7). The lowest number of plasmodia was observed in case of kale 'Verheul'.

Discussion and conclusions
Resistance to clubroot is one of the most important objectives in Brassica breeding programmes. The main problems in this task is large variations of pathogenicity of Plasmodiophora brassicae, which can quickly break the resistance of cultivars intended for planting on large areas (LeBoldus et al. 2012;Tanaka and Ito 2013;Diederichsen et al. 2014). Therefore, in clubroot resistance breeding programmes it is of a great importance to evaluate the available sources of resistance to most, or at least to most of locally occurring pathotypes of Means followed by the same letter for time of sampling within a P. brassicae race category do not differ at P < 0.05 based on Newman-Keuls test P.brassicae and apply gene pyramiding approach. In Poland, research on clubroot resistance was concentrated on vegetable Brassicas and on winter oilseed rape, where disease causes considerable losses (Robak and Gidelska 2009;Kurowski et al. 2009), among others due to high infestation of local soils by P.brassicae (Kopecký et al. 2012;Jędryczka et al. 2014).
The genotypes selected for the study have been found in other researches as resistant to a number of P. brassicae pathotypes, but our results indicate that in most of them resistance was broken down by the Polish isolates of the pathogen. Only genotypes with ID of less than 20 are considered as highly resistant (Chytilová and Dušek 2007). In our experiments only B. rapa var. capitata line AABBcc (ECD 03) fulfilled this criterion in reference to all three P. brassicae pathotypes isolated in Poland. High resistance of this genotype confirms earlier results of Kuginuki et al. (1999), Donald et al. (2006) and Hasan et al. (2012).
Some of the other genotypes tested were resistant only to selected pathotypes, e.g. B. napus var. rapifera 'Wilhelmsburger' to Pb2 and Pb3, B. oleracea var. capitata 'Kilaton F 1 ' and B. rapa subsp. pekinensis 'Bilko F 1 ' to Pb2. High resistance of B. napus var. rapifera 'Wilhelmsburger' to P. brassicae pathotype 3 found in our experiments confirms earlier results of Kuginuki et al. (1999), Donald et al. (2006) and Hasan et al. (2012). But research partially contradicts results of other studies. Headed cabbage 'Kilaton F 1 ' in our experiments was fully resistant to Pb2 infection, but was highly susceptible to Pb3 and Pb9 (IDs 68 and 72, respectively), but in the threeyear field studies done by in Polish conditions by Cebula et al. (2010) this cultivar was fully resistant to clubroot. However, they have not determined the pathotype(s) of P. brassicae infesting the experimental plots. B. oleracea var. acephala 'Verheul', which according to Cao et al. (2014) was highly to moderately resistant to Pb3, in our experiments was highly to moderately susceptible to all the pathotypes tested (IDs 71, 46 and 40 in reference to Pb2, Pb3 and Pb9, respectively). Also B. napus var. napus 'Mendel F 1 ', which carries a major dominant gene and two recessive genes conferring clubroot resistance and, according to Diederichsen et al. (2006) and Rahman et al. (2014), is fully resistant to majority of P. brassicae pathotypes dominant in Europe, in our research it was moderately or highly susceptible to Polish isolates of Pb3 and Pb9 (IDs 32 and 100, respectively). This shows that the pathotype of P. brassicae isolates collected in various areas and determined using differential sets does not necessarily reflect its actual level of virulence toward various Brasssica genotypes. Thus, prior to the introduction for cultivation of new resistant Brassica species/ cultivars or selecting them as a source of resistance traits in breeding programmes their resistance/susceptibility to local Plasmodiophora brassicae isolates shall be tested before hand, irrespective of the pathotypes of the pathogen determined elsewhere with differential set. In the available literature there is no information about resistance of the genotypes tested in our study to P. brassicae pathotype 9. According to our best knowledge this is the first such a report.
Our research shows that of seven clubroot-resistant genotypes of Brassica evaluated, only B. rapa var.
capitata line AABBcc, B. napus var. rapifera 'Wilhelmsburger', B. oleracea var. capitata 'Kilaton F 1 ' and B. rapa subsp. pekinensis 'Bilko F 1 ' can be recommended as a source of clubroot resistance in Brassica breeding programmes or for cultivation in low-to moderately-infested soils in Poland.
In our study, the primary root infection was not correlated with plant's resistance to clubroot. In fact, the root hairs of B. rapa var. capitata line AABBcc, which was resistant to all the pathotypes tested, had the highest population of P. brassica plasmodia of all the genotypes tested. This confirms the findings of Kageyama and Asano (2009) who suggested that susceptibility of root hairs to It was noted that the average number of plasmodia in the root depended both on the plant genotype and on the pathotype P. brassicae. In the case of pathotype 2 the appearance of mature plasmodia containing spherical sporangia was observed earlier than in the case of Pb3 and Pb9. Morgner and Sacristán (1995) also reported colonization of c l ub r o o t -r e s i s t a n t B r a s s i c a g e n o t y p e s b y P. brassicae. In their quantitative study pathogen structures were observed 12 days after inoculation of B. napus and B. oleracea. Voorrips (1992) demonstrated large differences in the levels of primary root hair infection among thirteen host genotypes with varying levels of resistance, however did not observe correlation between resistance to root hair infection and resistance to clubroot. According to this author, only complete resistance to root hair infection would confer resistance to clubroot. It is widely believed that the clubroot incidence can be limited by planting the bait, decoy or catch plants, which can stimulate germination of resting spores and thus reduce their quantity of in the soil (Yamagishi et al. 1986;Murakami et al. 2000). According to Ikegami (1985), cultivation of clubroot-resistant Japanese radish (Raphanus sativus) during 5 years resulted in a considerable decrease of resting spore populations in the soil. Resting spores population was also slightly reduced after two cycles of other cruciferous nonhost crops (Ahmed et al. 2011). However, in our research, cultivation of the clubroot-resistant genotypes contributed to the increase of the number of spores in the growing media inoculated with pathotypes 2 and 9. These results suggest that the cultivation of resistant cultivars in heavily infected areas does not guarantee a reduction of the soil infestation with P. brassicae resting spores shall not be recommended as a measure to control clubroot.
Acknowledgements Special thanks to Dr. Piotr Kamiński from Department of Genetics and Breeding Vegetable Plants Research Institute of Horticulture for sharing seeds of selected Brassica genotypes.
This work was supported by the Ministry of Agriculture and Rural Development within Basic research for the biological progress in plant production programme -Task 99.
Compliance with ethical standards This article does not contain any studies with human or animal subjects performed by any of the authors.
Conflict of interest All authors listed have contributed to the work, have read the manuscript and declare that there are no potential conflicts of interest.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons. org/licenses/by/4.0/.