Expression and Role of Response Regulating, Biosynthetic and Degrading Genes for Cytokinin Signaling during Clubroot Disease Development

The obligate biotroph Plasmodiophora brassicae causes clubroot disease in oilseeds and vegetables of the Brassicaceae family, and cytokinins play a vital role in clubroot formation. In this study, we examined the expression patterns of 17 cytokinin-related genes involved in the biosynthesis, signaling, and degradation in Chinese cabbage inoculated with the Korean pathotype group 4 isolate of P. brassicae, Seosan. This isolate produced the most severe clubroot symptoms in Chinese cabbage cultivar “Bullam-3-ho” compared to three other Korean geographical isolates investigated. BrIPT1, a cytokinin biosynthesis gene, was induced on Day 1 and Day 28 in infected root tissues and the upregulation of this biosynthetic gene coincided with the higher expression of the response regulators BrRR1, on both Days and BrRR6 on Day 1 and 3. BrRR3 and 4 genes were also induced during gall enlargement on Day 35 in leaf tissues. The BrRR4 gene, which positively interact with phytochrome B, was consistently induced in leaf tissues on Day 1, 3, and 14 in the inoculated plants. The cytokinin degrading gene BrCKX3-6 were induced on Day 14, before gall initiation. BrCKX2,3,6 were induced until Day 28 and their expression was downregulated on Day 35. This insight improves our current understanding of the role of cytokinin signaling genes in clubroot disease development.


Introduction
Clubroot is a devastating disease of Brassica vegetables and oilseed crops. This infectious disease is caused by the obligate parasite, Plasmodiophora brassicae Woronin. This pathogen is a Cercozoan biotrophic protist of Plasmodiophoraceae family. Clubroot disease is a serious problem of oilseed rape growers throughout the world [1]. The pathogen is variable among geographic regions and therefore regional difference in severity of this disease is reported [2]. According to pathogenic reaction of 12 geographical isolates and the reactions of those isolate to Williams' differential set the Korean P. brassicae isolates were divided into four pathotypes [2,3]. In addition, the intron 1 of 18S smaller subunit of those isolates were found variable after ribosomal DNA sequencing [4].
The life cycle of P. brassicae is complex ( Figure 1) [5,6]. This pathogen is difficult to control because of the durability of spores and lack of suitable chemical control [1,7,8]. Despite the complex life cycle, two distinct phases of infection to its Brassica hosts are noted, primary and secondary phases of infection [9,10]. The way of infection and types of roots that are infected differ between primary and secondary phases of infection [11,12]. Primary infection starts from the root hairs and this phase usually lasts first three days (Figure 1). In this phase, the resting spores residing in soil germinates, infects root hairs and releases secondary zoospores. In the secondary phase of infection, the pathogen enters into the cortex of the roots and steles of roots and hypocotyls of the infected plants. When the secondary infection process continues root tissues divide abnormally and that results in rapid swelling in the root zone which is termed as "galls" or "clubs" (Figure 1) [9,11]. The life cycle of P. brassicae is complex ( Figure 1) [5,6]. This pathogen is difficult to control because of the durability of spores and lack of suitable chemical control [1,7,8]. Despite the complex life cycle, two distinct phases of infection to its Brassica hosts are noted, primary and secondary phases of infection [9,10]. The way of infection and types of roots that are infected differ between primary and secondary phases of infection [11,12]. Primary infection starts from the root hairs and this phase usually lasts first three days (Figure 1). In this phase, the resting spores residing in soil germinates, infects root hairs and releases secondary zoospores. In the secondary phase of infection, the pathogen enters into the cortex of the roots and steles of roots and hypocotyls of the infected plants. When the secondary infection process continues root tissues divide abnormally and that results in rapid swelling in the root zone which is termed as "galls" or "clubs" (Figure 1) [9,11].  [13] after Dekhuijzen [14], Müller and Hilgenberg [15]). dpi, days post inoculation.
It is generally believed that an alteration in the usual hormonal balance has a vital role in the clubroot formation [13,16,17]. The contribution of hormones in club formation in the infected and hypertrophied roots during the invasion of P. brassicae was of an area of interest of the many clubroot researchers over the past several decades [14,[16][17][18]. Both auxins and cytokinins have key regulatory role at the beginning of hypertrophy in the infected root tissues at the early secondary phase of infection [16,17,[19][20][21].
Hormonal homeostasis of auxin and cytokinin is pivotal in the formation of clubroot [21][22][23][24][25]. Stimulation of the cell division which is associated with club formation in the infected roots requires the activity of cytokinin and auxin. In Arabidopsis thaliana and Brassica spp. an increased level of cytokinin and faster cell division were observed [16,17,26]. Further, cytokinin increases availability of nutrients including carbohydrates, amino acids, and lipids which are essential for the multiplication of pathogen during gall formation [22,26]. At the later stage of club development, a decreasing level of cytokinin primarily reflected a repression of cytokinin biosynthesis genes although few response regulator and receptor genes were induced [16,[26][27][28].  [13] after Dekhuijzen [14], Müller and Hilgenberg [15]). dpi, days post inoculation.
It is generally believed that an alteration in the usual hormonal balance has a vital role in the clubroot formation [13,16,17]. The contribution of hormones in club formation in the infected and hypertrophied roots during the invasion of P. brassicae was of an area of interest of the many clubroot researchers over the past several decades [14,[16][17][18]. Both auxins and cytokinins have key regulatory role at the beginning of hypertrophy in the infected root tissues at the early secondary phase of infection [16,17,[19][20][21].
Hormonal homeostasis of auxin and cytokinin is pivotal in the formation of clubroot [21][22][23][24][25]. Stimulation of the cell division which is associated with club formation in the infected roots requires the activity of cytokinin and auxin. In Arabidopsis thaliana and Brassica spp. an increased level of cytokinin and faster cell division were observed [16,17,26]. Further, cytokinin increases availability of nutrients including carbohydrates, amino acids, and lipids which are essential for the multiplication of pathogen during gall formation [22,26]. At the later stage of club development, a decreasing level of cytokinin primarily reflected a repression of cytokinin biosynthesis genes although few response regulator and receptor genes were induced [16,[26][27][28].
Despite the well-recognized importance of auxins and cytokinins in gall formation, the fluctuation in plant hormone contents does not correspond to the severity of infection in B. rapa roots [23]. Transcriptome analysis revealed a wide range of variation in the expression of cytokinin biosynthesis-related genes between infected and non-infected tissues. Laser micro-dissection and pressure catapulting (LMPC) detected up-regulated genes associated with cytokinin metabolism in A. thaliana during gall formation [29]. Microarray analysis during early phase of root infection at 4, 7, and 10 days revealed significant alteration in expression of a relatively lower number of genes in A. thaliana [30]. A few of those genes were found to change the level of expression during infection of the cortex from day 4 to day 10 [30]. By contrast, transcriptomes at two time-points, 10 and 23 days after inoculation, in the root cortex revealed that more than 1000 genes were differentially expressed between infected and control roots [28]. RNAseq data reported 2089 differentially expressed genes in resistant Chinese cabbage line at 30 days post inoculation compared to susceptible line where 188 and 138 genes were associated with plant-pathogen interaction and hormone signal transduction, respectively [24]. In another study, isobaric tags for relative and absolute quantitation (iTRAQ) analysis detected a total of 5003 differentially expressed proteins in resistant vs. susceptible line in secondary phase of disease infection [31].
Malinowski et al. [16] reported that the genes associated with cytokinin biosynthesis, signaling, and degradation show a remarkable repression in host during gall initiation at the secondary phase of infection in A. thaliana at 16 and 26 days post inoculation. However, to date, little is known about the behavior of cytokinin regulating, synthesizing, and degrading genes in plants during both primary and secondary phase of clubroot infection and gall expansion in Chinese cabbage. In the current study, we have characterized cytokinin regulating, synthesizing, and degrading genes from seven gene families in silico. We, then, analyzed the expression of 17 cytokinin-related genes, selected based on Schuller et al. [29] in Arabidopsis thaliana, in a Chinese cabbage cultivar inoculated with a highly virulent Korean P. brassicae isolate during both primary and secondary phase of clubroot infection. However, the specific role of BrCRR1, involved in protein phosphorylation, and BrPYL1, a receptor of abscisic acid (ABA), in cytokinin accumulation is not obvious. The pattern of expression of cytokinin-related genes was discussed to explain their role during primary infection, gall initiation, and gall expansion. We have also explored an association between increased expression of cytokinin biosynthesis genes and increase/decrease of cytokinin contents (in the published data) at the adjacent time-points in the clubroot infected root tissues to discern the role of cytokinin-related genes. The results of this study shed light on the changes in transcript levels for cytokinin-related genes, both in primary and secondary phases of clubroot infection, involved in metabolism, transport, and signaling in the host, during infection by this important pathogen in Chinese cabbage.

Disease Severity Index of Four Korean Field Isolates
A small but visible gall appeared at 21 DAI onwards. The isolate Seosan produced the largest galls at both 28 and 35 DAI, followed by Daejon, Gangneung1, and Yeoncheon ( Figure 2). Disease severity index (DSI), significantly varied among the four isolates ( Figure 2).

Properties of Cytokinin-Related Proteins
The 17 selected cytokinin-related genes were from seven different families: cytokinin oxidase, response regulator, histidine kinase, protein kinase, purine permease, transcriptional adaptor, isopentenyl transferase, and cytokinin-specific binding protein. The isoelectric point of purine permease, transcriptional adaptor, and isopentenyl transferase proteins are >7.0, whereas those of the other protein families are generally <7.0 ( Table 1). The molecular weights of these proteins range from 18.37 to 115 kD. Histidine kinase has the highest molecular weight (115 kD). All proteins except isopentenyl transferase, kinase, and cytokinin oxidase were predicted to localize to the nucleus.

Properties of Cytokinin-Related Proteins
The 17 selected cytokinin-related genes were from seven different families: cytokinin oxidase, response regulator, histidine kinase, protein kinase, purine permease, transcriptional adaptor, isopentenyl transferase, and cytokinin-specific binding protein. The isoelectric point of purine permease, transcriptional adaptor, and isopentenyl transferase proteins are >7.0, whereas those of the other protein families are generally <7.0 ( Table 1). The molecular weights of these proteins range from 18.37 to 115 kD. Histidine kinase has the highest molecular weight (115 kD). All proteins except isopentenyl transferase, kinase, and cytokinin oxidase were predicted to localize to the nucleus.

Expression Level Difference of Cytokinin-Related Genes in Leaf vs. Root
Nine cytokinin regulating, synthesizing, and degrading genes exhibited significant variation in their relative expression levels between leaf and root samples as the average of both inoculated and mock-treated samples at five different time-points ( Figure 3). Another eight genes showed non-significant variation between leaves and roots (Table S1). Two cytokinin degrading genes, BrCKX4 and BrCKX6 showed 13.3-and 6.4-folds higher expression, respectively, in root tissues compared to leaf tissues ( Figure 3). Both of the two response regulating genes, BrRR1 and BrRR6 accounted for 1.9-fold higher expression in root tissues compared to leaf tissues ( Figure 3). By contrast, both BrRR3 and BrRR5 genes showed 3.6-fold lower expression in root tissues compared to leaf tissues ( Figure 3). Similarly, BrPUP1 and BrIPT1 exhibited 5.5-and 2.7-folds lower expression, respectively, in root tissues compared to leaf tissues ( Figure 3). But the BrHK1 gene showed 6.5-fold higher expression in root tissues compared to leaf tissues ( Figure 3).

Expression Profiles of Response Regulator Genes
Six response regulator genes, BrRR1-6, showed significant variations in time-points, treatment within time-points and sample types within treatment and time-points (Table 2). These response regulators, except BrRR3 and BrRR4, also exhibited significant treatment difference (Table 2). Among the six response regulator genes, BrRR1 gene exhibited a 2.9-and 2.4-fold increase in expression on Day 1 and Day 28, respectively, in roots compared to mock-treated root samples at the same timepoint (Figures 4 and 5; Table 2). BrRR2 was upregulated 4.7-fold in root tissues on Day 28 in Seosaninoculated plants compared to mock-treated roots ( Figure 4). BrRR3 gene exhibited a 2.9-fold increase in expression in the leaves of Seosan-inoculated plants compared to mock-treated leaves at the same

Expression Profiles of Response Regulator Genes
Six response regulator genes, BrRR1-6, showed significant variations in time-points, treatment within time-points and sample types within treatment and time-points (Table 2). These response regulators, except BrRR3 and BrRR4, also exhibited significant treatment difference (Table 2). Among the six response regulator genes, BrRR1 gene exhibited a 2.9-and 2.4-fold increase in expression on Day 1 and Day 28, respectively, in roots compared to mock-treated root samples at the same time-point (Figures 4 and 5; Table 2). BrRR2 was upregulated 4.7-fold in root tissues on Day 28 in Seosan-inoculated plants compared to mock-treated roots (Figure 4). BrRR3 gene exhibited a 2.9-fold increase in expression in the leaves of Seosan-inoculated plants compared to mock-treated leaves at the same time-point (Figures 4 and 5). In Seosan-inoculated plants,

Expression Profiles of Other Cytokinin-Related Genes
In addition to investigating the expression patterns of response regulator and cytokinin oxidase genes, we analyzed the expression of five other genes, including histidine kinase, protein kinase, purine permease, transcriptional adaptor, and isopentenyl-transferase genes. At most time-points, the histidine kinase gene BrHK1 was more highly expressed in root vs. leaf tissue ( Figure 6). BrHK1 gene was expressed at 3.1-fold higher levels in Seosan-inoculated roots on Day 28 compared to mocktreated roots at the same time-point ( Figure 6). The protein kinase gene BrCRR1 was upregulated 3.5fold in the leaves of Seosan-inoculated plants on Day 3 compared to mock-treated leaf samples ( Figure 6). The

Expression Profiles of Other Cytokinin-Related Genes
In addition to investigating the expression patterns of response regulator and cytokinin oxidase genes, we analyzed the expression of five other genes, including histidine kinase, protein kinase, purine permease, transcriptional adaptor, and isopentenyl-transferase genes. At most time-points, the histidine kinase gene BrHK1 was more highly expressed in root vs. leaf tissue ( Figure 6).

Expression Profiles of Cytokinin Oxidase Genes
Cytokinin oxidase genes, BrCKX2-6, exhibited significant variations for time-points, treatment, time-points x treatment and sample types within treatment and time-points (Table 2). BrCKX2 was highly expressed on Day 1, with 3.4-fold higher expression in leaves of Seosan-inoculated plants compared to mock-treated leaves at the same time-point (Figures 5 and 7; Table 2). This gene also showed 225-, 2.5-, and 8.4-fold higher expression in roots on Day 3, Day 14, and Day 27, respectively,

Expression Profiles of Cytokinin Oxidase Genes
Cytokinin oxidase genes, BrCKX2-6, exhibited significant variations for time-points, treatment, time-points x treatment and sample types within treatment and time-points (Table 2). BrCKX2 was highly expressed on Day 1, with 3.4-fold higher expression in leaves of Seosan-inoculated plants compared to mock-treated leaves at the same time-point (Figures 5 and 7; Table 2). This gene also showed 225-, 2.5-, and 8.4-fold higher expression in roots on Day 3, Day 14, and Day 27, respectively, compared to mock-treated roots at the same time-point (Figures 5 and 7). BrCKX3 gene was expressed at 3.3-and 15.4-fold higher levels in Seosan-inoculated roots on Day 14 and Day 28, respectively, compared to mock-treated roots at the same time-point ( Figure 5). In general, BrCKX4 was expressed at higher levels in roots vs. leaves in control plants and in both mock-and Seosan-inoculated plants up to Day 14 compared to leaves (Figure 7). The expression of BrCKX4 increased 1.7-fold in Seosan-inoculated roots on Day 14 compared to mock-treated roots at the same time-point (Figures 5 and 7). BrCKX5 was upregulated 5.3-, 1.8-, and 1.3-fold in leaves of Seosan-inoculated plants on Day 1, Day 14, and Day 28 compared to mock-treated leaves at the same time-point (Figures 5 and 7). This gene also showed 2.1-fold increase in expression in Seosan-inoculated roots on Day 14 compared to mock-treated roots at the same time-point ( Figure 5). This gene was expressed at significantly (1.7-fold) higher levels in Seosan-inoculated roots on Day 14 compared to mock-treated roots at the same time-point (Figure 7). compared to mock-treated roots at the same time-point (Figures 5 and 7). BrCKX3 gene was expressed at 3.3-and 15.4-fold higher levels in Seosan-inoculated roots on Day 14 and Day 28, respectively, compared to mock-treated roots at the same time-point ( Figure 5). In general, BrCKX4 was expressed at higher levels in roots vs. leaves in control plants and in both mock-and Seosan-inoculated plants up to Day 14 compared to leaves (Figure 7). The expression of BrCKX4 increased 1.7-fold in Seosaninoculated roots on Day 14 compared to mock-treated roots at the same time-point (Figures 5 and 7). BrCKX5 was upregulated 5.3-, 1.8-, and 1.3-fold in leaves of Seosan-inoculated plants on Day 1, Day 14, and Day 28 compared to mock-treated leaves at the same time-point (Figures 5 and 7). This gene also showed 2.1-fold increase in expression in Seosan-inoculated roots on Day 14 compared to mocktreated roots at the same time-point ( Figure 5). This gene was expressed at significantly (1.7-fold) higher levels in Seosan-inoculated roots on Day 14 compared to mock-treated roots at the same timepoint (Figure 7).

Discussion
The main objective of this study was to assess the transcript levels of cytokinin-related genes in Chinese cabbage inoculated with P. brassicae isolate "Seosan" at five different time-points. Cytokinins are one of the major hormone group present in plants those regulated numerous physiological and biochemical processes in plants including growth, development, and metabolism [32][33][34]. We selected "Seosan" isolate because of its high virulence compared to the three other Korean isolates examined.

Expression Level Difference in Leaf vs. Root Tissue
Cytokinins play a crucial role in maintaining root-shoot homeostasis in plants through transduction of nutritional signals [35,36]. In this study, relative expression of several genes including BrHK1 and BrIPT1 dramatically changed in root and leaf tissues of both mock and Seosan-inoculated plants at different time-points indicating that hormonal homeostasis within plant is dynamic and that may be changed over time between organs (Figures 3 and 6).

Expression Level Difference of BrIPT1 Gene
Isopentenyl-transferase (IPT) is a critically important enzyme for the biosynthesis of cytokinin [38][39][40][41]. This enzyme is vital for cytokinin homeostasis under the infection of P. brassicae [42]. In B. rapa, Ando et al. [42] found that several BrIPT genes, including BrIPT1, were induced in the inoculated plants with P. brassicae at the first appearance of gall after 20 dpi. In this study a high level of expression of BrIPT1 gene by 11.5-and 8.3-fold in leaf and root tissues, respectively, on Day 1 after inoculation and a marginal upregulation in inoculated root tissues by 3.3-fold on Day 3 ( Figure 5) indicated that primary infection of P. brassicae in the root tissues transiently stimulated the expression of this gene. Expression of this gene was repressed on Day 3 and Day 14, during gall initiation ( Figure 6). Malinowski et al. [16] reported a comparatively lower expression of BrIPT1 gene on Day 16 compared to the expression level of BrIPT3 and BrIPT5. A quadruple mutant of IPT-ipt1 ipt3 ipt5 ipt7, measured a very low content of cytokinin in a previous study in A. thaliana indicating the importance of IPT1 gene in cytokinin biosynthesis [43]. This gene again showed 5.7-and 2.7-fold higher expression in leaves and roots of the treated plants, respectively, on Day 28 after inoculation, during gall enlargement, in infected plants indicating its role in gall expansion through induction of cell division ( Figure 6). Devos et al. [27] in Arabidopsis and Dekhuijzen [44] in B. rapa measured significantly higher contents of cytokinin at 4 dpi and 35 dpi, respectively (Figure 8). Increased expression levels of BrIPT1 gene, in this study, during primary infection of root hairs (on Day 1 and 3) and during gall formation (on Day 28) could be associated with higher biosynthesis of zeatin, isopentenyl adenine, or zeatin riboside in the infected root tissues compared to non-inoculated plants at the adjacent dpi (Figure 8).

Expression Level Difference of BrRR Genes
The BrRR proteins with two receptor (signal transduction) components have a specific histidine kinase to sense environmental changes to mediate response of a cell to environmental homeostasis [45]. The BrRR1-BrRR6 response regulator genes are the Type-A members bearing a short C-terminal extension in addition to the conserved receiver domain [45][46][47]. Exogenous application of trans-zeatin (tZ) transcriptionally upregulated the Type-A gene members [45]. A higher expression level of BrRR genes in Chinese cabbage might negatively regulate the expression of BrHK and BrHP genes instantly [45,48]. Induced expression of BrRR1 and BrRR6 during primary infection in roots, coincided with the expression of BrIPT1 gene. A 20.4-fold higher expression of BrRR4 gene on Day 35 indicated its vital role during gall enlargement in the infected root tissues as Type-A RR genes switch of cytokinin signal (Figures 5 and 8) [16,17,21,26]. In this study only BrRR4 gene was induced in leaf samples on Day 1, 3, and 14 ( Figure 6). The unregulated BrRR4 gene in leaves may enhance photosynthesis by interacting with phytochrome B after the infection of P. brassicae [49].

Expression Level of CKX Genes
The cytokinin oxidase (CKX) enzymes have proven role in irreversible degradation of cytokinins cleaving side-chain in roots and leaves of higher plants [50]. The CKX genes in both A. thaliana and B. rapa differ in subcellular localizations, biochemical properties, and level of expression from spatial and temporal perspectives (Table 1) [51,52]. In transgenic A. thaliana, overexpressed plants with six different CKX genes showed repressed accumulation of cytokinin that affected root and shoot growth [35]. Exogenous treatment of cytokinin was found to alter the responses of CKX genes in B. napus [53].
In this study, we analyzed the expression patterns of cytokinin-related genes in leaf and root tissues at different time-points in both non-inoculated and Seosan-inoculated plants. In a previous study, in transgenic Arabidopsis plants, the CKX1and CKX3-overexpressing plants exhibited increased resistance to P. brassicae [26]. We detected notably higher expression of BrCKX2, BrCKX3, BrCKX4, BrCKX5, and BrCKX6 on Day 14 (before gall formation) in Seosan-treated roots compared to mock-treated root indicating that cytokinin biosynthesis and signaling was repressed on Day 14 ( Figure 8) [16,27,28]. Our prediction is supported by the significantly lower accumulation of zeatin and isopentenyl adenine in the clubroot infected root tissues of Arabidopsis at 10 dpi in Malinowski et al. [16]. BrCKX2 and BrCKX3 were highly induced by 8.4-and 15.4-folds, respectively, on Day 28 when gall appeared but both of these two genes were repressed on Day 35 indicating that cytokinin signaling was enhanced during rapid gall enlargement phase ( Figure 5). Previous studies reported that down-regulation of CKX genes in A. thaliana caused biosynthesis of higher levels of cytokinin upon the infection of P. brassicae [22,28].

Gall Sample Collection
Four isolates from four different pathotype groups obtained from infected seedlings were previously collected from four different regions of South Korea by Jo et al. [54]. These isolates included Gangneung 1, Daejon, Yeoncheon, and Seosan, belonging to pathotype group 1, 2, 3, and 4, respectively [2].

Spore Preparation from Galls and Infection of Plant Materials
P. brassicae spores were extracted from each isolate collected from four different regions of South Korea according to Feng et al. [9] and Laila et al. [4]. The concentration of spores was adjusted to 1 × 10 7 spores per mL after counting the number of spores in a hemocytometer. Two-week-old plants were inoculated with the four selected isolates via the root cutting method, followed by culture in a controlled environment. Specifically, the plants were inoculated with 6 mL spore suspension (10 7 spores mL −1 ) for 10-15 min after cutting the root branches with scissors. The plants were transferred to a 200 cm 3 pot containing artificial soil in a controlled growth chamber. Mock treatment was performed using 6 mL water.

Disease Scoring Criteria
Inoculated plants were inspected weekly to observe the severity of infection (Figure 9). Each sample was washed with running tap water. The disease severity of clubroot infection was rated on a scale of 0-4 as described by Laila et al. [55] (Figure 9). Disease severity index (DSI) was calculated as the average disease scores of 10 plants (replicates) under each treatment.

Gall Sample Collection
Four isolates from four different pathotype groups obtained from infected seedlings were previously collected from four different regions of South Korea by Jo et al. [54]. These isolates included Gangneung 1, Daejon, Yeoncheon, and Seosan, belonging to pathotype group 1, 2, 3, and 4, respectively [2].

Spore Preparation from Galls and Infection of Plant Materials
P. brassicae spores were extracted from each isolate collected from four different regions of South Korea according to Feng et al. [9] and Laila et al. [4]. The concentration of spores was adjusted to 1 × 10 7 spores per mL after counting the number of spores in a hemocytometer. Two-week-old plants were inoculated with the four selected isolates via the root cutting method, followed by culture in a controlled environment. Specifically, the plants were inoculated with 6 mL spore suspension (10 7 spores mL −1 ) for 10-15 min after cutting the root branches with scissors. The plants were transferred to a 200 cm 3 pot containing artificial soil in a controlled growth chamber. Mock treatment was performed using 6 mL water.

Disease Scoring Criteria
Inoculated plants were inspected weekly to observe the severity of infection (Figure 9). Each sample was washed with running tap water. The disease severity of clubroot infection was rated on a scale of 0-4 as described by Laila et al. [55] (Figure 9). Disease severity index (DSI) was calculated as the average disease scores of 10 plants (replicates) under each treatment. Figure 9. Scoring criteria for clubroot development in Chinese cabbage cultivar "Bullam-3-ho." Scale of 0-4: 0 = root system without clubs, 1 = one or a few small clubs <5 mm in diameter on lateral roots, 2 = a few medium-sized, separate globular clubs >5mm in diameter on lateral roots, 3 = medium-sized clubs 5-10 mm in diameter on the main roots, 4 = severe clubbing, with clubs >10 mm in diameter on the lateral and main roots [55].

Collection of Leaf and Root Samples from Plants Inoculated with Seosan Isolate
For qRT-PCR expression analysis of cytokinin-related genes, 14-day-old seedlings of Chinese cabbage cultivar "Bullam-3-ho" were inoculated with spores of the Seosan isolate (10 7 spores mL −1 ). Root and leaf samples were harvested from mock and inoculated plants for qRT-PCR analysis at 1, 3, 14, 28 and 35 days after inoculation (DAI). A list of sampling combinations is given in Table 3. Three biological replicates were collected for each sample. During sample collection, any adhering soil . Scoring criteria for clubroot development in Chinese cabbage cultivar "Bullam-3-ho." Scale of 0-4: 0 = root system without clubs, 1 = one or a few small clubs <5 mm in diameter on lateral roots, 2 = a few medium-sized, separate globular clubs >5 mm in diameter on lateral roots, 3 = medium-sized clubs 5-10 mm in diameter on the main roots, 4 = severe clubbing, with clubs >10 mm in diameter on the lateral and main roots [55].

Collection of Leaf and Root Samples from Plants Inoculated with Seosan Isolate
For qRT-PCR expression analysis of cytokinin-related genes, 14-day-old seedlings of Chinese cabbage cultivar "Bullam-3-ho" were inoculated with spores of the Seosan isolate (10 7 spores mL −1 ). Root and leaf samples were harvested from mock and inoculated plants for qRT-PCR analysis at 1, 3, 14, 28 and 35 days after inoculation (DAI). A list of sampling combinations is given in Table 3. Three biological replicates were collected for each sample. During sample collection, any adhering soil residue was carefully removed from the roots using running tap water. The excess water from the roots were then dried between two pieces of filter paper and immediately frozen in liquid nitrogen. The leaf and root samples were stored at −80 • C.

RNA Extraction and cDNA Synthesis
Approximately 100 mg of leaf or root tissue was ground to a powder in liquid nitrogen. Total RNA was extracted from all 22 types of samples using an RNeasy Plant Mini Kit (Qiagen, Hilden, Germany). RNase-free DNase (Qiagen, Hilden, Germany) treatment was performed to remove DNA contamination. The purity of the extracted RNA was determined at a 260/280 nm ratio using a NanoDrop ® ND-1000 (Thermo Scientific, Hudson, NH, USA). The cDNA was synthesized from RNA following the manufacturer's instructions using a Superscript ® III First-Strand Synthesis Kit (Invitrogen, CA, USA).

Identification of Cytokinin-Related Genes
Schuller et al. [29] published a list of genes involved in cytokinin metabolism, signaling, and transport in Arabidopsis thaliana. In this study, genes were selected based on their high expression level in Arabidopsis thaliana. The corresponding Brassica rapa orthologs were retrieved from the BRAD database (http://brassicadb.org/brad/; Cheng et al. [56] against each cytokinin-related gene (Table 1, Figure S1). The subcellular localization of each corresponding cytokinin biosynthetic protein was predicted using both UniProt (https://www.uniprot.org/uniprot/) and ProtComp 9.0 from Softberry (http://linux1.softberry.com/berry.phtml). A phylogenetic tree was constructed using Neighbor-joining method in MEGA7.0 software [57]. ProtParam tool [58] was used to determine the properties of the proteins. However, a database research in The Arabidopsis Information Resource (TAIR) (https://www.arabidopsis.org/) identified that At3g55950 (BrCRR1, Bra007202) is a protein kinase that is involved in protein phosphorylation. In addition, At5g05440 gene (BrPYL1, Bra009113) is an abscisic acid (ABA) binding protein that acts as a receptor of ABA.

qRT-PCR Expression Analysis
We performed qRT-PCR to investigate the relative expression patterns of cytokinin-related genes in 20 samples representing five different time-points, two treatments, and two sample types: leaves and roots. Gene expression levels in leaves from non-treated plants on Day 0 were set to 1. There were two technical replicates for each sample. Cytokinin-related genes-specific primers were used to analyze the gene expression in Seosan-inoculated leaf and root samples via quantitative reverse-transcription PCR (qRT-PCR) ( Table 4). The primers were designed based on the coding sequence of each gene using Primer3 software (http://frodo.wi.mit.edu/primer3). The specificity of the primers was checked following Robin et al. [59]. Two different ACTIN genes (GenBank Accession Nos. XM_009147610.2 and FJ969844.1, housekeeping genes) from B. rapa were used to normalize the expression levels of cytokinin-related genes. Each 20 µL PCR reaction mixture contained 1 µL template cDNA at 50 ng µL −1 concentration, 1 µL each of forward and reverse primers at 10 pmol concentration, 10 µL qPCR BIOSyGreen Mix Lo-ROX (PCR Biosystems, London, UK), and 7 µL ultra-pure double distilled water. The qRT-PCR conditions were as follows: denaturation for 10 min at 95 • C, followed by 40 cycles of amplification at 95 • C for 20 s, 58 • C for 20 s, and 72 • C for 25 s. The quantification cycle value (a measure of fluorescence) was recorded at the end of each cycle. The amplified products were detected and the data were analyzed using a LightCycler96 system (Roche, Mannheim, Germany). The 2 −∆∆CT method was used to calculate the relative expression levels of cytokinin-related genes because equal primer efficiency of the tested genes was used, two stable housekeeping genes were used for standardization, and the selected genes were tested at both control and experimental conditions [60].

Statistical Analysis
Analysis of variance (ANOVA) of relative expression levels of five time-points (dpi), two treatments (mock and Seosan-inoculated), and two sample types (leaf vs. root) was conducted following a nested-ANOVA under general linear model using MINITAB 17 Statistical Software (Minitab Inc., State College, PA, USA) to assess the variation among time-points, treatments, time-point x treatment, and sample type within time-point and treatment. Separate statistical analyses were conducted for disease severity scores and relative expression levels of cytokinin-related genes at two different sample types (leaf vs. root) via one-way ANOVA using MINITAB 17 Statistical Software (Minitab Inc., State College, PA, USA). For pairwise comparisons of means of disease severity scores or relative expression values, Tukey's pairwise comparisons were performed.

Conclusions
In this study, the Korean pathotype group 4 isolate Seosan was found to be more virulent than three other isolates, Gangneung1, Daejon, and Yeoncheon, with respect to DSI. The expression patterns of cytokinin-related genes involved in response regulation, biosynthesis, and degradation differed greatly between leaf and root tissues, pointing to their organ-specific expression. Cytokinin biosynthetic gene BrIPT1 and response regulator genes BrRR1, BrRR4, and BrRR6 showed a pattern of upregulation at the beginning of primary infection and during gall expansion. The induced expression of BrCKX genes before gall formation indicated their possible role in repression of gall formation. Altered cytokinin contents in root tissues, in the published data, upon the infection of P. brassicae temporally coincided with altered expression levels of biosynthetic and catalytic genes of this study. We conclude that the genes BrIPT1, BrRR1, BrRR6, and BrCKXs have vital role in clubroot disease development. Multi-genic manipulation of susceptible Chinese cabbage plants with BrCKX2-6 genes could be screened for clubroot disease resistance in future. The results of this study improve our current understanding of the possible role of relevant genes in cytokinin signaling during clubroot formation.
Supplementary Materials: The following are available online at http://www.mdpi.com/1422-0067/21/11/3896/s1, Figure S1. Phylogenetic tree showing association between cytokinin biosynthesis genes (black) in Arabidopsis thaliana and orthologues in Brassica rapa. BrCRR1 is involved in protein phosphorylation and BrPYL1 is a receptor of ABA. Table S1. Variation in relative expression of cytokinin regulating, synthesizing and degrading genes in root and leaf samples showing non-significant variation between leaves and roots. Each data is the average (±sd) of both inoculated and non-inoculated samples at five different time points. BrCRR1 is involved in protein phosphorylation and BrPYL1 is a receptor of ABA.