The early stages of infection of canola roots by the clubroot pathogen Plasmodiophora brassicae were investigated. Inoculation with 1 × 105 resting spores mL−1 resulted in primary (root hair) infection at 12 h after inoculation (hai). Secondary (cortical) infection began to be observed at 72 hai. When inoculated onto plants at a concentration of 1 × 104 mL−1, secondary zoospores produced primary infections similar to those obtained with resting spores at a concentration of 1 × 105 mL−1. Secondary zoospores caused secondary infections earlier than resting spores. When the plants were inoculated with 1 × 107 resting spores mL−1, 2 days after being challenged with 1 × 104 or 1 × 105 resting spores mL−1, secondary infections were observed on the very next day, which was earlier than the secondary infections resulting from inoculation with 1 × 107 resting spores mL−1 alone and more severe than those produced by inoculation with 1 × 104 or 1 × 105 resting spores mL−1 alone. Compared with the single inoculations, secondary infections on plants that had received both inoculations remained at higher levels throughout a 7-day time course. These data indicate that primary zoospores can directly cause secondary infection when the host is under primary infection, helping to understand the relationship and relative importance of the two infection stages of P. brassicae.
Clubroot, caused by Plasmodiophora brassicae, is one of the most destructive soilborne diseases of cruciferous crops worldwide and is an emerging threat to canola (Brassica napus) production in Canada (Hwang et al., 2012). The pathogen is an obligate parasite with a life cycle partitioned into three stages: survival in the soil as resting spores, primary infection of root hairs, and secondary infection of and development within the root cortex (Ingram & Tommerup, 1972; Naiki & Dixon, 1987). Each resting spore germinates to release one primary zoospore. These zoospores swim to and infect root hairs by penetrating the cell wall. Within the root hairs, primary plasmodia develop and cleave into zoosporangia, each containing 4–16 secondary zoospores. Primary infections do not produce macroscopic symptoms (Howard et al., 2010). The secondary zoospores are either released into the rhizosphere and then infect the root cortex from outside, or they infect neighbouring cortical cells directly from inside the root hair (Kageyama & Asano, 2009). Secondary infection is followed by the development of secondary plasmodia within the root cortex, which results in the production of clubroot symptoms, i.e. club-shaped malformations of the roots. Each secondary plasmodium will eventually be cleaved into large numbers of resting spores within the clubbed root. As the root tissues disintegrate, the resting spores are released into the soil to complete the disease cycle.
Many strategies have been proposed for clubroot control (Donald & Porter, 2009) and among others, the use of resistant cultivars is believed to be most effective (Diederichsen et al., 2009). Breeding for resistance requires a better understanding of the mechanisms of pathogenesis. However, many aspects of P. brassicae pathogenesis are not fully understood (Hwang et al., 2012), one of which is the relative importance of primary and secondary infections in clubroot establishment, or, in other words, of primary and secondary zoospores in the pathogenicity of P. brassicae.
Primary and secondary zoospores cannot be differentiated based on morphology (Kageyama & Asano, 2009). The abilities to cause different infections suggest that these two kinds of zoospores differ at a molecular level, most likely in gene expression. Data on this aspect is lacking, although this may be essential for resistance selection and development of resistant cultivars. It has been recognized that secondary zoospores can cause both primary and secondary infections in both hosts and non-hosts (Naiki et al., 1984; Ludwig-Müller et al., 1999; Feng et al., 2012). It seems as though secondary zoospores can supersede primary zoospores and play a monodrama during the infection stage of the pathogen life cycle. A known contribution of primary zoospores to secondary infection is the generation of multiple secondary zoospores through primary infection (Kageyama & Asano, 2009). However, this function could be compromised by the huge number of resting spores produced by secondary infection. Thus, it is reasonable to ask a question: why does P. brassicae produce primary zoospores?
One hypothesis is that P. brassicae uses primary infection to break down the basal resistance of the plant, in order to enable secondary infection; in this case, primary infection (rather than secondary zoospores) would be the real prerequisite for secondary infection. To provide information supporting this hypothesis, experiments were conducted in the present study to determine: (i) whether secondary zoospores directly cause secondary infections in healthy plants, and (ii) whether primary zoospores directly cause secondary infections in plants that have already developed primary infections.
Materials and methods
Plant material and P. brassicae inoculum
The canola cultivar Westar was used as host. Westar roots with galls developing from natural infections were collected from an experimental field plot near Leduc, AB, Canada. The galls were kept at −20°C and resting spores from these galls were used as the initial inoculum.
Preparation of seedlings
Canola seeds were surface-sterilized in 1% sodium hypochlorite for 5 min, washed with distilled water, and germinated on moistened filter paper for 7 days. The resulting seedlings were used in all inoculation experiments.
Preparation of resting spores
Galls were homogenized in 10% (w/v) sucrose in a blender. The slurry was passed through eight layers of cheesecloth and the suspension was centrifuged at 50 g for 5 min. After transfer into a new tube, the supernatant was centrifuged at 2000 g for 5 min. The resulting pellet consisted of two distinct layers: the lower black layer consisted of soil particles, and the upper white to brownish layer of resting spores (Bryngelsson et al., 1988). The upper layer was suspended in 5 mL water by gentle pipetting and transferred into a new tube containing 40 mL water. After centrifugation at 2000 g for 5 min, the supernatant was removed. Spores in the pellet were adjusted to 1 × 107 spores mL−1 and surface-disinfected by using 1 μg mL−1 colistin sulphate (Sigma-Aldrich) and 1 μg mL−1 vancomycin hydrochloride (EMD Millipore) at 25°C in the dark for 24 h (Asano et al., 2000). The spores were used as inoculum after removal of the antibiotics by washing twice with 40 mL distilled water.
Preparation of secondary zoospores
Canola seedlings were transplanted in 40- × 30- × 15-cm plastic trays filled with sand that had been inoculated with 1 × 107 resting spores mL−1 (see below for sand inoculation).
After 7 days, about 400 plants were dug out and the roots were washed with tap water. The foliage was then cut off at the soil level, and the roots were rinsed three times by shaking at 150 rpm for 20 min in 500 mL distilled water in a 1000-mL flask. The roots were then shaken in 100 mL distilled water at 100 rpm for 20 h to stimulate release of secondary zoospores. After removing the roots, 10 mL of the zoospore suspension was concentrated by centrifugation at 5000 g for 5 min and adjusted to 2·5 × 106 spores mL−1. Ten samples of the concentrated suspension were examined microscopically to confirm the absence of resting spores. The original suspension was then adjusted to 1 × 105 spores mL−1 and used immediately for inoculation.
Inoculation of sand
Sea sand was washed with distilled water, autoclaved at 121°C for 60 min, and then dried at 80°C overnight. For each inoculation, 4·5 volumes of sand were mixed with 1 volume of spore suspension, which saturated the sand with water. Based on the post-inoculation volume, the concentrations of spore suspensions were adjusted to ensure that the planned concentrations were applied in the sand. Throughout this study, all plant inoculations were conducted by transplanting pre-germinated seedlings into the inoculated sand.
Ten canola seedlings were transferred into one well of a 24-well cell culture cluster (Corning Life Sciences). The roots and the lower stem were buried with 3 mL inoculated sand. Two experiments were conducted. In experiment A1, plants were challenged with 1 × 104 resting spores mL−1 (treatment 1) or 1 × 104 secondary zoospores mL−1 (treatment 2). In experiment A2, treatment 2 was the same as in A1, but for treatment 1 the resting spore concentration was 1 × 105 mL−1. Each of the two experiments was conducted as a randomized complete block (RCB) with 10 replicates. The cell culture clusters were sealed individually in transparent zipper bags and kept in a growth chamber maintained at 24/18°C (day/night), with a 16-h photoperiod and 80% RH. No watering was applied during the experiment. One plant sample was collected from each replicate (well) at 12, 24, 36, 48, 72, 96 and 120 h after inoculation (hai). The roots from the samples were either examined immediately (12–72 hai) or preserved in FAA fixative (100 mL contained 50 mL ethyl alcohol, 5 mL glacial acetic acid and 10 mL 37–40% formaldehyde) for later observation.
Experiment B: resting spores + resting spores
Petri dishes (100 × 25 mm) were filled with 100 mL sand. Twenty seedlings were transplanted into each dish. Two experiments were conducted. In B1, plants were challenged with resting spores either at 1 × 104 resting spores mL−1 at day 0 (treatment 1), 1 × 107 resting sporesmL−1 at day 2 (treatment 2), or 1 × 104 resting spores mL−1 at day 0 followed by 1 × 107 resting spores mL−1 at day 2 (treatment 3). In B2, treatment 1 and the first inoculation in treatment 3 were conducted with 1 × 105 resting spores mL−1; in treatment 2 and the second inoculation in treatment 3, the concentration of resting spores was the same as in B1. Inoculation in treatment 2 and the second inoculation in treatment 3 were conducted by pipetting 5 mL resting spore suspension into each of the Petri dishes. For the dishes in treatment 1, the same amount of water was added. Both experiments were conducted as RCBs with 10 replications. The plants were watered every second day with 7·5 mL distilled water and maintained in the same growth chamber as in experiment A. One plant was sampled from each dish every day from day 2 to day 7. The samples were preserved in the FAA fixative.
Primary and secondary infections were investigated on a 1-cm fragment of root from each plant using a Zeiss AXIO microscope (Carl Zeiss). On each sample, five fields of view using the ×20 objective lens were examined. In each field of view, two sets of data were collected: the percentage of root hairs with primary infection, and the total number of secondary plasmodia.
The percentage of root hairs with primary infections and the total number of secondary plasmodia were calculated for each root sample. Data from each time point were subjected to analysis of variance using the Microsoft Excel add-in dsaastat developed by Dr Andrea Onofri at the University of Perugia, Italy (http://www.unipg.it/~onofri/index.htm). Differences between/among treatments were assessed using Fisher’s LSD test (P ≤0·05).
Infections after inoculation with resting spores
After inoculation with 1 × 105 resting spores mL−1, the attachment of primary zoospores to root hairs was observed at 12 hai (Fig. 1a), although whether or not infection was successful could not be determined at this time. Nonetheless, attachment was considered as infection for the quantitative assessment. Young primary plasmodia developed in root hairs at 24 hai (Fig. 1b) and enlarged at 36 hai (Fig. 1c). The formation of zoosporangia started at 48 hai (Fig. 1d). Secondary infection, indicated by the presence of plasmodia in the cortical tissue, was evident at 72 hai (Fig. 1e). Also at 72 hai, some zoosporangia became partially evacuated (Fig. 1f), while some others contained well-developed secondary zoospores that appeared to be moving towards the cortex (Fig. 1g). These were continuously observed at 96 and 120 hai (data not shown). Secondary infections were commonly observed at 96 hai, with round or oval-shaped secondary plasmodia embedding in the cortical cells (Fig. 1h). Fusion of two plasmodia was observed at 96 hai (Fig. 1i). At 120 hai, mature secondary plasmodia commonly occurred in groups (Fig. 1j).
Infections after inoculation with secondary zoospores
Primary infections caused by secondary zoospores started at 12 hai (Fig. 2a,b). Compared with inoculation with 1 × 104 resting spores mL−1, inoculation with secondary zoospores produced similar primary infection rates until 48 hai, but at the later time points, secondary zoospores produced higher rates of primary infection (Fig. 2a). Throughout the time course, secondary zoospores and 1 × 105 resting spores mL−1 produced similar primary infection rates (Fig. 2b). The morphologies and the time of appearance of primary infections caused by secondary zoospores were similar to those caused by resting spores (Fig. 1a–d and f–g).
Inoculation with either 1 × 104 or 1 × 105 resting spores mL−1 did not produce secondary infections until 72 hai (Fig. 2c,d). In contrast, secondary infections were produced by inoculation with secondary zoospores at 12 hai (Fig. 2c,d), suggesting that the secondary zoospores caused secondary infections without going through the primary infection process. An accelerated increase in secondary infections caused by secondary zoospores was noticed after 72 hai (Fig. 2c,d), suggesting that the increase in secondary zoospore number was a result of primary infection. The morphologies of secondary infections caused by secondary zoospores or resting spores were similar. The fusion of secondary plasmodia (Fig. 1i) was observed at 36 hai, i.e. 36 h earlier than it was observed after inoculation with resting spores.
Secondary infections caused by primary zoospores
Compared to inoculations with 1 × 104 resting spores mL−1 (treatment 1), inoculations with 1 × 107 resting spores mL−1 (treatments 2 and 3) produced higher numbers of primary infections (Fig. 3a). The latter two treatments produced similar numbers of primary infections to those produced by inoculation with 1 × 105 resting spores mL−1, especially during the later stages of the time course (Fig. 3b), suggesting that amounts of primary infections produced by each of these three treatments may have represented the maximum levels of primary infection.
No secondary infections were observed on plants inoculated with either 1 × 104 or 1 × 105 resting spores mL−1 until day 3 (Fig. 3c,d). Inoculation with 1 × 107 resting spores mL−1 produced secondary infections at day 4 (2 days after inoculation; dai) in one experiment (Fig. 3c), but no secondary infections in the other (Fig. 3d). The inoculations with 1 × 104 resting spores mL−1 +1 × 107 resting spores mL−1, and with 1 × 105 resting spores mL−1 +1 × 107 resting spores mL−1 produced secondary infections at 3 dai, higher than single inoculations with 1 × 104 or 1 × 105 resting spores mL−1, respectively (Fig. 3c,d). Because in these treatments the inoculation with 1 × 107 resting spores mL−1 was performed on day 2, there was insufficient time for the production of secondary zoospores through primary infection. Thus, the extra secondary infections must have resulted from infection by primary zoospores generated from the inoculated resting spores.
In the later points of the time course, secondary infections produced in treatment 3 were higher than in the other two treatments, and this was observed in both experiments. The difference between treatments 2 and 3 also suggests a contribution by primary zoospores to the extensive secondary infections observed in treatment 3 (Fig. 4).
After inoculation with resting spores, primary infections began to be noticed as early as 12 hai. Previous studies reported longer periods prior to the identification of primary infections; for example, primary infections were not identified until 10 dai in Chinese cabbage (Tanaka et al., 2006) and cauliflower (Donald et al., 2008), or 6 dai in canola (Hwang et al., 2011). However, in these studies, assessments were either not conducted earlier, or seeds rather than seedlings were sown in the P. brassicae-infested growth medium. Using hairy root culture, Asano et al. (2000) observed that the primary infection of turnip root hairs began at 4 dai with resting spores, rapidly increasing up to 6 dai, and continuing to increase more slowly until 10–12 dai. In contrast, Deora et al. (2012) observed that nearly 50% of canola root hairs were infected at 4 days after inoculation with resting spores in sand culture. Although no observations were conducted prior to 4 days in that study, the infection may have started earlier. In another study, Dobson & Gabrielson (1983) observed primary infections on Chinese cabbage within 1 day after inoculation. The consistent observations in the four independent experiments conducted in the present study suggest that primary infection is initiated earlier than generally used to be thought.
Primary infections caused by secondary zoospores were observed in previous studies (Naiki et al., 1984; Feng et al., 2012). The present study indicates that primary and secondary zoospores can produce primary infections synchronously that are similar in morphology, supporting the hypothesis that primary and secondary zoospores may essentially be the same in their identities. On the other hand, in the early stages of infection when both spores are able to cause primary infection, recognition of each other as non-self would result in competition between the two spore types, which would force the pathogen to suffer an extra cost to balance the amounts of spores produced.
In the present study, secondary infections were observed at 72 h after inoculation with resting spores, when the release of secondary zoospores from sporangia was only beginning. Dobson & Gabrielson (1983) indicated that the time required by secondary zoospores for cortical infection was 3 h from their release. This may allow the simultaneous observation of secondary zoospore maturation and secondary infection. Another study conducted in Arabidopsis thaliana (Mithen & Magrath, 1992) indicated that primary zoospores can cause secondary infections without the generation of secondary zoospores. Based on these data, a motile myxamoeboid phase was suggested: pseudopodia-like structures (myxamoeboid) develop from primary zoospores during primary infection and then move into the cortical cells to cause secondary infection. Nevertheless, production of a putative myxamoeboid phase would still be dependent on primary infection. Data from the present study suggest that primary zoospores could cause secondary infections without first causing primary infections. Indeed, inoculation of high concentrations of resting spores on plants that were already under primary infection resulted in higher levels of secondary infection within 1 day. The increase in secondary infections could only be attributed to the second inoculation. Moreover, while the possibility that the additional secondary infections 1 day after the second inoculation were caused by secondary zoospores that were quickly released after primary infection (or derived from a myxamoeboid phase) cannot be excluded, the consistently higher levels of secondary infection later in the time course (Fig. 3c,d) could only be explained by the contribution of resting spores. This is because after 3 dai, treatment 3 exhibited similar numbers of primary infections, but higher numbers of secondary infections relative to treatments 1 and 2.
After inoculation of healthy plants with secondary zoospores, primary and secondary infections could be observed simultaneously (Fig. 2). This makes it difficult to verify the requirement of a primary infection phase for the production of secondary zoospores to cause secondary infections. If primary infection is a process that serves to increase the susceptibility of plants to secondary infection, then a small number of primary infections might be sufficient for this purpose. In the present study, the two-inoculation treatments in two independent experiments produced similar amounts of secondary infections (Trt 3 in Fig. 3c,d), albeit that the numbers of primary infections caused by the initial inoculations were different (Trt 1 in Fig. 3a,b). Thus, whether or not secondary infections caused by secondary zoospores require the presence of primary infections remains unclear.
From the available data, it is difficult to elucidate exactly what may have been happening during the co-evolution between P. brassicae and its hosts with respect to pathogenesis. Based on the present study, one hypothesis can be proposed: the pathogen uses primary infection to overcome the basal resistance of the plant to cortical infection. Infection of root hairs is rather unspecific and has been observed in resistant (Diederichsen et al., 2009) and non-host plants (Feng et al., 2012). These root hairs not only serve as a niche for the pathogen to survive, but also provide an opportunity for close interaction between host and pathogen, facilitating the evolution of mechanisms by which P. brassicae can break down the basal resistance to cortical infection. Under such a scenario, it can be postulated that the pathogen has succeeded in overcoming this resistance in hosts, but has failed in non-hosts. More support for this hypothesis comes from a transcriptome analysis of A. thaliana clubroots (Siemens et al., 2006), in which most known defence- or resistance-related genes were found to be either not differentially expressed or down-regulated during pathogenesis. Compared to 10 dai, more genes were down-regulated at 23 dai. The authors suggested that there was an initial equilibrium between pathogen attack and host plant defence, which later shifted in favour of the pathogen. Although the precise time point at which this shift occurred is unclear, it is consistent with the above hypothesis.
In summary, the present study demonstrates for the first time that primary zoospores can cause secondary infections and suggests that primary and secondary zoospores may be essentially the same with respect to their ability to cause infection. However, additional data, including studies at the molecular level, will be required to support this hypothesis.
The authors acknowledge the funding from the Canola Agronomic Research Program (Alberta Canola Producers Commission, Manitoba Canola Growers Association, SaskCanola, and the Canola Council of Canada), the Alberta Crop Industry Development Fund (ACIDF) and the Clubroot Risk Mitigation Initiative (AAFC/CCC).