Analysis of the Oxidative Burst and Its Relevant Signaling Pathways in Leptosphaeria maculans—Brassica napus Pathosystem

An oxidative burst is an early response of plants to various biotic/abiotic stresses. In plant-microbe interactions, the plant body can induce oxidative burst to activate various defense mechanisms to combat phytopathogens. A localized oxidative burst is also one of the typical behaviors during hypersensitive response (HR) caused by gene-for-gene interaction. In this study, the occurrence of oxidative burst and its signaling pathways was studied from different levels of disease severity (i.e., susceptible, intermediate, and resistant) in the B. napus–L. maculans pathosystem. Canola cotyledons with distinct levels of resistance exhibited differential regulation of the genes involved in reactive oxygen species (ROS) accumulation and responses. Histochemical assays were carried out to understand the patterns of H2O2 accumulation and cell death. Intermediate and resistant genotypes exhibited earlier accumulation of H2O2 and emergence of cell death around the inoculation origins. The observations also suggested that the cotyledons with stronger resistance were able to form a protective region of intensive oxidative bursts between the areas with and without hyphal intrusions to block further fungal advancement to the uninfected regions. The qPCR analysis suggested that different onset patterns of some marker genes in ROS accumulation/programmed cell death (PCD) such as RBOHD, MPK3 were associated with distinct levels of resistance from B. napus cultivars against L. maculans. The observations and datasets from this article indicated the distinct differences in ROS-related cellular behaviors and signaling between compatible and incompatible interactions.


Introduction
Canola, as a brand of cultivars from rapeseed and field mustard, has become a very important crop to Canada as well as to the world, since they have low amounts of erucic acid and glucosinolate, which have bad tastes and harmful effects on human and animal consumption [1,2]. Blackleg (caused by fungus Leptosphaeria maculans) has become such a great threat since 1975 in Canada [3] that it may cause a 50% reduction in the production of canola. The major symptoms caused by the pathogen include stem canker, root rot, leaf lesion (with pycnidia), and pycnidia and pseudothecia on stubble [4].
Traditionally, the blackleg disease is managed by crop rotation, fungicide, etc., however, the development of genetically resistant rapeseed (Brassica napus) is the most costeffective and environmentally friendly strategy for dealing with the disease [5]. L. maculans possesses various enzymes to reinforce its infection upon hosts. Three cell-wall-degrading enzymes (CWDEs): endopolygalacturonase (pg1) and two cellulases (cel1 and 2), and cel2 transcripts are found in the cotyledons and leaves from B. napus and B. juncea [6]. Moreover, L. mauclans also secretes phytotoxins such as sirodesmin PL, which causes leaf lesion [7,8].
Generally, there are two types of disease resistance in plants (triggered by phytopathogenic infection): qualitative and quantitative resistance. Qualitative resistance cultivars. The observations of cytological behaviors were also able to visualize the effects of ROS signaling on disease resistance.
To understand how oxidative burst works at a physiological level, the measurement of electrolyte leakage is a useful tool. As one of the earliest responses to various stresses, electrolyte leakage is found to trigger multiple defensive mechanisms in planta, which includes hormonal secretion, programmed cell death, oxidative burst, etc. [17,19,20]. For this study, the electrolyte leakage was measured from the excised cotyledons, the voltage caused by the leaked electrolytes from both mock and inoculated samples were measured with the VWR sympHony conductivity meter. The goal of the study is to describe how ROS production and signaling function in the Brassica napus-Leptosphaeria maculans interaction, we intended to elucidate the role of ROS signaling with different severities of B. napus defense. By analyzing and comparing the pivotal genes within ROS signaling pathways, the onset patterns and expression levels of those studied genes can explain the various disease severities among different canola cultivars. The observations of cytological behaviors were also able to visualize the effects of ROS signaling on disease resistance.
(a)  As shown in Figure 1, the two resistant genotypes Surpass400 and 01-23-2-1 exhibited a significantly higher level of electrolyte leakage (compared with mock-inoculated cotyledons) as early as 5 dpi, when the susceptible Westar cotyledons did not have the induction of significant electrolyte leakage. Westar started to induce higher leakage at 7 dpi, and at 11 dpi, the cotyledons were collapsed and severely damaged to perform a further measurement, therefore, there was no data about Westar leakage at 11 dpi. Surprisingly, Surpass400-H75 8-1 seemed to have retained the secretion of electrolyte at 11 dpi according to the conductivity measurement. The results suggested that resistant genotypes had earlier activation of electrolyte secretion (at 5 dpi) while the compatible interaction (i.e., Westar) had a later triggering process (at 7 dpi).

Distinct Detection of Hydrogen Peroxide in Susceptible, Intermediate and Resistant B. napus Plants
As a stable and reactive ROS molecule, H 2 O 2 plays multiple roles in plants during normal physiological functioning and stress resistance, its membrane-permeable property makes it a useful messenger in cell-cell communication, thus coordinates cellular signaling mechanisms which are time/space-dependent [12,14,34]. By 3,3 -Diaminobenzidine (DAB) staining, the diffusion of hydrogen peroxide (H 2 O 2 ) was visualized as brown-colored stains.
As shown in Figure Figure 2b, red arrows). Adequate H 2 O 2 accumulation induces considerable signaling, which triggers defense responses at the cellular level, such as MAPK cascades and Ca 2+ signaling [14,16,35]. Since the accumulation of H 2 O 2 plays central roles in the activation of plant defense signaling, the intense accumulation of H 2 O 2 on Surpass400 (7 dpi) and 01-23-2-1 (5 and 7 dpi) cotyledons indicated that the gene-for-gene interaction (for both intermediate and resistant cases) can induce early H 2 O 2 outburst to trigger anticipated and localized defense activities to inhibit fungal development.

The Impacts of ROS Upon Cell Death
Followed by H 2 O 2 , various physiological activities can be triggered to stop further pathogenic progression, those activities include callose deposition, and cell wall crosslinking [11,12,14,24,36]. Another biological process highly regulated by H 2 O 2 is the programmed cell death (PCD) [11,12,16] . The brownish stains (black arrows) suggested the diffusion regions of H2O2. Westar at 11 dpi also showed some pycnidia production (red arrows). (b). The representative microscopic images taken from the origins of inoculation (magnitude: 50×), the images showed the details about H2O2 accumulation when the fungus progressed from the puncture holes for inoculation. The brownish color shown from Surpas400 and 01-23-2-1 had more captured H2O2 around the origins of inoculation (red arrows) and some pycnidia from Westar and Surpass400 were also captured (black arrows). The microscopic images were taken at 5, 7, and 11 days post-inoculation (dpi). Bars = 100 µm.

The Impacts of ROS Upon Cell Death
Followed by H2O2, various physiological activities can be triggered to stop further pathogenic progression, those activities include callose deposition, and cell wall crosslinking [11,12,14,24,36]. Another biological process highly regulated by H2O2 is the programmed cell death (PCD) [11,12,16].
By observing the cotyledons treated with trypan blue staining (TBS), the bulk of stained senescent cells were visible from both Surpass400 and 01-23-2-1 cotyledons around the origins of inoculation at 7 dpi (Figure 3a,b). At 11 dpi, Surpass400 and 01-23- Westar at 11 dpi also showed some pycnidia production (red arrows). (b). The representative microscopic images taken from the origins of inoculation (magnitude: 50×), the images showed the details about H 2 O 2 accumulation when the fungus progressed from the puncture holes for inoculation. The brownish color shown from Surpas400 and 01-23-2-1 had more captured H 2 O 2 around the origins of inoculation (red arrows) and some pycnidia from Westar and Surpass400 were also captured (black arrows). The microscopic images were taken at 5, 7, and 11 days post-inoculation (dpi). Bars = 100 µm.
By observing the cotyledons treated with trypan blue staining (TBS), the bulk of stained senescent cells were visible from both Surpass400 and 01-23-2-1 cotyledons around the origins of inoculation at 7 dpi (Figure 3a,b). At 11 dpi, Surpass400 and 01-23-2-1 had a further enlargement of death regions, which was an enhanced situation to what happened at 7 dpi. From the microscopic images, the incompatible interaction did not hinder the hyphal formation of L. maculans fungus but formed a buffering zone with dead cells (Figure 3a, black arrows, Figure 3b, yellow arrows) to inhibit the chance for hyphae to invade more living tissues for nutrition. On the other hand, Westar only had hyphae (5 dpi) and pycnidia (7 and 11 dpi) formed around the punctured holes for inoculation (Figure 3b, red arrows), suggesting that the LCD was not observed in compatible interaction and this defense mechanism must be the feature for incompatible interaction (HR cell death) (Figure 3b).

Signal Allocation Patterns in ROS Production and Subsequent Responsive Factors among Susceptible, Intermediate and Resistant B. napus Plants
Triggered by oxidative burst, the plant body can trigger a series of defensive mechanisms including expression of responsive genes in hindering further pathogenic progression. These defensive mechanisms include the early apoplastic accumulation of ROS by membrane-bound NADPH oxidases [11].
As shown in Figure 4, Surpass400 and 01-23-2-1 exhibited an earlier induction of RBOHD and F compared with Westar. 01-23-2-1 showed the relatively high expression of RBOHD/F as early as 1 dpi. Surpass400 (both H75 8-1 and H77 7-2) did not show straightforward early induction of both RBOHD and F (1, 3 and 5 dpi). Remarkably, both Sur-pass400 H75 8-1 and H77 7-2 showed higher expression RBOHD at 5 dpi compared with Westar (Surpass400-H77 7-2 is not significant enough). For Westar, both genes were not expressed until 7 dpi and displayed a high expression level at 11 dpi. As early as 3 dpi, the blackleg fungus started to secrete cell wall degrading proteins (CWDBs) in B. napus [37]. Becker et al., (2017) [31] also found the early cell collapse in resistant B. napus (incompatible interaction against L. maculans) at 3 dpi, and at the same time point, genes related to SA and JA signals are also induced. The results indicated that RBOH enzyme may be an important factor to initiate ROS production during plant defense in B. napus since early defense against L. maculans seems to be one of the features for effective defense.

Signal Allocation Patterns in ROS Production and Subsequent Responsive Factors among Susceptible, Intermediate and Resistant B. napus Plants
Triggered by oxidative burst, the plant body can trigger a series of defensive mechanisms including expression of responsive genes in hindering further pathogenic progression. These defensive mechanisms include the early apoplastic accumulation of ROS by membrane-bound NADPH oxidases [11].
As shown in Figure 4, Surpass400 and 01-23-2-1 exhibited an earlier induction of RBOHD and F compared with Westar. 01-23-2-1 showed the relatively high expression of RBOHD/F as early as 1 dpi. Surpass400 (both H75 8-1 and H77 7-2) did not show straightforward early induction of both RBOHD and F (1, 3 and 5 dpi). Remarkably, both Surpass400 H75 8-1 and H77 7-2 showed higher expression RBOHD at 5 dpi compared with Westar (Surpass400-H77 7-2 is not significant enough). For Westar, both genes were not expressed until 7 dpi and displayed a high expression level at 11 dpi. As early as 3 dpi, the blackleg fungus started to secrete cell wall degrading proteins (CWDBs) in B. napus [37]. Becker et al., (2017) [31] also found the early cell collapse in resistant B. napus (incompatible interaction against L. maculans) at 3 dpi, and at the same time point, genes related to SA and JA signals are also induced. The results indicated that RBOH enzyme may be an important factor to initiate ROS production during plant defense in B. napus since early defense against L. maculans seems to be one of the features for effective defense.   Since PCD is one of the mechanisms of HR defense, the pathogen is not able to get enough nutrients to replicate when it is surrounded by dead cells [16,36,37]. ENHANCED DISEASE SUSCEPTIBILITY 1 (EDS1) and PHYTOALEXIN DEFICIENT 4 (PAD4) are found to play pivotal roles in R gene-mediated signaling of resistance [2,36]. By analyzing the expression of EDS1 (Figure 5), 01-23-2-1 (inoculated with H75 8-1 and H77 7-2) displayed higher expression at 3 and 5 dpi while Westar (with H75 8-1 and H77 7-2) had the peak expression at 11 dpi. Intermediate interaction for cotyledons of Surpass400 with H75 8-1 had early induction of the same gene at 3 dpi, and also exhibited up-regulation at 11 dpi like Westar. The onset patterns of EDS1 expression suggested that resistant interaction had earlier activation of EDS1, possibly due to the earlier recognition of the pathogen by gene-for-gene interaction.  (MPK3 and 6). The level of the bars are the expression levels from the inoculated cotyledons (three genotypes: Westar, Surpass400 and 01-23-2-1, two isolates: H75 8-1 and H77 7-2) normalized with the cotyledons inoculated with water (assuming that the expression of each studied gene in the cotyledons inoculated with water is 1). For time point, different lowercase letters suggest the significant differences among mean values (Fisher's Least Significant Difference; p < 0.05). The results are based on three replicates in three independent experiments.
Since PCD is one of the mechanisms of HR defense, the pathogen is not able to get enough nutrients to replicate when it is surrounded by dead cells [16,36,37]. ENHANCED DISEASE SUSCEPTIBILITY 1 (EDS1) and PHYTOALEXIN DEFICIENT 4 (PAD4) are found to play pivotal roles in R gene-mediated signaling of resistance [2,36]. By analyzing the expression of EDS1 (Figure 5), 01-23-2-1 (inoculated with H75 8-1 and H77 7-2) displayed higher expression at 3 and 5 dpi while Westar (with H75 8-1 and H77 7-2) had the peak expression at 11 dpi. Intermediate interaction for cotyledons of Surpass400 with H75 8-1 had early induction of the same gene at 3 dpi, and also exhibited up-regulation at 11 dpi like Westar. The onset patterns of EDS1 expression suggested that resistant interaction had earlier activation of EDS1, possibly due to the earlier recognition of the pathogen by gene-for-gene interaction.  Surprisingly, PAD4 did not show co-expression with EDS1. According to other studies, PAD4 and EDS1 interact with each other to trigger basal resistance and HR [38,39]. PAD4 did not have a high expression at 5 dpi in Surpass400 and 01-23-2-1 when expression of EDS1 peaked at this time point in these two genotypes. EDS1 can induce other resistant activities without PAD4 [38,39] and EDS1 is found to bind multiple factors in plant defense [40].
MPK3 and MPK6 are also found to support ROS signaling, these two factors also assist the production of camalexin and ethylene [27,41]. An ethylene-responsive factor, ERF6, was phosphorylated and activated by MPK3/MPK6 cascade to induce WRKY33. PDF1.1 and PDF1.2, two defensins to enhance plant defense, were also activated [42]. MPK3 and MPK6 displayed induction from Westar at 11 dpi (with both HCRT75 8-1 and HCRT77 7-2), while Surpass400 and 01-23-2-1 did not show very high expression ( Figure  5). Westar depends more on the expression of MPK3 and MPK6 expression levels at a later period (necrotrophic) of the infection, when the fungal progression was too severe that Surprisingly, PAD4 did not show co-expression with EDS1. According to other studies, PAD4 and EDS1 interact with each other to trigger basal resistance and HR [38,39]. PAD4 did not have a high expression at 5 dpi in Surpass400 and 01-23-2-1 when expression of EDS1 peaked at this time point in these two genotypes. EDS1 can induce other resistant activities without PAD4 [38,39] and EDS1 is found to bind multiple factors in plant defense [40].
MPK3 and MPK6 are also found to support ROS signaling, these two factors also assist the production of camalexin and ethylene [27,41]. An ethylene-responsive factor, ERF6, was phosphorylated and activated by MPK3/MPK6 cascade to induce WRKY33. PDF1.1 and PDF1.2, two defensins to enhance plant defense, were also activated [42]. MPK3 and MPK6 displayed induction from Westar at 11 dpi (with both HCRT75 8-1 and HCRT77 7-2), while Surpass400 and 01-23-2-1 did not show very high expression ( Figure 5). Westar depends more on the expression of MPK3 and MPK6 expression levels at a later period (necrotrophic) of the infection, when the fungal progression was too severe that the host needed massive ROS signaling and other defense activities to stop further infection. The high expression of MPK3 and MPK6 on Westar cotyledons at 11 dpi also linked to the expression of RBOHD and -F, suggesting that susceptible Westar cotyledons lately activated massive ROS signaling to stop the necrotrophic phase of L. maculans, the ROS molecules are able to exert multiple factors and signaling pathways to activate plant defense activities [15,20].

Discussion
In this article, the genotypes with stronger resistance Surpass400 and 01-23-2-1 exhibited earlier emergence of electrolyte leakage, H 2 O 2 diffusion and cell death, compared with susceptible control Westar. Moreover, ROS-responsive genes such as RBOHD/F tended to be activated from 01-23-2-1.
Electrolyte leakage has been found in many studies as the early physiological signal for stress response. It is also observed from plant tissues during hypersensitive response and cell death [18,19]. Ions such as K + and Ca 2+ are transported via ion channels to induce signals related to stress tolerance. The efflux of K + is found in various biological processes including PCD, ROS, stomata closure, and hormonal secretion [17,43]. Besides, another remarkable electrolyte Ca 2+ is originated from the vacuole and induced as the second signal when the MAMP/DAMP (Microbe/Damage-Associated Molecular Pattern) factors are precepted, and the defense signals also lead to PCD [17,20]. Thus, electrolyte leakage becomes a reliable measure of cell death and stress response. The early observation of electrolyte voltage from the inoculated cotyledons of Surpass400 and 01-23-2-1 implicated the early activation of defense response since the HR is found to trigger defense mechanisms including ion leakage, ROS signaling, hormonal signaling, etc. [17,31,37]. On the other hand, Westar samples started to have higher conductivity at 7 dpi, as suggested in Becker et al., 2017 [31], the RNA sequencing data revealed that the susceptible B. napus genotype triggered the same defense-related genes as the resistant genotype, however, the incompatible interaction activates the earlier expression of those genes compared with the compatible interaction, causing the different disease severity between susceptible and resistant genotypes.
Since L. maculans is hemibiotrophic, it undergoes the biotrophic stage first and then reaches the necrotrophic stage. Biotrophs usually exploit the nutrient from the living cells, it penetrates the plant cell wall and membrane with fungal structures such as haustoria and hyphae [44][45][46]. Evidence also showed that around the early stage of L. maculans infection upon B. napus, the fungus secretes cell wall degrading enzymes (CWDEs), and this physiological process is considered as one aspect of its pathogenicity [37,47]. Sexton et al. (2000) [37] reported that the highly virulent L. maculans races secrete the CWDEs at an early stage.
When blackleg fungus infects successfully, fungal hyphae develop in intercellular space during the biotrophic stage, and no obvious damage was made upon host cells [48]. Thus, early cell senescence becomes an effective strategy against biotrophic pathogens, to prevent further colonization and exploitation of host nutrients [16,45,48].
The early intensive diffusion of H 2 O 2 from Surpass400 and 01-23-2-1 connect their resistant responses against the fungus, the accumulation of brownish discoloration (i.e., H 2 O 2 ) around the origins of inoculation indicated a series of defensive responses including cell senescence from the host, which will hinder the further fungal growth. This may explain the similar pattern of cell death that occurred around the sites of inoculation, which was validated by trypan blue staining (TBS). The intermediate and resistant cotyledons tended to induce a protective region together with early hyphal development, so that the further intercellular penetration by the hyphae could be suppressed. The regional secretion of H 2 O 2 and cell death were also found from other HR cases, this also accompanies other defense responses such as papillae development and cell wall alteration [21,49]. Moreover, the co-existence of regional cell death and H 2 O 2 accumulation was also found from other examples of HR [50]; Nováková et al., (2015) [51] also suggested the potential function of H 2 O 2 in restricting L. maculans development in B. napus. The findings from DAB and cell death assays revealed that the two types of incompatible interaction, intermediate (Surpass400-H75 8-1) and resistant (Surpass400-H77 7-2 and 01-23-2-1-H75 8-1/H77 7-2) were able to induce intense early (5 dpi) H 2 O 2 accumulation and cell death as the priming defense to achieve effective defense against fungal proliferation on the plant tissues.
ROS generation and signaling play versatile roles in stress tolerance in the plant body. The superoxide (O 2 − ) molecules are initially produced by NADPH oxidases or respiratory burst oxidase homologues (RBOH's) and converted into hydrogen peroxide (H 2 O 2 ) by superoxide dismutases (SOD's) [14,34]. The stable and membrane-permeable properties make H 2 O 2 molecules able to induce systemic responses against various biotic and abiotic stresses [12]. RBOHD and F are two the NADPH oxidases inducing ROS accumulation during plant defense response [52,53]. RBOHD and F are the two NADPH oxidases that have been well studied in Arabidopsis thaliana defense [30,[52][53][54]. Calcium leakage, reactive oxygen intermediates (ROI) and peroxide were reduced in rbohD, rbohF, and rbohD/rbohF double mutants [53]. RBOHs are regarded as the central factors to trigger ROS signaling in plant cells [11,15,54]. RBOHD initiates the cell-to-cell ROS signaling which is called "ROS wave", by transmitting H 2 O 2 extracellularly. Evidence suggested that RBOHD was involved in early acute ROS signaling in defense and tolerance against various challenges [12,15], and RBOHD plays important roles in ROS production when the host recognizes the pathogen successfully on the site of infection [52]. RBOHD and F working together can fully activate basal resistance, the mutation of both genes abolishes ROS production and makes it easier for pathogens to infect [52,53]. As shown in Figure 4, the early expression of RBOHD from Surpass400 and 01-23-2-1 implied that this gene might also play crucial roles in B. napus such as in Arabidopsis during pathogenic infection. There are also some differences between the onset pattern between RBOHD and F, as such, RBOHF did not show high expression at 5 dpi in Surpass400, and 01-23-2-1 cotyledons were found to have the most pronounced up-regulation at 5 dpi while for RBOHD, the gene was up-regulated earlier at 1 dpi. These two genes were found to be regulated differently in Arabidopsis thaliana, and RBOHD plays more dominant roles in activation against pathogenic invasion [52]. It is also noted that the Westar genotype also displayed high expressions of both RBOH genes during necrotrophic stage (11 dpi). When infected, both compatible and incompatible interactions can trigger an oxidative burst. Therefore, it is normal to see massive a regulation of ROS-related genes when the plant tissue is heavily infected, however, the timing of the coordination of various regulators seems to be more important. By analyzing host-cell-wall-degrading enzymes (CWDEs) from the pathogen, Sexton et al., (2000) [37] implied that early restriction of fungal development is a crucial factor for B. napus cotyledon to achieve resistance towards L. maculans.
ROS signaling plays important roles in lesion development and cell senescence on plants, on the other hand, other factors such as JA signaling at the same time, can also attenuate to prevent the excessive damage by ROS [19,29]. Moreover, Becker et al., (2017) [31] also listed various types of genes which were activated from resistant B. napus genotype at 3 dpi. They include the factors in pathogen perception, callose deposition, sulfur metabolism, and lignification, whereas at the same time point, genes related to the negative regulation of plant defense and senescence were also activated. It seems that the resistant genotypes trigger massive signals from both up-regulation and down-regulation sides of defense at an early stage of infection, which hinders the fungal development during hyphal stage. On the other hand, during late stage (necrotrophic), the pathogen colonizes too widely. Thus, it is impossible for the host to achieve effective resistance, and the defense signals including ROS tend to express in a large amount to halt the further development, which produced such high levels of RBOH genes at 11 dpi in Westar.
Therefore, the high amount of the fungal cells pushed the host to evoke more defense signals to cope with the pressure of self-defense, similar to the cases in susceptible Westar and intermediate Surpass400-H75 8-1 after 7 dpi. Surpass400-H75 8-1, as the intermediate interaction, displayed both resistant and susceptible traits, as the samples taken from the inoculation pair both had the anticipated activation of RBOHD expression and H 2 O 2 (same as resistant interaction), and the late induction of electrolyte leakage and EDS1 expression (same as susceptible interaction).
On the other hand, the absence of co-expression between PAD4 and EDS1 was not expected. According to previous studies, the expression of PAD4 is dependent upon EDS1, the interaction between EDS1 and PAD4 seems to enhance the HR by further SA accumulation [38,55]. In protein level, EDS1 and PAD4 interact each other, and trigger R gene-related resistance [38,39,46,56]. However, the function of EDS1 is not totally dependent on PAD4, EDS1 is also able to dimerize itself (i.e., EDS1-EDS1 interaction) or bind with another PCD factor SENESCENCE ASSOCIATED GENE 101 (SAG101), moreover, those types of interaction contribute to innate immunity [38,39,55]. EDS1 itself also triggers partial R gene-related defense and SA accumulation [38,55]. Therefore, in this study, EDS1 is highly expressed in defense response alone, without the cooperation with PAD4.
Finally, yet importantly, there was no strong trend of the early activation of MPK3/6 from Surpass400 and 01-23-2-1. Besides gene expression, the function of MAPK factors is also related to phosphorylation, which activates downstream defense factors [42,57]. It is necessary to postulate that early ROS activation in B. napus might promote more on phosphorylation than expression.

Pathogen Cultivation
Leptosphaeria maculans isolates were cultured on V8 juice (Campbell's, Camden, NJ, USA) at room temperature under the fluorent tube light. The isolates were cultured for 10 to 14 days to obtain a sufficient amount of pycnidiospores. Each plate was scraped off and washed with 2 mL of distilled water to collect pycnidiospores and make inoculum stock solutions. The stock solutions were stored at −20 • C.
The cotyledons of B. napus cultivars were inoculated 7 days after sowing (seedling stage) by puncture inoculation. Each lobe of cotyledons was punctured by a sterile needle twice from each side, to have 4 inoculation points on each seedling of the canola plant.

Electrolyte Leakage Measurement
The cotyledons (6 cotyledons from 3 biological replicates) were cut into small leaf disks (round, 5 mm in diameter) with the cork borer. The leaf disks were washed for 30 min in 10 mL ultrapure water and transferred into another round of fresh ultrapure water (25 mL). After 5 h, the electrolyte leakage was measured in voltage from the soaked ultrapure water by the VWR sympHony conductivity meter (Radnor, PA, United States).

3,3 -Diaminobenzidine (DAB) Staining
The DAB staining solution was prepared by dissolving 40 mg of DAB (Sigma-Aldrich, St. Louis, MO, USA) in 200 µL of dimethylformamide in 40 mL of water. Cotyledons were soaked in the staining solution in the dark and shaken overnight. The stained cotyledons were discolored by 95% ethanol. The experiment is followed by the protocol of Sašek et al. (2012) [6].

Trypan Blue Staining (TBS)
The trypan blue stock solution was prepared by mixing 10 mL of phenol, 10 mL of glycerol, 10 mL of lactic acid, 10 mL of water, and 0.02 g of trypan blue powder (Sigma-Aldrich, St. Louis, MO, USA). The working solution was prepared by dissolving the stock solution with ethanol (96%; 1:2, v/v). B. napus cotyledons were soaked in the working solution and boiled in a water bath for 1 min, incubated in the solution overnight and washed in chloral hydrate solution (2.5 g of choral hydrate in 1 mL of distilled water).

Gene Expression Analysis
Frozen cotyledons (12 cotyledons, 6 seedlings, 3 biological replicates) were ground in liquid nitrogen using pestles and mortars. Total RNA was extracted with TRI reagent (Sigma-Aldrich, St. Louis, MO, USA). Total RNA was purified by DNase I treatment with DNase I recombinant, RNase-free kit (Roche). Purified RNA was used to synthesize cDNA with the GOScript Reverse Transcription System (Promega). The cDNA stock solution was diluted into a concentration of 100 ng/µL. The quantitative-PCR was performed by mixing 1 µL of cDNA (100 ng) into the 10 µL reaction system of IQ TM SYBR ® Green Supermix (BioRad, Hercules, CA, USA).
The qPCR program used for all analyzed genes was: 95 • C for 3 min; followed by 39 cycles of 95 • C for 15 sec, and 60 • C for 20 sec; followed by a melting curve analysis.
All qPCR primers are listed in Supplementary Table S1. The relative level of gene expression was analyzed with the 2 −∆∆CT method described by Livak and Schmittgen, (2001) [58]. The expression of the studied genes was normalized by the house-keeping gene ACTIN (NM_001316010.1).

Statistical Analysis
Unless specified, the analyses of samples used at least three biological replicates. The statistical analyses were performed using the Fisher's Least Significant Difference (LSD) method with the SAS 9.4 software (SAS Institute, Cary, NC, USA) [59]. The Fisher's LSD was applied to gene expression and electrolyte leakage measurement, in order to observe the effectiveness of resistance in the three genotypes when inoculated with two isolates.

Conclusions
The data from this article revealed that ROS metabolism and signaling played potential roles in the host-microbe interaction in the B. napus-L. maculans pathosystem. Intermediate and resistant genotypes displayed intense hydrogen peroxide (H 2 O 2 ) diffusion and cell death around the site of inoculation. Moreover, ROS/PCD-responsive genes tended to express earlier in the intermediate and incompatible interactions. Those findings suggested that earlier activation of ROS-related defense mechanisms is an essential component of effective resistance in B. napus against L. maculans.