Effect of Elevated CO2 Concentration on the Disease Severity of Compatible and Incompatible Interactions of Brassica napus–Leptosphaeria maculans Pathosystem

Global warming by increased atmospheric CO2 concentration has been widely accepted. Yet, there has not been any consistent conclusion on the doubled CO2 concentration that in the future will affect plant disease incidence and severity. Blackleg disease, mainly caused by Leptosphaeria maculans, is a major disease on canola production globally. Brassica napus and L. maculans have a gene-for-gene interaction, which causes an incompatible reaction between canola plants carrying resistance genes and L. maculans isolates carrying corresponding avirulence genes. In this study, B. napus varieties and lines inoculated with different Leptosphaeria isolates were subjected to simulated growth conditions, namely, growth chambers with normal environments and with controlled CO2 concentrations of 400, 600, and 800 ppm. The results indicated that the elevated CO2 concentrations have no noticeable effect on the inferred phenotypes of the canola–blackleg interactions. However, the disease severity decreased in most of the B. napus–L. maculans interactions at extremely high CO2 concentration (800 ppm). The varied pathogenicity changes of the B. napus–L. maculans pathosystem under elevated CO2 concentrations at 400 or 600 ppm may be due to the genetic background or physiological differences in plants and pathogenicity differences in L. maculans isolates having different Avr gene profiles. The mechanisms by which elevated CO2 concentrations affect the B. napus–L. maculans pathosystem will help us understand how climate change will impact crops and diseases.


Introduction
Climate change is an important dynamic that will affect food production globally. The atmosphere's elevated concentration of CO 2 , one of the most important climate change influences, is due to anthropogenic emissions [1]. The concentration of CO 2 has been increasing and is predicted to double at the end of this century by the Intergovernmental Panel on Climate Change (IPCC) [2]. Levels will climb from the current level of 380 µmol/mol to 730-1020 µmol/mol [3]. The effects of rising CO 2 concentrations on crop growth and physiological processes have been extensively studied [4][5][6][7]. For instance, rising CO 2 can enhance the physiological performance of Brassica napus seedlings under optimal water supply [8]. Li et al. (2017) reported that increased CO 2 could alter green tea quality by the stimulation of primary and secondary metabolism [9]. They found that the higher concentration of CO 2 could enhance photosynthesis, C:N ratio, Rubisco carboxylation activity, water-use efficiency, and final yield or quality [6,[8][9][10][11].
Plant disease symptoms in pathosystems are influenced by, or require, three components-a susceptible host, an aggressive or virulent pathogen, and a conducive environment for the pathogen to cause disease [12,13]. Thus, the variations in environmental conditions, including CO 2 concentration change, may potentially affect plant disease susceptibility and severity. A couple of studies have investigated the elevated CO 2 concentration effects on plant-pathogen interactions. The exposure of tomato plants to elevated CO 2 led to a lower disease incidence and severity caused by tomato (Solanum lycopersicum) mosaic virus (TMV) and Pseudomonas syringae. However, the plant's susceptibility to necrotrophic Botrytis cinerea was increased [14]. Ferrocino et al. (2013) found that increased CO 2 concentration had no detectable impact on the Fusarium wilt of lettuce (Lactuca sativa) or the abundance of Fusarium spp. [15]. The leaf spot disease incidence and severity always increased in redbud (Cercis canadensis) and sweetgum trees (Liquidambar styraciflua) under increased CO 2 concentration in a five-year survey [16]. Elevated CO 2 reduced the disease incidence and severity significantly in the red maple fungal pathogen Phyllosticta minima [13]. The species diversity of foliar fungal plant disease decreased when the CO 2 concentration increased [17]. The disease incidence and severity of rocket plants (Eruca sativa) caused by Fusarium oxysporum increased under elevated CO 2 conditions with controlled temperatures [18]. Considering the above, it is difficult to find a consistent pattern to the effects of elevated CO 2 concentration on disease incidence and severity in plants.
Canola (Brassica napus) is one of the most important cash crops in Canada, reaching a production of 20.3 MMT in 2018 (Canola Council of Canada). Blackleg, caused by the fungal pathogen Leptosphaeria maculans, is one of the most important and devastating diseases in canola worldwide, which causes significant yield loss [19]. The incidence and severity of blackleg disease in canola vary with the geographic distribution, varieties, crop rotation, and climate conditions. This disease has been controlled historically by crop rotation and the utilization of canola varieties with major gene resistances [20,21]. A gene-for-gene interaction has been reported between B. napus with a certain R gene and its corresponding avirulence gene (Avr) in the L. maculans isolate [22,23]. This pathosystem brings incompatible and compatible interactions in the canola plant, with and without a certain R gene, respectively, that can defend against the L. maculans pathogen with the corresponding Avr gene. Currently, a total of 16 Avr genes have been identified in L. maculans [24][25][26][27][28]. Corresponding to the Avr genes in L. maculans, various major resistance genes in Brassica spp. have been identified [29][30][31][32][33]. Apart from the qualitative resistance (major gene resistance), the effect of quantitative resistance against an isolate of L. maculans in selected Canadian canola cultivars under increased temperature was investigated [34]. Very few studies have reported on the effects of environmental changes, especially the CO 2 concentration, on the interactions of the B. napus-L. maculans pathosystem. In this study, we conducted experiments in controlled-environment rooms with three different CO 2 concentrations of 400, 600, and 800 ppm. The aims were to understand the effects of elevated CO 2 on (1) the compatible and incompatible interactions between B. napus and L. maculans isolates; (2) the disease severities of the susceptible and resistant canola varieties; and (3) the pathogenicity of different isolates carrying different Avr gene races.

Inferred Phenotyping
The mean rating scores and their inferred disease resistance for the seedlings of B. napus varieties and lines had different variations based on the inoculum and CO 2 concentrations (Table 1). L. maculans isolate D5, which carries AvrLm1-2-4-7-S-LepR1-LepR2, showed susceptible (S) and intermediate resistance (IR) on B. napus line 1065 under CO 2 concentrations of 400 and 800 ppm, respectively, whereas it showed a resistance (R) reaction under a normal environment (NE) and a CO 2 concentration of 600 ppm. The inferred plant phenotypes of L. maculans isolate D10 (AvrLm5-6-7-LepR1) inoculated onto B. napus varieties and lines 1135 and 01-23-2-1 carrying LepR2 and Rlm7 genes, respectively, were changed from susceptible into intermediate resistance at the highest CO 2 concentration (800 ppm).
The DM118 isolate showed a high disease rating score of 6.00 (IR) at the 600 ppm concentration of CO 2 when inoculated onto the B. napus 1065 line. The disease rating score was decreased to 5.25 under 800 ppm CO 2 when the DS103 isolate was inoculated onto B. napus line 1065. A CRISPR mutant isolate umavr7 produced from DS103 showed intermediate resistance (5.25) in B. napus variety or line 1065 under 400 ppm, reduced from a susceptible reaction (8.33, NE; 6.67, 600 ppm; 8.00, 800 ppm). In addition, it was also reduced when CO 2 concentrations increased from 400 to 800 ppm (Table 1). Usually, the L. biglobosa isolate is considered as non-aggressive or less virulent to B. napus. In this study, we found that L. biglobosa showed intermediate resistance on B. napus line 1135 under a normal environment and 600 ppm CO 2 , and under a normal environment in both B. napus varieties and lines 02-22-2-1 and Goé Land (Table 1). The L. biglobosa isolate, which is avirulent to the canola plants, did not show any significant lesion size change under different CO 2 concentrations on B. napus variety Westar. The lesion size decreased slightly when the CO 2 concentration was increased to 800 ppm ( Figure 1). All the other seven involved L. maculans isolates showed virulence and clear disease symptoms on B. napus variety Westar. The cotyledon lesion size caused by L. maculans isolate D10 was significantly decreased under 400 ppm of CO 2 . The largest lesion size caused by DM118 was under 600 ppm of CO 2 . This was significantly larger than those produced under a normal environment, 400 ppm, and 800 ppm. The D5 isolate caused a significantly larger lesion size when the CO 2 concentration increased to 800 ppm. In contrast, the lesion size caused by isolate DS103 was significantly smaller under 800 ppm ( Figure 1). The lesion size caused by L. maculans isolates CDS-13, DM96, and umavr7 had no significant change under the four different CO 2 concentrations (Figure 1).

Pathogenicity Evaluation of Compatible and Incompatible Interactions between Leptosphaeria spp. and B. napus
The two isolates umavr7 and L. biglobosa, which have virulence and avirulence, respectively, to all studied B. napus varieties and lines, were used to investigate the compatible and incompatible reactions and the effect of CO 2 concentrations (Table 1 and Figure 2). The lesion sizes were significantly decreased on B. napus varieties and lines Jet Neuf (Rlm4) and Goé Land (Rlm9) inoculated by umavr7 as the CO 2 concentrations increased ( Figure 2A). umavr7 caused significantly smaller lesions on B. napus line 02-22-2-1 (Rlm3) under 600 and 800 ppm of CO 2 compared with the normal environment and 400 ppm. In B. napus line 1065 (LepR1), the lesion size was significantly decreased under 400 ppm of CO 2 . The lesion sizes were all smaller in B. napus line 1065 under elevated CO 2 concentrations compared to those in the normal environment ( Figure 2A). However, the increased CO 2 concentrations enhanced the pathogenicity of umavr7 significantly when inoculated onto B. napus line 01-23-2-1 (Rlm7) (Figure 2A). There were no significant changes in lesion size when the umavr7 isolate was inoculated onto B. napus varieties and lines Westar and 1135 under different CO 2 concentrations (Figures 1 and 2A).
For the incompatible interaction, the lesion sizes were decreased under the highest CO 2 concentration (800 ppm) when inoculated onto seven B. napus varieties and lines, except 01-23-2-1 (Rlm7), compared to those under the normal environment ( Figure 2B). For example, the lesion sizes were significantly smaller under 800 ppm CO 2 when inoculated onto B. napus varieties and lines 1135, Jet Neuf, and 1065 ( Figure 2B). The smallest lesions caused by the L. biglobosa isolate from B. napus line Goé Land and 02-22-2-1 were under CO 2 concentrations of 600 and 400 ppm, respectively ( Figure 2B). As in the compatible interaction, the higher CO 2 concentration enhanced the pathogenicity of the L. biglobosa isolate when inoculated onto B. napus line 01-23-2-1 (Figure 2A Figure 3E,F). The lesion sizes were decreased at 400 ppm and then increased at 600 ppm when DM118 was inoculated onto B. napus varieties and lines 1135, Jet Neuf, and Goé Land ( Figure 3E). DS103 caused a steady decline of lesion sizes on most of the B. napus varieties and lines when the CO 2 concentration was elevated ( Figure 3F).

Discussion
Plants usually benefit by increasing the photosynthetic products and/or utilization efficiencies of water and nutrients when growing under rising temperatures and CO 2 concentrations [35]. However, the effects of elevated CO 2 concentration on plant-pathogen interactions have undergone limited studies, producing inconsistent results [13,[15][16][17][18]36]. In this study, we aimed to investigate how elevated CO 2 concentration affects the canola-blackleg interaction using different B. napus varieties and lines with different R genes and Leptosphaeria isolates with various Avr gene profiles. We found that the CO 2 concentration has no impact on the inferred phenotypes, which are determined by the qualitative resistance (R gene) in canola. However, the extremely high CO 2 concentration (800 ppm) inhibits the pathogenicity of Leptosphaeria spp. on the canola plant with the exception of D5 isolate on the susceptible B. napus variety Westar. The variability of the enhanced or inhibited pathogenicity in B. napus-L. maculans interactions will provide a starting point for future studies on the physiology and genetics of certain B. napus lines in response to fungal disease under different CO 2 concentrations.
B. napus has both qualitative (R gene resistance in the seedling) and quantitative (adult plant) resistance to blackleg disease [37]. Currently, most of the canola cultivars grown in Canadian canola fields have single or multiple R genes (Canola Council of Canada). Therefore, the main resistance of canola seedlings to blackleg disease is determined by the qualitative resistance (R gene(s)) [38,39]. There were slight changes in the disease rating scores of different B. napus varieties and lines in responding to various Leptosphaeria spp. isolates under different elevated CO 2 concentrations, and these did not shift the inferred phenotypes of the interactions in this study. This result supports the conclusion that the R gene-mediated resistance is the major factor for canola seedlings in conferring the environmental element variations (i.e., elevated CO 2 concentration). This is similar to a previous review that indicated that the abiotic stress tolerance may or may not correlate with the plant's resistance to disease [40]. There was no change of inferred phenotypes of the susceptible B. napus variety Westar conferred by different Leptosphaeria isolates under different CO 2 concentrations. The interactions between canola varieties and lines carrying different R genes and Leptosphaeria isolates had slight changes, that is, from susceptible to intermediate resistance/susceptible or from resistance to intermediate resistance, indicating that physiological stresses and variations occurred in the host plant or that the fungal pathogen may affect the disease severity aside from the major R gene mediation.
Global climate warming caused by the elevated CO 2 concentration is well documented and widely accepted [2]. Although the elevated CO 2 concentration has no impact on the major R gene-determined resistance of canola seedlings, it should have profound effects on plant and fungi growth, development and reproduction. Therefore, the degree of disease severity quantified by lesion size had significant variations under different CO 2 concentrations. The main finding of this study was that the lesion sizes were significantly smaller in most of the B. napus varieties inoculated with different L. maculans isolates at extremely high CO 2 concentration (800 ppm). The disease incidence or severity was decreased under elevated CO 2 concentrations in previous studies [13,[41][42][43][44]. For example, resistance was induced at 700 ppm and associated with the canopy size in the Fitzroy-Colletotrichum gloeosporioides pathosystem [41]. McElrone et al. (2005) found that elevated CO 2 significantly reduced the disease incidence and severity of a red maple (Acer rubrum) fungal pathogen (Phyllosticta minima) through the changes in leaf chemistry and host physiology, that is, stomatal conductance reduction and smaller openings for fungi germ tubes or reduced nutritive quality [13]. Conversely, the elevated CO 2 concentration increased the disease incidence and severity of rice blast (Magnaporthe oryzae) and sheath blight (Rhizoctonia solani) in rice (Oryza sativa), probably because a higher number of tillers was observed, which may increase the chance for sclerotia to adhere to the leaf sheath [45]. The leaf spot disease incidence and severity in redbud (Cercis canadensis) and sweetgum trees (Liquidambar styraciflua) were increased under elevated CO 2 concentration, since the enhanced photosynthetic efficiency in the remaining leaf tissues minimized or mitigated any increases of the disease symptoms [16]. An interesting study conducted by Eastburn et al. (2010) found that increased CO 2 concentration significantly reduced and increased the disease severity of downy mildew and brown spot disease in the soybean, respectively [42]. They also found that CO 2 or O 3 treatment had no effects on the incidence of sudden death syndrome (SDS) of the soybean [42]. All these findings indicated that the effects of elevated CO 2 concentration on the disease incidence and severity depended on the plant and pathogen species. The effects of elevated CO 2 concentration on physiological processes, leaf chemistry, the stomatal opening system, and even on L. maculans growth rate, sporulation and aggressiveness, probably caused the decreased disease severity of canola seedlings inoculated by different L. maculans isolates. Further studies are needed to investigate causes of the inhibition of Leptosphaeria isolates' pathogenicity on the canola plant under elevated CO 2 concentration and at the molecular level.
There were different lesion size variation patterns for the canola plants inoculated with different L. maculans isolates carrying different Avr gene profiles under CO 2 concentrations of 400 and 600 ppm. For example, compatible interactions (i.e., CDS-13 on B. napus varieties and lines 02-22-1 and Goé Land; D5 on 1135; D10 on B. napus varieties and lines 1135, 02-22-2-1, and 01-23-2-1; DM118 on B. napus varieties and lines Westar, 1135, and Goé Land) showed largest lesion sizes at an elevated CO 2 concentration of 600 ppm. Several studies reported that elevated CO 2 concentration stimulated fungal growth, aggressiveness, sporulation, and fecundity [13,[46][47][48]. The increased disease expression at elevated CO 2 concentration may induce the larger lesions in certain plant species, varieties, or lines. This may be the reason that L. maculans isolates displayed increased pathogenicity on B. napus varieties and lines specifically at elevated CO 2 concentration. However, several compatible interactions showed the highest lesion size at a CO 2 concentration of 400 ppm or a normal environment. For incompatible interactions, the disease expression also increased at different CO 2 concentrations (normal environment, 400, or 600 ppm), and the disease severity mostly was inhibited at extremely high CO 2 concentration (800 ppm) ( Figure 3). Therefore, the increased pathogenicity of Leptosphaeria spp. observed here did not produce a consistent expression pattern under certain CO 2 concentrations, indicating that the disease development could result from the characteristics of Leptosphaeria or other changes in the host plant. It is well known that L. maculans isolates usually carry different Avr gene profiles, thus they show different pathogenicity and interactions with canola plants. L. maculans isolates carrying more Avr genes may show less virulence on the B. napus varieties and lines since they only cause disease symptoms in the varieties and lines without the corresponding resistance genes. Therefore, the effects of elevated CO 2 concentration on the effector or elicitor gene in Leptosphaeria may differ and induce different disease severity in canola plants. On the other hand, differences in the host (canola) plant traits (i.e., genetic background, resistance genes, plant morphology, and leaf and cotyledon structure) probably initiate various defense mechanisms under high CO 2 concentration stresses. Long-term studies including fungal genetics and biology, plant physiology, and genetics, etc., are needed to disclose the mechanisms of B. napus and Leptosphaeria spp. interactions under elevated CO 2 concentration.
Global warming at present and in the future could reduce the disease severity of the canola-blackleg pathosystem through elevated CO 2 at an extremely high level of 800 ppm. However, in this study, the pathogenicity of Leptosphaeria spp. seems increased in a couple of interactions with different B. napus varieties and lines at 400 ppm or 600 ppm CO 2 . We cannot make specific predictions on the change patterns from field trials since the CO 2 concentration can be easily controlled only in a growth chamber. The B. napus varieties and lines used in this study are differential lines used for the characterization of Avr gene profiling in L. maculans isolates. We were trying to understand the effects of elevated CO 2 concentration on different resistant gene (R) in response to various L. maculans isolates harboring different Avr genes. Since the most commercial canola varieties in fields in Canada have the Rlm3 gene [38], it is not practical to use cultivars with one R gene to represent the effects of elevated CO 2 concentration. However, the results obtained from this study provide a clue of how elevated CO 2 concentration affects the R-Avr interactions of the B. napus-Leptosphaeria maculans pathosystem in both compatible and incompatible reactions. The numbers of B. napus varieties and lines and L. maculans isolates employed in this study will also trigger a discussion of how the different canola plant varieties show different resistance mechanisms in response to the different Leptosphaeria spp. infections under elevated CO 2 concentration.

Brassica napus Varieties and Lines and Leptosphaeria maculans Isolates
B. napus varieties and lines carrying known R genes were used to test the compatible and incompatible interactions with different isolates (Table 1). For example, B. napus line 02-22-2-1, having the Rlm3 gene, showed compatible interaction with the isolate without the AvrLm3 gene, while showing incompatible reaction with the isolate carrying the AvrLm3 gene. Accordingly, several other B. napus varieties and lines (i.e., Jet Neuf, 01-23-2-1, Goé Land, 1065, and 1135 harboring Rlm4, Rlm7, Rlm9, LepR1, and LepR2, respectively) were included in this study (Table 1). These B. napus varieties and lines from different countries (Canada, France, and others) were extensively used for the characterization of Avr gene profiles in L. maculans isolates [38]. A susceptible B. napus variety Westar, which is a spring canola variety that has none of the resistant genes to blackleg disease, was also included as a control (Table 1).

Plant Cotyledon Inoculation and CO 2 Treatment
The Leptosphaeria spp. isolate inocula were harvested by flooding eleven-day-old cultures from a single pycnidiospore using distilled water. The single pycnidiospore was picked from culture of a stocked paper disc of each isolate [38,39]. The final concentration of inoculum of each isolate was adjusted to 2 × 10 7 spores/mL for the later cotyledon inoculation test. The seven B. napus varieties and lines with or without different R genes were seeded into 96-well flats (53 × 27× 7 cm as length, width and height) filled with commercialized soil (Pro-Mix BX, Premier Tech, Rivière-du-Loup, QC, Canada) and placed in a growth chamber at 16 • C (night) and 21 • C (day) with a 16 h photoperiod/day. The seven-day-old seedlings were used for inoculation and grown in different CO 2 concentrations. The cotyledons were punctured and inoculated with a prepared 10 µL inoculum droplet at each of two wound sites per cotyledon [38]. Therefore, at least 24 wound sites from 6 plants were inoculated and rated at 14 days post-inoculation.
Four flats of each of the B. napus varieties and lines were inoculated with different Leptosphaeria isolates. After one or two hours drying, one flat of one B. napus variety or line was moved to the growth chamber with normal growth conditions after inoculation of Leptosphaeria isolates. To investigate the elevated and extremely high CO 2 concentration effects on disease severity, the other three flats with each of the B. napus varieties and lines were placed into controlled-environment growth chambers with different CO 2 concentrations of 400, 600, and 800 ppm, respectively, with a 16 • C/21 • C night/day and a 16 h photoperiod/day (Conviron, Winnipeg, MB, Canada). The CO 2 concentration of each growth chamber was monitored by CO 2 /Temp./RH Data Logger (Professional Instruments, Huizhou, Guangdong, China). All the experiments were repeated once.

Disease Rating and Lesion Size Measurement
Based on the lesions, chlorosis, necrosis and signs of pycnidia, a rating scale from 0 to 9 was used to evaluate the disease severity at 14 days post-inoculation [38,39]. An average score was calculated from 24 inoculation sites of six plants (four wound sites per plant). Therefore, an average rating score ≤ 4.5 was considered as a resistant (R) reaction, 4.6 to 6.0 as an intermediate resistance (IR) reaction and 6.1 to 9.0 as a susceptible (S) reaction [38,49]. The lesion sizes of the infected cotyledons, under different CO 2 concentrations for each variety or line with different isolates, were quantified at 14 dpi using Assess 2.0 image software (American Phytopathological Society, St. Paul, MN, USA).

Statistical Methodology
Statistical analysis of the lesion sizes caused by the different isolates under different CO 2 concentrations, and the F measurement of distance between two repeats, were conducted using SAS version 9.4 (SAS Institute, Inc., Cary, NC, USA). The data were subjected to ANOVA and the mean lesion sizes were compared by Tukey's HSD studentized range test at p < 0.05.