Response of Alnus glutinosa to Phytophthora bark infections at ambient and elevated CO 2 levels

showed a decreasing trend in all inoculated plants with no signiﬁcant differences between Phytophthora species at both CO 2 levels. Chemical analyses of root samples showed high variability in sugars and phenolic compounds related to the plant’s health status. Discussion: This is the ﬁrst study to examine the response of alder seedlings to Phytophthora pathogens at different CO 2 levels. The ﬁndings demonstrate high aggressiveness of P . × alni and P. uniformis and weaker aggressiveness of P. plurivora to alder seedlings regardless of the CO 2 level.

Introduction: Mortality of the riparian alder population caused by Phytophthora pathogens has been studied for over years throughout Europe, recently gaining more importance in the context of evident climate change.The main objective of this study was to examine the pathogenicity of species from the "Phytophthora alni complex" present in the Czech Republic (P.×alni and P. uniformis) and P. plurivora to Alnus glutinosa seedlings grown at ambient and elevated CO concentration.
Methods: An underbark inoculation test was performed with seedlings grown from seeds collected from two Czech alder populations, one su ering from severe Phytophthora decline and the other disease-free.

Results:
The results showed significant di erences in lesion development and seedling mortality.After a -week experimental period, at both CO levels P. ×alni and P. uniformis showed high aggressiveness to A. glutinosa seedlings causing lesions of variable sizes and mortality of .%, and .% of plants, respectively.In contrast, P. plurivora did not cause mortality to any plant, and lesion sizes did not di er significantly from those in control plants.Physiological measurements did not reveal any significant di erences between Phytophthora species except for plants inoculated with P. plurivora showing increased values in specific physiological parameters weeks post-inoculation.Net photosynthesis decreased over the measurement period in all treatments with significant di erences found between measurements conducted and weeks after the inoculation.Transpiration showed a decreasing trend in all inoculated plants with no significant di erences between Phytophthora species at both CO levels.Chemical analyses of root samples showed high variability in sugars and phenolic compounds related to the plant's health status.
Surveys in several European countries demonstrated that PAA is the most frequently isolated species, whereas PAU and especially PAM are comparatively rare (Nagy et al., 2003;Jung and Blaschke, 2004;Aguayo et al., 2013).Accordingly, pathogenicity tests revealed significant differences between species within the "P.
alni complex" with PAA being more aggressive to A. glutinosa than PAU when infecting black alder (Alnus glutinosa) (Brasier and Kirk, 2001;Santini et al., 2003).Interestingly, PAU was shown to play an increasing role in alder dieback with increasing latitude in Scandinavia and increasing altitude in the Alps which is related to the lower frost tolerance of PAA as compared to PAU (Schumacher et al., 2006;Černý and Strnadová, 2010;Redondo et al., 2015;Corcobado et al., 2023).This is particularly important in the context of ongoing climatic changes and potential spread of the aggressive PAA strains to the currently unfavorable climatic regions.
Climatic changes are characterized by shifts in annual rain patterns, increased frequencies of extreme weather events, and the increase of average annual temperatures and atmospheric CO 2 concentration (Moore et al., 2015).The most rapid increase of CO 2 concentration was recorded during the second half of the 20 th century and the beginning of the 21 st century (Belmecheri and Lavergne, 2020).According to the Intergovernmental Panel on Climate Change (IPCC) the projected concentration of CO 2 in the year 2100 will range in different scenarios from 540 to 970 ppm (IPCC, 2001).Pessimistic scenarios predict an increase of CO 2 levels to 900-1,000 ppm by the end of the 21 st century (Van Vuuren et al., 2011).
Elevated CO 2 levels can have a significant impact on hostpathogen interactions (Percy et al., 2002;McElrone et al., 2005).Generally, a stronger resistance against plant pathogens is predicted in relation to elevated CO 2 level in the atmosphere, as the excess of carbon can promote the production of plant defensive secondary chemicals (Kazan, 2018).However, plants grown at elevated CO 2 can often alter constitutive and induced defense-related phytohormone levels, reflecting complex and dynamic biochemical interactions (Bazinet et al., 2022).So far, existing studies reported positive, negative, and neutral effects of elevated CO 2 level on plant immune responses to fungal pathogen infection (Smith and Luna, 2023).The majority of studies on plant-biotic interaction investigate above-ground plant parts and related fungal pathogens (Kazan, 2018).However, only a limited number of studies have been performed on Phytophthora pathogens and underbark or root infections.These studies showed both positive and negative effects of elevated CO 2 level on disease incidence depending on the Phytophthora and host species and the length of the experiment (Jwa and Walling, 2001;Fleischmann et al., 2010;Tkaczyk et al., 2014;Oszako et al., 2016;Milanović et al., 2020).
defense is suppressed, as Phytophthora deactivates salicylic-acid response genes, resulting in the reduction of lignin, phenols, and PR proteins in infected roots compared to healthy ones (Osswald et al., 2014).
Changes in physiological parameters are reflected in saccharides translocation.For example, in beech seedlings inoculated by PLU, the content of total saccharides decreased significantly compared to the control plants (Corcobado et al., 2022).Even though the sugar alcohols were even higher in the roots of control plants, the differences between infected and control seedlings became less pronounced after three weeks of inoculation (Corcobado et al., 2022).There is a different response to Phytophthora attack in leaves and roots at the biochemical level (Corcobado et al., 2022).
While the influence of temperature on the activity of different species from the "P.alni complex" was discussed within previous studies (Schumacher et al., 2006;Černý and Strnadová, 2010;Redondo et al., 2015) suggesting possible further spreading due to climatic changes, studies on Phytophthora-Alnus interaction at elevated CO 2 levels are, however, still missing.This study aimed to compare the responses of A. glutinosa plants to infections by PAA, PAU, and PLU at ambient and elevated CO 2 levels.
Prior to the experiment, the following hypotheses were raised: (i) the survival of alders is influenced by the interaction between Phytophthora species inoculation and CO 2 level, and (ii) variations in physiological and biochemical responses are expected in inoculated alders depending on the Phytophthora species and the CO 2 level.

Plant material and growth conditions
Alder (A. glutinosa) tree cones were collected in autumn 2020 in the Czech Republic.To achieve higher variability of alder genotypes, two different riparian populations were selected: one showing typical alder dieback and collar rot symptoms (Jung and Blaschke, 2004;Jung et al., 2018), and the second being asymptomatic (Table 1).The seedlings were grown in a ready-mixed substrate containing peat and perlite (Forestina, Czech Republic) in individual 2-liter containers.The plants were watered to field capacity with tap water once per week.Plants were fertilized 2 months before the inoculation (Osmocote, ICL Specialty Fertilizers).No fungicides were applied during the experiment.

Underbark inoculation test
Isolates of three Phytophthora species, PAA, PAU, and PLU were sourced from the CZU/MM Oomycetes collection (Table 2).Their identity was previously confirmed by sequence analysis of the internal transcribed spacer (ITS1-5.8S-ITS2)region of the ribosomal DNA according to Jung et al. (2019), for the PAU isolate the β-tubulin gene was also sequenced using primers TUBUF2 and TUBUR1 (Kroon et al., 2004) (Table 2).Underbark inoculation was performed in June 2021 according to Milenković et al. (2018).The bark surface of the alder saplings at a distance of 10-15 cm from the collar was cleaned with 70% ethanol.Plants were wounded using a sterilized 7-mm metal cork borer.Same-sized plugs cut from the edges of 5 to 7-day-old Phytophthora colonies grown on V8-agar medium (V8A) (Jung et al., 1996) were placed with the mycelial side onto the exposed wood.The agar plugs were covered with the removed piece of bark and cotton moistened with sterile distilled water and sealed with Parafilm and aluminum foil.
In total, 12 six-months-old plants per Phytophthora species and CO 2 level were inoculated, including six plants originating from populations 1 and 2, respectively (Table 1).The control treatment also had 12 plants, but they were mock-inoculated with sterile V8A plugs and sealed in the same way.After the inoculation the plants were moved back to their growth chambers.The plants were inspected every 2 weeks for the appearance of symptoms, such as seedling mortality, vertical length of stem lesions and extent of girdling (Table 3).Girdling was visually evaluated on a five-point scoring system, according to Zamora-Ballesteros et al. (2017).
Thirteen weeks after inoculation, when 50% of plants in one treatment died, the experiment was finished and evaluated.Necrosis lengths were measured using a precise ruler, while necrosis widths were measured using a flexible measurement tape.Reisolations were made from all inoculated and control plants by plating small pieces from the upper and lower margins of necrotic lesions or, in the absence of necroses, from the margins of the inoculation points onto PARPNH selective agar medium (Jung et al., 1996).Pathogen identification was confirmed by comparing the colony morphologies on V8A with those of the original isolates.
Biomass of above-and below-ground plant tissues was weighed and expressed as the above/below-ground dry biomass ratio.The aerial tissues of each seedling were divided into stems and leaves and dried in the oven at 65 • C for 48 hr.Roots were dried after thoroughly washing to remove adhering soil.

Physiological and chemical analysis Physiological analysis
Six plants from each treatment (Phytophthora species/CO 2 level) were randomly selected for the physiological measurements.These were performed six times during the experiment in 2-week intervals, always using the same plants: one measurement prior to the inoculation and five measurements after the inoculation.Physiological measurements included chlorophyll a fluorescence and gas exchange measurements.
Chlorophyll a fluorescence was measured using FluorPen FP 110 (Photon Systems Instruments, Brno, Czech Republic).
. / gc. .Measurements were performed on adaxial surfaces of 3-4 leaves per plant.Before the fluorescence measurement, dark acclimation of the three selected leaves was assured by special leaf clips with a retractable metal plate.After 20 min of dark adaptation, the instrument was mounted to the clip and the actinic light (3,000 µmol m −2 s −1 ) was applied to the leaves.For statistical analysis, the following indexes were selected to estimate plant health and performance (according to Živčák et al., 2008): (a) maximum quantum yield of primary PSII photochemistry (1) and (b) performance index expressing energy conservation from absorption of light by an antenna (2).Its expressions are detailed below: Φ PSII -maximum quantum yield of primary PSII photochemistry F V -variable fluorescence F M -maximum fluorescence F 0 -fluorescence intensity at 50 µs F j -fluorescence intensity at the J step (at 2 ms) F M -maximal fluorescence intensity V j -relative variable fluorescence at 2 ms M 0 -initial slope of fluorescence kinetics.
An open portable photosynthesis system with infrared gas analyser LI-6400 XT (LICOR, Lincoln, NE, USA) was used for in situ gas exchange measurements.Net photosynthetic rate (PN), transpiration (E), stomatal conductance (Gs), and internalto-ambient CO 2 concentration ratio (Ci/Ca) were measured at photosynthetic photon flux density of 1,500 ± 1 µmol m −2 s −1 and an ambient CO 2 concentration of 415/830 ± 1 µmol mol −1 (ambient/elevated CO 2 treatments) which was above the saturation point measured in advance (350-400 µmol m −2 s −1 ).A standard leaf chamber with a red/blue LED light source was used.The samples were taken between 9 am and 5 pm.Measurements were performed on 2 leaves per plant.

Chemical analysis
Three root samples per treatment (Phytophthora species/CO 2 level) were taken for chemical analysis during the final assessment.Selected plants were gently removed from the planting pots, and the root systems washed in distilled water to remove all substrate particles.Immediately after cleaning the samples were frozen and kept at −32 • C until further processing.After freeze-drying and homogenization, approximately 0.03 g per sample were used for the extraction of selected sugars and phenolic compounds.The detailed method is given in Supplementary Table 1.

Data analysis Lesion development and mortality, plant biomass
The assessment of normality within each group was conducted using the Shapiro-Wilk test.Given the observed deviation from the assumptions of normal data distribution, differences in various parameters, including lesion length and width, lesion area, above-/below-ground biomass ratio, percentage of lesion length from plant height, final plant height, and collar diameter, were subsequently analyzed using the paired samples Wilcoxon test.Significances of pairwise differences were computed by the pairwise Wilcoxon test implemented in "pairwise.wilcox.test"base R statistical software (R development CORE team, 2021) with pvalue adjustment set as BH-i.e., false discovery rate adjustment (Benjamini and Hochberg, 1995).Differences between groups were investigated at the significance level α = 0.05.Hedge's g effect sizes were computed using "cohens.d"function from "effsize" package with "hedges.correction"parameter set to TRUE.
For girdling analysis, the Area Under Disease Progress Curve (AUDPC) was used for the disease assessment.The AUDPC was measured using the trapezoidal integration method, according to Madden et al. (2007), using the expression: where: y i : initial score of girdling; y i+1 : final score of girdling; t i+1 : final time; t i : initial time.
AUDPC values were analyzed using ANOVA and Fisher's LSD test performed in STATISTICA 13.0 software.
Survival curves describing the relationship between the mortality rate and time after the infection were generated using the Kaplan-Meier Estimates method in STATISTICA 13.0 software.Statistical significance of survival differences among groups was assessed employing the Chi-Square Test.Pairwise comparisons were conducted using the Log Rank test with a significance threshold set at α = 0.05.
Mortality rate was evaluated using the Chi-Square Test.Correlations between mortality and selected parameters (lesion area, percentage of lesion length from plant height, final plant height, and above-/below-ground biomass ratio) were analyzed using Spearman Rank Order Correlations because of the inherent characteristics of the data.Relationships between stem girth and girdling/mortality were analyzed using either linear regression models or Welsh test, employing the R software.The data used for the analysis of the correlation between stem girth and lesion girdling are originated from an equally spaced five-point scoring system of lesion girdling, hence suitable for any kind of analysis including linear regression models.

Physiological and chemical analyses
For physiological parameters, the R software was used.Since the measurements were conducted six times, a repeated measurements approach was adopted with each plant serving as the measurement unit.The experimental design encompassed treatment groups with four levels (control, PAA, PAU, and PLU), CO 2 concentration at two levels (415/830 ppm).The evaluation of physiological traits involved the application of a mixed linear model, accounting for the repeated measurements.
PAA-and PAU-inoculated plants showed gradual mortality, leading to a decrease in the number of measured plants; therefore, the measurements performed at 6, 8, and 10 weeks post inoculation (p.i.) were unsuitable for statistical analyses of gas exchange measurements.
The model was structured to assess the influence of timevarying factors and their interactions on the response variable y.The fixed effects in our model encompassed Time, Treatment, and CO 2 , as well as their interactions with Time, formulated as: To adequately capture the within-subject variation typical in repeated measures designs, the model controlled variation among replicated boxes by including random effect formulated as Time:Box.Furthermore, to address the autocorrelation typically present in repeated measures data, the residual structure of the model was specified as: id(ID):cor(Time), allowing for correlations between time of the same experimental unit.
All mixed models were fitted with ASReml-R v4.1 (Butler et al., 2017), which uses restricted maximum likelihood methods to estimate variance components.In all cases, diagnostic plots were reviewed for normality and to detect potential outliers.
In the statistical analysis of chemical parameters (sugars and phenolic compounds), the Kruskal-Wallis ANOVA was used to examine intergroup differences.Prior to analysis, the normality of the data distribution was assessed, revealing a deviation from the assumption of normality.To further test significant variations among groups, a post hoc Dunn's test was conducted (significance level α = 0.05).

Lesion development and survival analysis
Representative symptoms 13 weeks post inoculation (p.i.) are shown in Figure 1 and explained below.Lesions started to appear within the first 2 weeks p.i..The lesion size varied between the Phytophthora species and among the plants within the same treatment.Lesions developed rapidly in plants inoculated with PAU and PAA.In most plants the lesions reached their final length within 4 weeks p.i. with mean lesion lengths ranging from 7.8 to 8.7 cm in PAA, 7.7 to 11.4 cm in PAU and 1.8 to 1.9 cm in PLU (Table 4, Figure 2).For PAU a statistically non-significant trend of higher average lesion lengths was recorded in the ambient as compared to the elevated CO 2 level (11.43 ± 7.33 cm vs. 7.66 ± 6.87 cm, p = 0.249; Table 4, Supplementary Figure S1).For PAA a similar trend was observed (8.68 ± 7.11 cm vs. 7.76 ± 5.91 cm, p = 0.951; Table 4, Supplementary Figure S1).On the contrary, plants inoculated with PLU had only very small lesions that were not significantly different from the control (Table 4,  Twelve plants were inoculated in each treatment (CO2/Phytophthora species). 1Different letters indicate significant differences between values within each CO2 level.For significance of differences between CO2 levels see Figure 2 and Supplementary Figures S1, S2. Figure 2, Supplementary Figure S1).The re-isolations confirmed the pathogen presence in all inoculated plants (Table 4).
The lesion area, calculated as the ellipse area (A = πab), showed differences between the Phytophthora species and CO 2 levels which were partly statistically significant (Table 4, Figure 2).At both CO 2 levels, PAA and PAU caused lesions with significantly different lesion areas compared to both the control and PLU-inoculated plants (p ≤ 0.05; Table 4).In contrast to PAA-inoculated plants, in PAU-inoculated plants the effect of CO 2 was statistically significant (p = 0.049) with bigger lesions found at the ambient as compared to the elevated level (18.23 ± 12.19 cm vs. 8.13 ± 7.92 cm) (Figure 2).PLU differed significantly from the control at both CO 2 levels, but .
the variability was much lower and no significant difference existed between different CO 2 levels (Table 4).PAA and PAU had significantly different lesion width compared to both the control and PLU, whereas PLU did not differ from the control at both CO 2 levels (ambient: p = 0.276, elevated: p = 0.572) (Table 4).However, the effect of CO 2 was significant only in PAU (p = 0.007) with bigger lesion width at ambient compared to the elevated level (1.94 ± 0.52 cm vs. 1.22 ± 0.35 cm, respectively) (Supplementary Figure S2).
For both PAA and PAU-inoculated plants a statistically significant positive correlation was found between lesion area and mortality.A larger lesion area was consistently associated with a higher mortality rate (PAA r = 0.72; PAU r = 0.78) (data not shown).
The statistical analyses revealed significant differences in girdling caused by PAA and PAU compared to the control regardless of the CO 2 level (Figure 3).The first dying plants developed lesions very quickly and complete girdling of the collar occurred within the first weeks p.i.On the contrary, PLU girdled in all plants <25% of collar circumference.Complete girdling was significantly correlated with smaller stem circumference (stem girth) in both PAA (R 2 = 0.272, p = 0.0052) and PAU-inoculated plants (R 2 = 0.368, p = 0.001; Figure 4).Complete girdling also induced the swelling of sleeping buds above the girdling lesions a strong reaction in several plants (Figures 1I, J).

Seedling mortality
The first symptoms of severe dieback and mortality appeared in PAU-inoculated plants 2 weeks p.i., followed by PAA-inoculated plants 6 weeks p.i.All control and PLU-inoculated plants survived until the end of the experiment (Figure 1, Table 4).Overall, PAA caused mortality in 33.3% (8/24 plants) and PAU in 45.8% (11/24) of the inoculated plants (Table 4).However, there was no significant effect of the CO 2 level on the mortality rates of either PAA or PAU (p > 0.05).
Log-rank test of equality from survival analysis revealed significant differences in survival probability among experimental groups (χ 2 = 29.38,p < 0.001).At ambient CO 2 level, survival probabilities in PAA and PAU-inoculated plants were 70% and 45%, respectively.At elevated CO 2 level, both isolates had survival rates lower than 60%.First mortality associated with PAU was recorded 3 and 5 weeks after inoculation at elevated and ambient CO 2 levels, respectively.The seedlings inoculated with PAA started to die 8 and 9 p.i. at elevated and ambient CO 2 levels, respectively.There were no differences in seedling survival probability between PAA and PAU isolates at both ambient and elevated CO 2 level.Compared to the control groups, PAU showed significant differences at ambient (χ 2 = 8.67, p = 0.003) and elevated CO 2 level (χ 2 = 6.07, p = 0.014).Also, PAA showed a significant difference from the control groups at ambient (χ 2 = 4.02, p = 0.045) and elevated (χ 2 = 6.07, p = 0.014) CO 2 level (Figure 5).
At the beginning of the experiment, plant height and collar diameter were variable with a mean height of 47.6 ± 15.6 cm and a mean collar diameter of 0.8 ± 0.2 cm (Table 5).Wilcoxon test revealed no significant differences in plant height or collar diameter between Phytophthora species at any CO 2 level (p > 0.05).The majority of plants reached their final height by the time of inoculation, with minimal height increases observed until the end of the experiment.In contrast, surviving plants exhibited a noteworthy increase in collar diameter (16-70% after 13 weeks).Conversely, plants that died first showed a decrease in collar diameter as the tissue shrunk due to water loss.
A statistically significant negative correlation between mortality and stem girth was found for both PAA and PAU (Welsh t-test, p < 0.05), i.e., higher mortality was associated with plants of smaller stem girth (Figure 6).
The percentage of lesion length from total plant height ranged in dead plants approximately from 30% to 80%.In surviving plants, the lesions extended approximately over 1% to 31% of the total plant height.This percentage correlated significantly with the mortality in both PAA and PAU (r = 0.804 and r = 0.851, respectively) (data not shown).Moreover, plant height showed a significant negative correlation with mortality for PAA (r = −0.389),PAU (r = −0.357)and PAA-PAU combined (r = −0.27)(Figure 7), i.e., higher plants died later or survived until the end of the experiment.The first plants to die were those with the shortest height and lesion lengths exceeding 60 % of the total plant height.

Plant biomass
The above-ground biomass at the end of the experiment was affected by various plant reactions to the pathogen infection.In many plants, successive leaf loss occurred over the whole experiment.Symptoms, such as wilting, leaf drying, chlorosis, sudden leaf dropping, and subsequent regeneration leaf flush, caused variations in plant biomass across all treatments (Figure 1).No significant differences were observed in dry biomass between above-ground plant tissues and roots.The average values for above and below-ground biomass for all plants were 8.89 ± 6.50 g and 6.96 ± 5.06 g, respectively (Table 5).
The effect of Phytophthora species on dry biomass did not show statistical significances.However, when comparing the ratio between above-and below-ground biomass, a statistically significant negative effect of elevated CO 2 was observed in control (p = 0.0013) and PLU-inoculated plants (p = 0.0013) (Supplementary Figure S3).In both cases, higher ratios were found at the ambient level compared to the elevated level (control: 1.60 ± 0.24 vs. 1.14 ± 0.24, PLU: 1.64 ± 0.28 vs. 1.17 ± 0.24).In plants inoculated with PAA or PAU the ratio between above-and belowground biomass showed neither a significant effect of the CO 2 level (p > 0.05; Supplementary Figure S3) nor a statistically significant correlation with mortality (p > 0.05; data not shown).

Physiological measurements and chemical analysis
Physiological responses of alder seedlings to the three Phytophthora species showed significant variations in time vs. CO 2 levels, and their interplay across treatments for net photosynthesis (Pn), stomatal conductance (Gs), and intracellular CO 2 concentration (Ci).The Ci/Ca ratio, denoting the intra-and extracellular CO 2 concentration ratio, displayed  , p = .
differences between measurements and CO 2 concentrations.Net photosynthesis (Pn) exhibited a declining trend throughout the measurement period, with no significant differences between Phytophthora species and the control, excluding CO 2 concentration and week interaction.However, significant differences were found between the first and second measurements after the inoculation  (two and 4 weeks p.i., respectively) in all inoculated plants at each CO 2 level (p < 0.05) (Figure 8).A non-significant decrease in Pn, observed between the first and second measurements in control plants at ambient CO 2 level, was induced by competition for light and space.Elevated CO 2 notably increased Pn, particularly in the pre-inoculation stage, but the values in inoculated plants were gradually decreasing over the experimental period.
Stomatal conductance (Gs) remained stable throughout the experiment, with the control having the lowest average values (0.27 ± 0.14 mol H 2 O m −2 s −1 ) and PLU exhibiting the highest values (0.32 ± 0.18 mol H 2 O m −2 s −1 ).Elevated CO 2 led to a 7% Gs reduction throughout the experimental period.Intriguingly, Ci responded to higher CO 2 levels, gradually reaching its peak during the third measurement (4 weeks p.i.), while Ci/Ca ratio increased in elevated CO 2 .Transpiration (Tr) showed decreasing trend in all inoculated plants till 4 weeks p.i. with no significant differences among Phytophthora species within each CO 2 level (p > 0.05) (Figure 8).Water use efficiency (WUE) ranged from 5.7 to 9.4 µmol CO 2 per mmol H 2 O, with higher values at the elevated CO 2 level.Fast kinetics of fluorescence indexes, quantum efficiency of PSII (Fv/Fm) and performance index (PI) exhibited common values for healthy plants.Similarly, the course of the OJIP curve did not reveal any significant differences in any step of the graph (Supplementary Figure S4).; PAU: p < . ).
In root samples, five sugars and eight phenolic compounds were identified, with no significant differences between treatments (Phytophthora species/CO 2 level) (Supplementary Table 2).Total sugar content was lower in infected alders, but no statistical significance was found between treatments.A higher content of mannitol was recorded in both PAA-and PAU-inoculated plants which were considered dead or dying compared to living plants.In PLU-inoculated plants approximately ten times lower values of mannitol were measured at elevated as compared to ambient CO 2 level.In contrast, the content of total sugars (mannitol excluded) was higher at ambient CO 2 level, and glucose and fructose showed approximately double concentrations compared to the elevated CO 2 level.Phenole concentrations in roots of infected and noninfected plants displayed extensive variation without significant differences (Supplementary Table 2).However, for several phenolic compounds an effect of CO 2 level was recorded in both control and PLU-inoculated plants.Higher mean values of catechin, epicatechin, gallic acid, and taxifolin were found at elevated CO 2 level compared to the ambient level.

Discussion
The devastating alder dieback epidemic, caused by the three species from the "Phytophthora alni complex, " was first noticed in the mid 1990s in the UK (Brasier et al., 1999;Gibbs et al., 1999) and is currently widespread along rivers and in alder plantings of most European countries (Jung et al., 2016(Jung et al., , 2018)).Phytophthora ×alni (PAA) has been identified as the main causal agent of the disease in most affected regions with temperate climatic conditions but P. ×multiformis (PAM) and P. uniformis (PAU) also cause bark lesions and mortality in the field (Brasier et al., 1999;Nagy et al., 2003;Jung and Blaschke, 2004;Thoirain et al., 2007;Solla et al., 2010;Aguayo et al., 2013;Jung et al., 2013Jung et al., , 2018;;Štěpánková et al., 2013;Redondo et al., 2015;Corcobado et al., 2023).Only two of the three species, i.e., PAA and PAU, have been recorded in the Czech Republic yet, with PAA strongly prevailing in declining alder stands (88% of 59 declining alder stands studied; Štěpánková et al., 2013).In one riparian forest in the Czech Republic, PAA and PAU were isolated from the same declining A. glutinosa tree (Macháčová, unpublished).A range of other Phytophthora species can contribute to the decline of riparian alder stands, most important P. plurivora (PLU) a widespread forest pathogen in Europe with a broad host range that includes A. glutinosa, A. incana and A. viridis (Jung and Blaschke, 2004;Jung and Burgess, 2009;Jung et al., 2013Jung et al., , 2016Jung et al., , 2018;;Mrázková et al., 2013;Haque et al., 2014;Corcobado et al., 2023;Tkaczyk et al., 2023).
The pathogenicity of species from the "Phytophthora alni complex" to all European alder species was demonstrated in several studies with aggressiveness of the different Phytophthora species varying between different studies (Brasier and Kirk, 2001;Santini et al., 2003;Jung and Blaschke, 2006;Černý and Strnadová, 2010;  Frontiers in Forests and Global Change frontiersin.orgHaque and Diez, 2012;Haque et al., 2015;Bjelke et al., 2016;Chandelier et al., 2016;Romportl et al., 2016;Zamora-Ballesteros et al., 2017).Extensive bark inoculation trials with mature logs of A. glutinosa revealed high aggressiveness of multiple PAA strains and of Dutch PAM strains whereas German PAM strains and Swedish PAU strains were only moderately pathogenic (Brasier and Kirk, 2001).Pathogenicity tests from other studies indicate, however, that PAU can be quite aggressive to A. glutinosa (Corcobado et al., 2017;Zamora-Ballesteros et al., 2017) and also to the North American red alder (A.rubra; Navarro et al., 2015).In this study, both PAA and PAU caused large bark lesions and high mortality in A. glutinosa seedlings, whereas PLU produced only small lesions not leading to mortality.The lesions caused by PAA and PAU did not differ significantly from each other.However, PAU lesions led to a higher mortality rate (45.8%) compared to PAA (33.3%), although this difference was not statistically significant.In a similar underbark inoculation study by Zamora-Ballesteros et al. ( 2017), PAA and PAU showed similar aggressiveness to A. glutinosa with lesion lengths of ca.35-40 mm.In their study, the survival probability at 37 weeks p.i. was 50-70% for PAA (depending on the isolate used) and 50% for PAU.Similar mortality rates were found in our study at 13 weeks p.i., but in contrast, the lesions were considerably longer reaching on average 110 mm in PAA and 117.8 mm in PAU.
Lesions extended rapidly in some plants during the first 2 weeks p.i. and sometimes reached their maximum length within this time.
A similar scenario was observed in other underbark inoculation studies (Zamora-Ballesteros et al., 2017;Marques Gomes et al., 2019).Gibbs (2003) reported on 15-months old A. glutinosa seedlings inoculated with species from the "Phytophthora alni complex" 3 weeks p.i. mean lesion lengths ranging from 28 to 68 mm.Comparison to results from other studies suggests a higher aggressiveness of the PAA and PAU isolates or higher susceptibility of the A. glutinosa genotypes used in our study.Isolate genotype seems to be an important factor affecting lesion development and mortality, as demonstrated by various experiments with PAA (Haque and Diez, 2012;Chandelier et al., 2016;Štochlová et al., 2016;Zamora-Ballesteros et al., 2017).The host plant susceptibility or resistance depends on many variables.Significant differences in susceptibility/tolerance of individual A. glutinosa trees to PAA were demonstrated in Belgium, Germany and the Czech Republic (Jung and Blaschke, 2006;Chandelier et al., 2016;Štochlová et al., 2016).In an extensive study including 90 genotypes of A. glutinosa sampled across the Czech Republic Štochlová et al.
(2016) found that differences in susceptibility to PAA were also depending on the geographic origin and altitude of the alder genotypes.Geographic patterns of resistance could be related to past exposures to invasive Phytophthora pathogens (Frampton et al., 2013), as trees in river corridors, including alders, are particularly exposed to many alien pathogenic oomycetes ( Černý et al., 2011;Milenković et al., 2018;Jung et al., 2019;Corcobado et al., 2023).
The age of a host plant is another important factor for the susceptibility to a pathogen.The thinner bark of young A. glutinosa shoots with smaller diameter may be more vulnerable to tissue colonization than mature bark tissues of older thicker shoots (Haque et al., 2015).This is consistent with the results of our study where we observed the first dying plants being those with the smallest collar diameter (ca.4-6 mm).On the contrary, plants with thicker collars (>9 mm) survived until the end of the experiment.
Phytophthora plurivora produced lesion lengths of ca.10-30 mm but induced no mortality.All inoculated seedlings healed the wounds with the callus tissue and did not show wilting or dieback symptoms.However, as a widely distributed pathogen, PLU was found quite aggressive to different host plants in numerous pathogenicity tests (Jung et al., 2003;Rytkönen et al., 2012;Henricot et al., 2014;Milenković et al., 2018;Oszako et al., 2018;Milanović et al., 2020;Ďurkovič et al., 2021).Several studies proved PLU to be pathogenic to alder seedlings, which is inconsistent with our results.Zamora-Ballesteros et al. (2017) reported in an underbark inoculation test on 1-year old A. glutinosa seedlings lesion lengths of 35-40 mm and a seedling survival probability of approximately 50% at 37 weeks p.i. (90% after 13 weeks).Similarly, Mrázková et al. (2013) observed 35-40 mm lesion length on 2years old alder seedlings after 6 weeks p.i.In another underbark trial, Haque et al. (2014) found all inoculated plants wilted and died 3 months p.i..In a zoospore inoculation test with excised A. glutinosa shoots, Jung and Nechwatal (2008) found PLU (under its former name P. citricola) causing within 3 weeks significantly longer lesions than PAA (123 vs. 72 mm).These contrasting results indicate that different genotypes of PLU show considerable differences in aggressiveness to A. glutinosa.In our study, despite having been constrained early by the formation of callus tissue the pathogen was successfully reisolated from all inoculated plants.Apparently, this PLU genotype has low virulence on A. glutinosa but has the ability to survive in dead tissue for quite a long time (13 weeks) which is probably related to the formation of oospores (Jung and Burgess, 2009).However, the pathogen would probably lose its viability and die after a certain time.This was demonstrated by Ďurkovič et al. (2021), who inoculated stems of 10-years old poplar trees under field conditions with P. cactorum (Lebert and Cohn) Schröt.and P. plurivora and found the wounds after 4 years completely healed with no presence of the pathogens.
Studies on Phytophthora-host interaction at elevated CO 2 levels are limited and showed contrasting results.Oszako et al. (2016) did not find any effect of elevated CO 2 concentration (800 ppm) on the root and shoot development of oak seedlings growing in non-infested soil or in soil infested with the oak-specific fine root pathogen P. quercina.In this study, a statistically significant negative correlation between the elevated CO 2 level and the above-/below-ground biomass ratio of A. glutinosa seedlings was found for the control and for plants inoculated with the moderately aggressive PLU strain but not for plants inoculated with the aggressive PAA and PAU strains.Although the mortality rate did not differ significantly between ambient and elevated (830 ppm) CO 2 levels, there was an effect of the CO 2 levels on the extent of the lesions for PAU-inoculated plants.Thus, at the elevated CO 2 level PAU caused a significantly smaller lesion area and width as well as a tendentially lower lesion length.Similarly, in a tomato trial, a tendency of increased tolerance to P. nicotianae Breda de Haan (referred to as P. parasitica) was found under elevated CO 2 (Jwa and Walling, 2001).However, in other studies, Phytophthora pathogens generally displayed higher aggressiveness to host plants at elevated CO 2 .Tkaczyk et al. (2014)  As the pathogen gradually colonizes the conductive tissues, the girdling extends, and the plant's response can be observed in both lesion development and physiological parameters.Many studies have shown that the common reactions after Phytophthora-induced bark infection are a gradual reduction in water uptake, the fast closure of stomata and in consequence a decrease in photosynthesis (Osswald et al., 2014).In our study, the lesions length reached their maximum in most of the inoculated plants 4 weeks p.i.Therefore, the most profound reactions were measured within this period −2 and 4 weeks post inoculation (p.i.).The biggest decrease of transpiration (Tr) was recorded in PAU 4 weeks p.i. when the amount of transpired water dropped to the half value (49%) of the control values (1.2 mmol m −2 s −1 ) at elevated CO 2 level, and to 68% (1.0 mmol m −2 s −1 ) at ambient level.Similarly, net photosynthesis (Pn) reached significantly lower values in PAA-and PAU-inoculated plants 4 weeks p.i. compared to 2 weeks p.i. at regardless the CO 2 levels.Generally, it would have been expected that Pn increases under elevated CO 2 conditions as it occurred for control plants and also in many studies (e.g., Pritchard et al., 1999;Badiani et al., 2023).It seems that the influence of the CO 2 conditions on Pn disappears under acute infections caused by aggressive pathogens.Surprisingly, no significant differences in Pn were recorded between control and inoculated plants at both CO 2 levels.At ambient CO 2 level, these results are consistent with Clemenz et al. (2008), who found no significant differences between control and inoculated plants in a trial with 3-years-old alders steminoculated with Phytophthora ×alni during the 1 st month after the inoculation.Only after 3 months were differences recorded, indicating the late effect of the pathogen on plant physiology.
The decrease of stomatal conductance (Gs) as a common physiological trait was observed across various Phytophthora and tree species (Gallego et al., 1999;Robin et al., 2001).In contrast, Gs remained relatively stable in our experiment, but PAA-and PAUinoculated plants decreased their Gs by approx.30% 4 weeks p.i. in comparison to 2 weeks p.i..This trend was not observed in PLU-inoculated plants, as these plants showed very small lesion sizes and no mortality.Moreover, no effect of elevated CO 2 on Gs was observed in any treatment.Our comparison of Pn and Gs trend 2 and 4 weeks p.i. revealed that Pn decreased with Gs simultaneously at ambient CO 2 level.A similar observation was made by Fleischmann et al. (2005) who performed soil inoculation of beech seedlings with P. plurivora (referred to as P. citricola).As reported in many studies, plants close their stomata at elevated CO 2 level due to surplus of CO 2 in the atmosphere (e.g., Kupper et al., 2006).Nevertheless, in our study, elevated CO 2 caused disbalance between Pn and Gs in PLU and PAA with lower Pn value and the same value of Gs.
Elevated CO 2 also resulted in higher WUEinst values in all treatments 2 weeks p.i. as expected according to general trends (Running and Nemani, 1991).However, the positive influence of CO 2 diminished over time and the later values were comparable between treatments.Similarly, Clemenz et al. (2008) reported that WUE of leaf gas exchange hardly differed between inoculation treatments and controls.On the contrary, another study with 1-year-old beech plants infected with P. plurivora showed that WUE data indicated that infected plants suffered from severe drought, even though a significant decrease in net assimilation and transpiration was not observed (Fleischmann et al., 2002).
Chlorophyll a fluorescence is a useful tool for evaluating the plant health status (Oxborough, 2004).Pfanz et al. (2015) 2022) might be influenced by the different type of applied inoculation methods.In soil infestation experiments with Phytophthora pathogens, root damage can be rapidly reflected in decreased foliage vitality, including decreasing performance of chlorophyll a.On the contrary, underbark inoculation leads primarily to necrosis of cortex and phloem tissue, resulting in limited assimilate transport (Osswald et al., 2014) that is minimally shown in chlorophyll fluorescence.Therefore, only when the pathogen starts to move deep into phloem and xylem tissue and the stem becomes completely girdled, chlorophyll a fluorescence would be possibly affected.
Plant responses to pathogen infection can be observed in changing levels of secondary metabolites.Levels of sugars may decrease notably, as reported by Corcobado et al. (2022) who found sugar pool decrease in beech roots in a soil infestation trial with different Phytophthora species, suggesting reduced phloem transport.Levels of mannitol may rise dramatically during plant infection by biotrophic or necrotrophic fungi, as mannitol has a multitude of functions in fungal metabolism (Calmes et al., 2013).In our study, at ambient CO 2 level, total sugar content was lower in infected alders, but statistical significance was only found for mannitol, which showed increased levels in PAA and PAUinoculated plants compared to the control.Similarly, Camisón et al. (2019) reported a temporary increase of sugar alcohols in chestnut clones resistant to P. cinnamomi during disease progression.In our study, plants with an advanced stage of dieback exhibited particularly high values of mannitol, which can be explained by the role of mannitol as an antioxidant agent in fungi and oomycetes that can suppress host defense responses by quenching the reactive oxygen species (Meena et al., 2015;Puig et al., 2018).Interestingly, in our study, an effect of elevated CO 2 level on the concentrations of several sugars and phenolic compounds in the roots was found, although the differences to ambient CO 2 were not statistically significant.The elevated CO 2 level showed ./ gc. .
an inhibiting effect on the total sugars content, with glucose and fructose reaching only half of the concentrations measured at the ambient CO 2 level.High variability in phenolic levels was recorded in all root samples tested in this study, regardless of the treatment.Generally, phenolic compounds level tends to rise in leaves in response to tissue damage, as demonstrated by other studies (e.g., Brown et al., 2018).In our study, PAA and PAUinoculated plants showed a high variability of phenolic compounds in roots.Regarding elevated CO 2 levels, a positive effect was recorded for several phenolic compounds in control and PLUinoculated plants.Similar results have been shown at elevated CO 2 by Roth et al. (1998).In general, the lack of differences in secondary metabolites between treatments was probably due to the low number of sampled plants which exhibited significantly different health status ranging from severe dieback to healthylooking plants.
In conclusion, this study demonstrated that PAU can show similar aggressiveness to A. glutinosa as PAA.The results indicate that inoculation experiments should be performed over longer periods to observe for different Phytophthora species statistically significant differences in disease incidences and defense responses of inoculated plants.Statistically significant effects of elevated CO 2 concentration on the disease symptoms of underbark-inoculated A. glutinosa plants, such as mortality, lesion development, plant biomass or chemical and physiological parameters were not recorded.On the contrary, PAU caused significantly larger lesions at ambient CO 2 and, although not significant, a trend of smaller lesion area at elevated CO 2 was also observed for PAA.Further experiments using soil infestation methods which simulate the natural infection process (Jung et al., 1996(Jung et al., , 1999(Jung et al., , 2003(Jung et al., , 2018) ) and longer exposition to various CO 2 levels to enable a significant effect on growth and above-ground biomass production of plants exposed to elevated CO 2 level (Ainsworth and Long, 2005) and a potentially more pronounced effect of Phytophthora infections on aboveground plant growth are needed to clarify the effect of elevated CO 2 on the Phytophthora root and collar rot disease of A. glutinosa.

FIGURE
FIGUREComparison of lesion area in the stems of Alnus glutinosa seedlings caused by three Phytophthora species at two CO levels (ambient ppm, elevated ppm).PAA, P. ×alni; PAU, P. uniformis; PLU, P. plurivora.The statistical significance of CO pairs comparisons is indicated by N.S. (non-significant) or an asterisk (p ≤ .).The tables on the right contain all pairwise comparisons; the upper triangle represents the p-value of pairwise comparisons, the lower triangle represents the e ect size of Hedges'g.

FIGURE
FIGUREComparison of area under the disease progress curve (AUDPC) in stems of Alnus glutinosa seedlings inoculated with Phytophthora ×alni (PAA), P. uniformis (PAU) and P. plurivora (PLU) at two CO levels ( ppm and ppm).Error bars show the standard deviation.Di erent letters indicate significant di erences (Fisher's LSD Test, p < .).

TABLE
TABLE Origin of the Alnus glutinosa seeds used in this study.Origin of the Phytophthora isolates used in this study and GenBank accessions of their ITS(a)and β-tubulin(b)sequences.
TABLE Pathogenicity of three Phytophthora species to Alnus glutinosa seedlings in the underbark inoculation test after weeks.
Fleischmann et al. (2010)P.plurivora causing more extensive fine root destructions of beech at elevated CO 2 level compared to the ambient condition.Similar results with increased susceptibility of beech seedlings to P. plurivora (referred to as P. citricola) at elevated CO 2 level were recorded byFleischmann et al. (2010). seedlings measured bark chlorophyll fluorescence in alders 10 weeks after basal stem inoculation with Phytophthora ×alni and revealed that Fv/Fm and DF/Fm ′ of the cortex chlorenchyma decreased to almost zero, indicating tissue necrosis.Similarly, in other studies at ambient CO 2 level, when Phytophthora soil inoculation was performed, differences were recorded in leaf chlorophyll fluorescence between control and inoculated plants (Angay et al., 2014; Dalio et al., 2017; Corcobado et al., 2022).Furthermore, Corcobado et al. (2022) observed differences in chlorophyl fluorescence of beech seedlings only 2 weeks p.i. but 3 weeks p.i. these differences disappeared.Similarly, at elevated CO 2 level, Fleischmann et al. (2010) found electron quantum yield of PSII decreased only shortly before the first wilting symptoms occurred in inoculated dying plants.In contrast, our results did not reveal any difference between treatments regardless of the CO 2 level.The discrepancy between our findings and those of Fleischmann et al. (2010) and Corcobado et al. (