Interactions between Phytophthora cactorum , Armillaria gallica and Betula pendula Roth. Seedlings Subjected to Defoliation

: The purpose of this study was to better understand the interactive impact of two soil-borne pathogens, Phytophthora cactorum and Armillaria gallica, on seedlings of silver birch ( Betula pendula Roth.) subjected to stress caused by mechanical defoliation, simulating primary insect feeding. This is the ﬁrst experimental conﬁrmation of silver birch seedling root damage (and in consequence shoot mortality) caused by the additive e ﬀ ect of defoliation stress and P. cactorum inoculation via soil. However, the most severe damage to roots occurred after A. gallica inoculation. One year after treatments, chlorophyll ﬂuorescence measurement, and gas chromatography coupled with mass spectrometry (GC-MS) were used to analyze the photosynthetic activity in leaves, the volatile organic compounds (VOCs) emitted by the birch leaves, and chemical compounds from the roots. The cumulative e ﬀ ect of the two pathogens and partial defoliation reduced photosynthetic activity, suggesting dysfunction of photosystem PSII due to the applied stresses. In summary, it seems that the main di ﬀ erences in photosynthetic performance could be attributed to Armillaria infection. The birch leaves in seedlings exposed to 50% defoliation, and inoculation with P. cactorum and A. gallica, emitted more aromatic carbonyls and alcohols, as well as half as much aliphatic esters, compared to controls. In infected birch roots, the production of phenols, triterpenes, and fatty alcohols increased, but fatty acids decreased. Higher levels of aromatic carbonyls and alcohols in leaves, as well as phenolic compounds in the roots of stressed birches to control) suggest an activation of plant systemic acquired resistance (SAR). held for 10 min. The detection was performed in a full scan mode from 41 to 800 amu. For identiﬁcation of extracted components, both mass spectra and retention indices were used. After integration, the content (%) of each component in the total ion current (TIC) was calculated. All measurements for leaves and root extracts were performed in 3 replicates, to determine the detection error.


Introduction
Silver birch (Betula pendula Roth.) is common throughout the lowlands and lower mountainous regions in Europe, and is subject to attack by Phytophthora cactorum (Lebert and Cohn) J. Schröt (as a primary pathogen) and Armillaria gallica Marxm. and Romagn. (as a secondary one) [1,2]. To date, over 150 Phytophthora species (Oomycetes) have been described from a broad range of hosts, including forest tree species and ornamental plants [3,4]. These groups of pathogens have not only led to extensive mortality of different forest tree species, but are also responsible for negative ecological impacts in many countries across the world. Moreover, climate warming and increased mean precipitation in the growing season have often been associated with increased invasiveness of phytophthoras in many plants [3,5]. In the last few decades, interest in the Phytophthora species as plant pathogens has been growing in Europe and on other continents, since these pathogens are causing emerging diseases of many forest tree species [6][7][8][9]. Phytophthora ramorum Werres, De Cock and Man in't Veld is an example of a harmful pathogen, which is responsible for sudden oak death (SOD) in North America and sudden larch death in Europe (SLD) [10][11][12][13]. Generally, the risk of transferring phytophthoras from nurseries to forest plantations is considered high, as P. ramorum was transmitted in seedlings from nurseries in California, and it affected production in over 20 US states [14].
In the 1980s birch stand decline was observed in Poland, but it was then considered as a complex disease involving water issues caused by drought. It is now known that species of Phytophthora causing damage to fine roots can mimic drought symptoms [20,21]. Some experiments demonstrated the possible susceptibility of one-year-old silver birch seedlings of Polish provenance to three species of Phytophthora, i.e., P. cinnamomi, P. citrophthora, and P. plurivora [22,23]. Nevertheless, the only published records of P. cactorum associated with birch in Poland are also of young birch seedling tissues [19], but not on adult trees [4].
Generally, tree damage and eventual mortality are caused, not by a single pathogen, but a whole suite of disturbance agents, like harmful fungi and defoliator insects, or unfavorable abiotic conditions. In Europe, among the root and butt rot fungi (Armillaria spp.), A. gallica is considered to be a weak pathogen of many forest tree species, increasing its activity after severe drought or defoliation [24][25][26]. However, Armillaria spp. are known to have a growing destructive impact in European ecosystems due to climate changing conditions, both in forest and in urban environments [27,28].
During the last few years, abundant research has been devoted to the combined effects of pests and pathogens on forest ecosystems [29]. Additionally, along with phytopathogens, birch trees are often subjected to attack by several insect defoliators, such as Deporaus betulae (L.), Phyllobius betulae F., and Ph. arborator (Herbst.), leading to severe damage of the crown [30,31]. A tree's response to direct damage (e.g., defoliation) consists not only of a consecutive production of new leaves from dormant buds, but also of physiological and biochemical reactions. Many plants, including forest tree species, change their photosynthetic activity and the level of volatile organic compounds (VOC) secretion in response to damage caused by external stress factors [32][33][34]. Plants weakened by partial defoliation very often become vulnerable to multi-pathogen infections [35].
The main purpose of this work was to better understand interactions between organisms which could be associated with birch decline syndrome. The interactions between soil borne pathogens on the production of secondary metabolites (such as priming) by birch have not been sufficiently studied, and an objective was to examine the effects of fungus (A. gallica) and oomycete (P. cactorum) on the chemical composition of birch volatiles, secreted by young birch seedlings. To our knowledge there are no scientific reports in Poland about interactions between P. cactorum and A. gallica, which cohabitate in local forests in the same ecological niche, the soil [36]. Furthermore, since defoliation may predispose trees to infection, we hypothesized that combinations of partial defoliation and infection by P. cactorum and A. gallica would cause more mortality than each stress alone. We attempted to mimic insect defoliation by the gradual cutting of leaves over time to generate a less abrupt physiological stress. We also analyzed photosynthetic activity, via chlorophyll fluorescence, and secondary metabolite content in leaves and roots to probe at the mechanism of physiological and biochemical plant response to separate, and combined, abiotic and biotic stresses. The main purpose of our investigation was to explore the effects of two soil borne pathogens (P. cactorum and A. gallica) on B. pendula seedlings subjected to stress caused by partial mechanical defoliation, simulating insect attack. We tried to determine the following: (i) whether P. cactorum and A. gallica can affect the health of birch seedlings; and (ii) whether stress from partial defoliation can stimulate enhanced root infection. We also attempted to figure out how the combined effects, caused by the aforementioned potential pathogens and partial defoliation, influenced the photosynthetic activity of the treated plants, and what chemical compounds were produced in the stressed leaves and roots of the birch seedlings.

Phytophthora Cactorum
Liquid medium, containing 250 cm 3 of vermiculite, 20 cm 3 of millet, and 175 cm 3 of V8 medium (100 mL of multi vegetable juice, 2 g of calcium carbonate CaCO 3 , and 900 mL of distilled water), was dispensed into 1 L Erlenmeyer flasks (890 cm 3 of medium per flask) and autoclaved for 15 min at 121 • C. An isolate of P. cactorum, from the IBL collection (GenBank accession number NCBI KX242303), was placed into the flasks using hyphae grown on V8 agar (ten pieces of agar about 0.5 × 0.5 cm covered with mycelium). The cultures had been incubated for 7 days at room temperature [37]. They were also used to inoculate soil samples (prepared from a mixture of peat, sand and perlite in a 1:1:1 ratio) and added to each individual 1 L pot in a volume equal to 2% of the volume of the soil. The untreated check did not contain P. cactorum inoculum.

Armillaria Gallica
To prepare the A. gallica inoculum, 1.5-2 cm thick stems of hazelnut (Corylus avellana L.) were first cut into 10 cm long sections, placed in a 630 cm 3 metal container, wrapped with aluminum, and autoclaved for 80 min at 121 • C. After cooling, hyphal plugs from 3 week old cultures of A. gallica (GenBank accession number NCBI DQ115578) on 2% malt agar were added to the sterilized stem containers, with the full contents of a 9 cm diameter petri plate per container of hazelnut cuttings. The inoculated stem segments were then incubated at 25 • C for 3 months until segments were completely overgrown by mycelium. After this, the stem segments were placed in the appropriate birch seedling pots, with 2 segments per pot. Autoclaved hazelnut segments were placed in pots as controls.

Verification of the Seedling Infection
The occurrence of inoculated pathogens in rhizosphere soil was checked as follows: (i) visually for Armillaria rhizomorphs attached to roots, following Oszako et al. [38], or (ii) by baiting with leaves, or (iii) culturing (P. cactorum) [39]. Attempts were made to isolate P. cactorum from baited leaves and from rhizosphere soil containing fine roots in order to test if the introduced pathogen was still present, and vital. For baiting, soil samples were taken from several pots of birch seedlings and were put into plastic boxes and watered (1 cm above soil level). On the surface of water, previously washed and dried 7-14 day-old oak and beech leaves were placed, and incubated at room temperature. They were observed for discoloration daily for a week. When brownish spots appeared on the leaf surfaces (generally after 3-7 days of incubation at 20 • C), leaves were cut into small pieces (approx. 5 × 5 mm) and transferred to a selective medium: PARPNH (agar containing V8) agar with various antibiotics, incubated at 20 • C in the dark, and the hyphae emerging after 48 h were transferred to V8, agar medium. Cultures were checked for axenicity over the following days. Initial identification, following morphological descriptions in Erwin and Ribeiro et al. [6], was confirmed molecularly (described below in the in-planta section).
To assess the presence of P. cactorum in planta, roots were collected a year after inoculation and washed thoroughly. Genomic DNA was extracted and purified using a NucleoSpin ® Plant II Midi kit (Macherey-Nagel, Düren, Germany). PCR was done using species-specific primers designed for P. cactorum isolate, JF300214.1 from GenBank, following Nowakowska et al. [40]. The TaqMan probe was labeled with the reporter dyes, i.e., JOE (6-carboxy-4,5-dichloro-2,7-dimethoxyfluorescein) at the 5'-end and HBQ1 quencher at the 3 -end (Sigma-Aldrich, St. Louis, MO, USA). PCR amplifications were performed in a total volume of 15 µL containing 1 × Master Mix (Sigma-Aldrich), 2 µM of each primer, 0.2 µM of probe, and 2 µL of diluted genomic DNA. PCR amplification included one cycle of denaturation at 95 • C for 3 min, 40 cycles of 95 • C for 30 s, 55-61 • C for 30 s, and 72 • C for 30 s. Amplifications were performed in a RotorGene 6000 (Qiagen, Hilden, Germany) apparatus, and fluorescence of newly synthetized amplicons was monitored in each PCR cycle during the annealing phase, following the manufacturer's instructions.

Defoliation of Birch Seedlings
To mimic progressive defoliation, such as that caused by foliar insects and possibly foliar pathogens, 32 seedlings were partially and progressively defoliated over four weeks. In the first week 50% of the leaves had 25% of their area removed. In the successive three weeks, 25% more each week was removed from these leaves, resulting in 50% defoliation after four weeks.

Evaluation of Birch Health Status
The last year of the experiment, in July 2019 the health status of all 64 birch seedlings was visually assessed on the following scale: 1, healthy with 30-50% of leaf and shoot loss; 2, diseased with 51-75% shoot loss; 3, dying with more than 76% of shoot loss [41]. We assumed that birch damage up to 50% can be scored as healthy, and those showing higher levels of damage (51-75%) are definitely diseased, while those damaged more than 75% are rated as dying, or are dead. Since the defoliation treatment occurred only in the first year, it was reasonable to assess defoliation in the succeeding years.

Biometric Parameters
After the 2-year-old seedlings had 13 months more growth, plant growth parameters were assessed. As soon as he plants were removed from their pots and cleaned, height and root collar diameter were measured, and fine roots were scanned on an EPSON Perfection V700 Photo Scanner (Regent Instruments Inc. Ltd., Quebec, QC, Canada), and the images processed using WinRhizo ® v. 2009 c 32-bit software (Regent Instruments Inc. Ltd., Quebec, QC, Canada). The number of living fine roots (<2 mm) per length of mother roots (2-5 mm) was calculated, and transferred to Excel for calculation of the following representative parameters: number of tips, fine root length, mother root length, and fine root length per major root length, total root length, fine root surface area, and fine root tips. After drying at 60 • C for 72 h (after which weight did not change), the dry biomass weight was measured in 8 replicates per treatment.

Fluorometric Analysis of Chlorophyll a
Fluorescence emitted by chlorophyll a is quantified by exposing a leaf (first kept in the dark) to light of a defined wavelength, and measuring the amount of light re-emitted at longer wavelength [42]. Twelve months after the start of the experiment, chlorophyll fluorescence of the leaves was measured directly with a Handy PEA (Handy Plant Efficiency Analyser) fluorimeter, following instructions from Hansatech Instruments Ltd. (King's Lynn, Norfolk, UK). In total, 10 measurements were made on each of three randomly chosen leaves for each of the eight treatments. Prior to measurements, leaves of 32 randomly chosen seedlings (half from each treatment) were kept for 30 min in the dark, where to the surface of each leaf, special clips were attached providing darkness, in order to slow down photosynthetic activity. The measurements were done on the central part of each leaf blade, on 4 mm 2 of each sample, i.e., in three replications per pot (from the top, middle, and lower part of the crown). The measurement conditions consisted of 1 s pulses of 3500 µmol m −2 s −1 light. To measure the physiological stress at the level of photosystem II (PSII), several photosynthetic parameters were assessed: F 0 , initial fluorescence; Fm, maximal intensity of fluorescence; F 0 /Fm, maximal photosynthetic activity; DI 0 /CS 0 , photosynthetic efficiency measured as energy dissipation in form of heat; and PI total, total performance index of photosystem II (PSII) [43].

Secondary Metabolites Contained in Birch Leaves and Roots
The leaf sample (1 g) was harvested from three birch seedlings. For each experimental treatment (1 to 8), one research sample was investigated. Similarly, eight samples of birch root were prepared. The volatile secondary metabolites emitted by birch leaves were analyzed using headspace solid phase microextraction and gas chromatography-mass spectrometry (HS-SPME/GC-MS) methods, following Isidorov et al. [44] and Oszako et al. [38]. The detection was performed in a full scan mode from 29 to 600 atomic mass units (amu).
Metabolites extracted from birch roots were analyzed with gas chromatography-mass spectrometry (GC-MS), a method previously developed by Stocki et al. [45,46], using GC-MS apparatus (Agilent Technologies Inc., Santa Clara, CA, USA). The samples were processed through an Agilent 7890A gas chromatograph with an Agilent 5975C mass spectrometer. Injection of each 1 µL sample was done using an Agilent 7693A autosampler. Chromatographic separation was performed on a capillary column HP-5MS (30 m × 0.25 mm × 0.25 µm) at a helium flow rate of 1 mL min −1 . The injector worked in a split (1:10) mode at a temperature of 300 • C. The initial column temperature was 50 • C, rising to 325 • C, at 3 • C min −1 , and the final temperature was held for 10 min. The detection was performed in a full scan mode from 41 to 800 amu. For identification of extracted components, both mass spectra and retention indices were used. After integration, the content (%) of each component in the total ion current (TIC) was calculated. All measurements for leaves and root extracts were performed in 3 replicates, to determine the detection error.

Statistical Analysis
For health data, growth data, and fluorescence data, normality of distribution was tested using the Shapiro-Wilk test, and homogeneity of variance with Levene's test. For data which did not satisfy the assumptions of ANOVA, and to enable parametric testing, data were subjected to Box-Cox transformation [47]. In order to assess which treatments differed significantly, two-way ANOVA (analysis of variance) and Tukey post-hoc tests were used. All analyses were performed in the STATISTICA 13.1 package (StatSoft Polska, Kraków, Poland) for α = 0.05.

Results
In 2018, the 64 three-year-old potted birch seedlings were subjected to eight different treatments involving partial defoliation, inoculation with two different pathogens, and combinations thereof. In 2019, after a year of growth in the greenhouse, the 64 four-year-old seedlings were subjected to the following analyses: (i) baiting-based, and molecular, detection of the P. cactorum and A. gallica in the soil and/or in plants; (ii) health state assessment of seedlings; (iii) evaluation of the response to stress factors, via biometric and photosynthetic measurements; and (iv) detection and measurement of volatiles emitted by leaves and compounds extracted from roots. The two-way ANOVA revealed no statistically significant interactions at p = 0.05.

Detection of P. cactorum and A. gallica
P. cactorum was detected from the rhizosphere soil of all inoculated plants via baiting techniques. The isolates obtained were identified as P. cactorum, following descriptions in Erwin and Ribeiro [6]. Real-time PCR reactions, performed with the specific P. cactorum probes and primers, yielded positive results for this pathogen presence in all roots taken from inoculated silver birch (Ct values ranged from 21.5 to 24.1), as was the case in isolation through baiting technique.
Successful infection was observed a year after inoculation with A. gallica. Typical rhizomorphs of the fungus, attached to the root system of birch seedlings, were observed in all infected potted plants (i.e., in treatments 5-8 of the experiment).

Growth Parameters
The comparison of seedling height between different experimental treatments showed no statistically significant differences (data not shown). However, the comparison of root collar thickness indicated some differences between treatments, especially with the additive effects of two pathogens and defoliation ( Figure 2). Figure 2 shows that neither P. cactorum nor A. gallica alone had effects on collar root diameter, and that the combined treatment had a greater effect. If those results are correct, the only plausible explanation is that there is a cumulative effect between both factors and defoliation.
Analysis of other growth parameters, such as the number of root tips, fine root length, fine root surface area, and fine root tips showed significant decreases in treatment (8) Armillaria gallica + Phytophthora cactorum, compared to the control ( Table 1). The most pronounced decrease of most of the parameters was observed for treatment 8 (A. gallica + P. cactorum), when seedlings were subjected to inoculation with both pathogens. In treatment 8, fine root length decreased by 70%, total root length by 69%, and fine root surface area by 58%. In total, five of eight root growth parameters significantly decreased in treatment 8, compared to the control.  Figure 2 shows that neither P. cactorum nor A. gallica alone had effects on collar root diameter, and that the combined treatment had a greater effect. If those results are correct, the only plausible explanation is that there is a cumulative effect between both factors and defoliation.
Analysis of other growth parameters, such as the number of root tips, fine root length, fine root surface area, and fine root tips showed significant decreases in treatment (8) Armillaria gallica + Phytophthora cactorum, compared to the control ( Table 1). The most pronounced decrease of most of the parameters was observed for treatment 8 (A. gallica + P. cactorum), when seedlings were subjected to inoculation with both pathogens. In treatment 8, fine root length decreased by 70%, total root length by 69%, and fine root surface area by 58%. In total, five of eight root growth parameters significantly decreased in treatment 8, compared to the control.   6 fine root length (<2 mm); 7 mother root length (2-5 mm); 8 total root length (mm); 9 fine root surface area; 10 fine root tips. * Results within a column with the same letter(s) are not significantly different at p = 0.05, using Tukey post-hoc test.

Chlorophyll Fluorescence
The physiological status of the birch seedlings was estimated based on selected chlorophyll fluorescence parameters. The initial fluorescence (F 0 ), in contrast to maximal fluorescence (Fm), was higher in all treatments compared to the control (Figure 3, Table S2). The highest ratios F 0 /Fm were observed in A. gallica alone and in combination with P. cactorum, and this indicated that the highest level of chlorophyll fluorescence was inversely proportional to photosynthetic activity. This was in concordance with the highest ratios, DI 0 /CS 0 , denoted in seedlings subjected to A. gallica infections, showing the highest amount of dissipated energy in the form of heat (i.e., the highest loss of absorbed light energy). Defoliation, 50%, alone did not heavily affect the photosynthetic activity (F 0 /Fm~1.0 and DI 0 /CS 0~1 .2).
1 Figure 3. Comparison of chlorophyll fluorescence parameters in different treatments of the experiment (colored lines, values in %). Explanation of the most important parameters (describing Clf and PI total) and their significance is given in the text and in Table S2.
In fact, all treatments showed reduced total photosynthetic activity (PI total <1.0) in comparison to the control plants (PI total = 1.0) except the treatment with P. cactorum alone, which did not adversely affect the photosynthetic activity, as PI total increased by 10% and F 0 /Fm~1.0 (Figure 3). Other combinations showed low photosynthetic efficiency, and among all treatments, defoliation combined with P. cactorum resulted in slightly higher values of PI total (0.9) compared to the other treatments ( Figure 3, Table S2). The interaction between the two pathogens and partial defoliation reduced PI total to 0.7, suggesting some compensation of the loss of activity in PSII due to the defoliation. In summary, it seems that the main differences in photosynthetic performance were given by the presence of Armillaria, confirmed by Tukey's post-hoc test. P. cactorum alone did not seem to have effects on the photosynthetic parameters different than that of the control (Table S2).

Secondary Metabolites of Leaves and Roots
Defoliation combined with inoculation with the two pathogens affected the chemical composition of the metabolites produced by leaves and roots of B. pendula, although no statistical differences were found except for sesquiterpenes (Tables 2, 3, S3 and S4). Nevertheless, qualitative and quantitative assessments of VOC emission are provided in supplementary materials (Tables S5 and S6). The amount of monoterpenes, e.g., citronellol, neral, (E)-geraniol, α-citral, and eugenol was the highest in leaves under the influence of stress caused by P. cactorum + defoliation (treatment 6) (Tables 2 and S5). Birch leaves infected by both pathogens (treatments 7, 8) had significantly less sesquiterpenes compared to the control, including α-copaene, β-bourbonene, or β-caryophyllene (Tables 2 and S5). The amount of aromatic esters (especially methyl salicylate) increased after infection by P. cactorum (3) and A. gallica + defoliation (5), compared to the control (Tables 2 and S5). The highest content of aromatic carbonyl compounds (7.12%) was found in the treatment with P. cactorum + defoliation (6), and the highest content of aromatic alcohols (12.47%) in the treatment with both A. gallica and P. cactorum (8) ( Table 2). Stressed birch leaves in treatments 7 and 8 emitted more aliphatic acids and aliphatic alcohols, but remarkably reduced the emission of alkanes and alkenes ( Table 2).

Birch Damage Caused by P. cactorum and A. gallica
As far as we know, this is the first report in Poland concerning interactions between the oomycete P. cactorum and the fungus A. gallica, inoculated on silver birch seedlings. The pathogen P. cactorum was already previously recorded on gooseberry [48]. However, in Finland [49] it was known as a primary agent responsible for destroying birch fine roots, but in our experiment A. gallica appeared to be more harmful than P. cactorum. Conventionally, it is presumed that Armillaria species follows primary infection as an opportunistic pathogen. The later pathogens are often stimulated by root wounds and biochemical changes caused by defoliation [50]. Our previous investigations confirmed that A. gallica prefers to attack roots, first affected by Phytophthora infection [38]. However, in this study we did not find any significant interactions between those organisms (Tables 1 and S1), although we found interactions in treatment 6 ( Table S1). This is an interesting result, because defoliation should affect the normal functioning of the hydraulic system of the tree. In our experiment, it seems that the joint effect of both stresses is then capable of causing greater damage to roots. In our experiment P. cactorum, inoculated via soil, caused more severe damage than when plants were weakened by defoliation (behaving as an opportunistic pathogen). For the moment we are not able to demonstrate by statistical analysis that P. cactorum significantly alters the physiology or the behavior of birch seedlings, nevertheless we confirmed its activity (by baiting) and presence (by isolations and qPCR) in the rhizosphere and root tissues of B. pendula. This is in line with other experiments performed on ash seedlings [51]. Inoculation of potted ash seedlings, Fraxinus excelsior, by P. plurivora via soil two months before shoot inoculation via wounds by Hymenoscyphus fraxineus, demonstrated induction of plant resistance and less mortality (by 40%) than expected, and achieved in H. fraxineus treatment alone (100%). However, P. cactorum is known as a primary pathogen, causing disease in silver birch seedlings, and can often be isolated in forest nurseries, e.g., using baiting (plant traps) from irrigation water ponds, or directly from necrotic lesions on stems [52].
Silver birch decline in Poland became more prevalent starting in the 1980s, but phytophthorosis was not detected, although cankers on their stems and reddish exudates were commonly noticed at the time [53,54]. The mortality of birches was instead connected at that time with severe drought and water uptake problems by roots damaged by Armillaria fungi. Armillaria root disease caused by A. ostoyae (Romagn.) Herink is a common disease in many parts of the Northern hemisphere, including western North America, and Europe [27]; and with changing climate the incidence of root disease and butt rot caused by A. ostoyae is likely to increase, causing direct and indirect tree mortality [55]. Nevertheless, most Armillaria fungi existing in broad-leaved forests are considered to be opportunistic secondary pathogens (like A. gallica), so they usually act after primary pathogens such as Phytophthora species. The combination of oomycetes (P. alni, P. plurivora, P. cambivora, and P. quercina) as primary pathogens and fungi (Armillaria spp.) as secondary ones was already noticed in alder, ash, beech, and oak stands in the UK and Serbia [2,56]. It was confirmed that A. gallica mainly attacks fine roots, which start to rot and separate from the main roots, while at the separation point, elongated wounds are followed by callus formation and characteristic scars [2]. Subsequently, the substances secreted from the scars stimulate the growth of rhizomorphs from the inoculum of A. gallica. The rhizomorphs usually attach to the surface of thicker roots, feeding saprophytically, but when plants are weakened, the fungus begins infection [57,58]. Based on our observations, we suggest that similar interactions take place in Polish birch stands. Although many birches died rapidly once the crown symptoms appeared (die-back), in other stands the disease took a chronic form with the loss of foliage and branch dieback increasing over time, suggesting the activity of some resistance or tolerance mechanism [59]. B. pendula can be infected by P. cactorum [48] and A. gallica in inoculation tests, but a clear pathogenic relationship needs to be demonstrated in the field. Our future work will focus on the additive effects between pathogen infections and defoliation of birch seedlings in lab and field studies.

Birch Damage (Defoliation) Caused by Insect Pests
In general, insects responsible for the defoliation of birch generally have one generation in June, such as Scolionera betuleti (Zdd.) [60], Messa nana (Kl.), which can also feed on silver birch [61], and Agromyza alnibetulae (Hend.) [62]. In our experimental design, the defoliation of birch seedlings was performed in June (similarly to insect attack in the forest), and aimed to mimic the defoliation caused by endophagous and folivore insects of birch, with defoliation of 50% of the leaves gradually occurring over a four-week period (25% of each leaf cut off per week). The endophagous and folivorous species very often weaken trees, but rarely cause mortality of deciduous trees species [63,64], and this was also observed in our experiments.
In our study, the partial defoliation of the host (B. pendula), probably simulated damage by an aphid group of folivorous insects, often occurring in broad-leaved forests, which do not cause direct total defoliation or immediate death of the host, but provoke weakening of the host through gradual and partial defoliation.

Effects of Stress Factors on Physiological Response of Plants
The impact of harmful biotic factors on forest tree species is often assessed by visual estimation of the general health condition of the crown [65,66]. P. cactorum and A. gallica directly destroy the plant root system, and in consequence influence the condition of foliage. After one year of our experiment, the influence of defoliation stress on plant health status was more pronounced than the influence of each pathogen alone. The lowest health status estimation was found with the treatment of both pathogens (Phytophthora and Armillaria) combined with defoliation. Perhaps the damage to roots caused by pathogens was masked by partial defoliation, which caused less visual disease symptoms. Since all investigated stress factors can affect birch root or foliage, we first attempted to examine the plant general response at the level of fine root damage, and later, at the level of photosynthetic activity. We discuss the results in the following sections.

Root Status of Treated Plants
In our study, we placed P. cactorum directly into the soil, to simulate infections in the natural conditions in forests, without wounding, because that is a very invasive method. After inoculation and observation of disease symptoms, and through baiting techniques and observing infection of leaves (oak and beech), this demonstrated the viability of P. cactorum inoculum. By re-isolation on artificial media (PARPH), and obtaining pure cultures, and comparing them with the original P. cactorum inoculum we showed that the pathogen was present and alive. The observed increase of phenolic compounds in birch roots, and the development of rhizomorphs of A. gallica attached to roots, demonstrated that the infection method we had chosen was successful [50].
The combination of the stress factors (two soil borne pathogens and defoliation), decreased significantly the development of root collar thickness in birch seedlings. Surprisingly, P. cactorum alone seemed not to play an important role in the development of this growth parameter. Statistically significant decrease of birch fine root length by ca. 75% and fine root surface areas by ca. 73% took place only in the cases of P. cactorum + A. gallica + 50% defoliation, and P. cactorum + A. gallica. Lower level damage (15%) of the root system caused by rot has been observed by Mauer and Palátová [25] on root systems of 20-year-old birches affected by defoliation, and on root systems of 15-year-old birches in the Ore Mountains (Czech Republic). In the case of the Ore Mountains study, the dominant fungi developing on roots were A. gallica (like in the Polish case) and A. ostoyae (also very common in Polish forests). Defoliated trees showed a clear loss of small roots, deteriorated longevity of small roots, and a switch from ectomycorrhizae to ectendomycorrhizae composition [25].

Photosynthetic Activity
The only observed increase of total PI was in plants infected by P. cactorum, and this was probably due to a triggered natural mechanism of compensation, as a response of birch leaves to the primary agent of fine root infection. Figure 3 shows average the change regarding control treatment, which was supported by statistical analysis (Table S1). Defoliation caused by pests and pathogens are often reflected in changes in fluorescence parameters [67][68][69]. In our experiments, the interaction with the two pathogens and partial defoliation diminished by 30% the value of the total performance index (PI total), compared to untreated plants. The only observed increase of total PI was in plants infected by P. cactorum, and this was probably due to the triggered natural mechanism of compensation, as a response of the birch leaves to the primary agent of fine root infection. At least at the beginning, the strategy of plants fighting against a recognized pathogen can be increasing the efficiency of photosynthesis [70,71]. Similar decrease of the PI parameter was observed in wheat, after infection by Fusarium sp. Due to the high sensitivity of the photosynthetic apparatus, chlorophyll fluorescence has been used for early detection of plant disease in asymptomatic plants, enabling prevention measures, e.g., against fusarioses in crops [72]. Physiological stress assessed by photosynthesis effectiveness, resulting in PI total <1, can be informative about weakened plants, susceptible to natural infections by nearby pathogens.

Effects of Stress Factors on Host Chemical Compound Production
Plants respond to harmful biotic factors by multiple mechanisms of defense, including production of specific defense compounds, such as aromatic esters (benzyl acetate, methyl salicylate), aromatic alcohols (p-cresol), and monoterpenes (neral, geraniol, eugenol), as well as sesquiterpenes (α-copaene, β-bourbonene), and oxygen derivatives of these two groups of chemicals [73][74][75]. Generally, the secondary metabolites have two major roles in stressed plants: (i) the development of induced systemic resistance (ISR), or systemic acquired resistance (SAR) mechanisms, and (ii) the direct protection of plants against harmful pests and pathogens [76].
In our experiment, the signaling molecules secreted by birch leaves comprised molecules involved in SAR development, i.e., monoterpenes (geraniol and eugenol), triterpenes (lupeol, betulin) as well as aromatic esters, including methyl salicylate. The amount of compounds known as important signaling molecules in SAR increased substantially in birch leaves and roots subjected to defoliation, or pathogen infection coupled with defoliation treatment. Insect attack is known to trigger SAR responses [77] and ISR responses [78], which presumably involve the production of volatile compounds that may deter or enhance further insect attack. Defoliation-associated induction of resistance has been found for birch [79,80].
The effects of interactions between soil borne pathogens on the production of secondary metabolites by birch (such as those leading to SAR development) have not been sufficiently studied, and therefore we focused on the effects of the fungus (A. gallica) and the oomycete (P. cactorum) on the chemical composition of compounds secreted by leaves and roots. Although statistical differences between treatments were found only for sesquiterpenes, qualitative and quantitative assessment of VOC emission is provided in supplementary material, and discussed in a similar way to other related work [44]. The significantly lower content of sesquiterpenes in the leaves of the treated birches (especially during A. gallica infection) compared to the control, may have resulted from the redirection of these substances from leaves to the roots, where they can directly interact with the pathogen. Due to their lipophilic properties, sesquiterpenes and phyto-flavonoids produced after pathogen penetration into plant cells, take part in the direct inhibition of harmful microorganisms [81]. We contend that A. gallica benefited from the weakening of the plants (after plant defoliation), and this was reflected in the level of defensive substances (phenolic compounds and triterpenes), which were reduced (compared to defoliation alone, or Armillaria treatment alone), hence the infection process. Phytophthora could also mask its presence in plants, because the birch seedlings did not react like in the case of root infection by Armillaria, as the level of phenolic compounds in the roots rose only by about 5-6 times (not 15 times like in the case of the fungus). Keča et al. [51] showed that P. plurivora, introduced into the soil two months before stem inoculation of ash seedlings with Hymenoscyphus fraxineus (T. Kowalski, Baral, Queloz, Hesoya, comb. nov.), increased plant tolerance to fungal disease.
Qualitative and quantitative compositions of extracts of naturally and artificially defoliated birch leaves in Russia revealed that the content of free sterols, triterpene compounds, and flavones decreased one year after defoliation, but the amount of flavanones and flavanonols increased [82]. The increased production of such secondary compounds has protective functions, as a part of the plant's strategy in the fight against harmful insects [83,84]. The concentration of antimicrobial compounds increased in the leaves of defoliated red oaks, in comparison to the non-defoliated plants, but the total amounts of nitrogen, sugars, proteins, starch, lignin, and hemicellulose were lower after one year of defoliation [85,86]. The defoliated trees had reduced amounts of nutrients, i.e., monosaccharides, amino acids, or fatty acids, in order to reduce their attractiveness for herbivorous insects [87].
The content of secondary metabolites in birch tissues may be influenced by many other factors, such as: species, age, or health condition [88]. We only dealt with young B. pendula, growing under different stress conditions, and probably cannot fully extrapolate these results directly to adult trees. Nevertheless, defoliation can invoke the SAR mechanism of defense, especially if it occurs for the first time and is accompanied by simultaneous root infection [89]. Another aspect worth studying in the future is coordinated defense mechanisms by distance, e.g., several VOCs or terpenoids may be involved in signaling among neighboring poplar species attacked by insects or herbivorous animals [33].

Summing up and Conclusions
In our study we focused only on the species of Armillaria as fungi, and these often associate with trees (rhizomorphs are superficially attached to the external bark at the root collar); and when host plants are weakened, the infection process usually starts. However, the role of other fungi in this process needs more investigation. They may accelerate the speed of tree death (synergy) or slow it down if fungi show antagonistic interactions, such as Trichoderma spp. killing rhizomorphs of Armillaria [90]. The occurrence of P. cactorum in a forest nursery usually causes problems with plant health because it can be pathogenic to many forest tree species. In our research we wanted to demonstrate changes in the physiology or the chemical composition of the plant due to the effect of P. cactorum. We were wondering if it were possible that, even if P. cactorum colonizes the root, the mere presence of the pathogen does not produce changes in the hydric balance and the physiological activity of the tree, providing that there is enough water in the soil. Therefore, we stressed plants (defoliation, inoculation) to recognize changes in the physiology and in the chemical composition of birch seedlings, due to the effect of P. cactorum alone, or in combination with Armillaria. In our experimental conditions we confirmed the damage to roots and as a consequence, to shoots, but the photosynthetic efficiency did not differ between the control and P. cactorum treatment. However, inoculation of birch seedlings with P. cactorum increased the amount of phenolic compounds compared to the non-inoculated control, indicating a plant response to infection. The heavily stressed plants also showed changes in triterpenes in roots and leaves. Similar observations were made for A. gallica, inoculation alone or in combination with artificial defoliation. In forests, these factors, among many others, can appear together in the same place and time, leading to the death of a tree or whole ecosystem [91].
Finally, the combination of stress factors in our experiment diminished the health of the investigated birch seedlings, including decreased root collar diameter, decreased number of tips (including fine roots), and decreased root length for fine roots and mother roots, among other effects. Taken together, the cumulative effect of different stress factors in a forest are major determinants of health. We hypothesized that P. cactorum is not more aggressive in combination with A. gallica, because A. gallica does not enhance the effect of the P. cactorum, or vice versa, nor does P. cactorum infection make the host more sensitive to the action of the Armillaria. After statistical analysis we can affirm that the evaluated parameters (the number of root tips, fine root length, fine root surface area, and fine root tips) in treatment (8), Armillaria gallica + Phytophthora cactorum, significantly diminished compared to the control (cumulative effect). On the other hand, two-way ANOVA revealed no significant interactions, which implied that effects were additive, meaning that the presence of P. cactorum in addition to other stress factors, such as defoliating insects and Armillaria gallica would have increasingly aggravating effects on plant health.
In conclusion, this research on birch seedlings subjected to partial defoliation, fungal infection by A. gallica, oomycete infection by P. cactorum, and combination thereof, demonstrated the following: (i) the oomycete P. cactorum can worsen health of silver birch (B. pendula) seedlings, (ii) the mortality of birch seedlings may occur after one growing season due to an interaction of P. cactorum with A. gallica, (iii) the fungus A. gallica caused significant loss of the photosynthetic apparatus efficiency compared to the control, (iv) infection by A. gallica caused a decrease of sesquiterpenes content in the leaves of the treated birches, revealing possible redirection of these groups of chemical compounds from the leaves to the roots, where they can directly interact with the pathogen during SAR.

Funding:
The research was carried out as part of the work No. WZ/WB-INL/1/2019 at the Białystok University of Technology and financed from a research subsidy provided by the Minister of Science and Higher Education. Part of the study was financed by the Forest Research Institute as a scholarship to Joao Maria Caetano who contributed to this paper during his three-month stay in the Institute as its fellow.