White rot fungal impact on the evolution of simple phenols during decay of silver fir wood by UHPLC‐HQOMS

Abstract Introduction Silver fir ( Abies alba Mill.) is one of the most valuable conifer wood species in Europe. Among the main opportunistic pathogens that cause root and butt rot on silver fir are Armillaria ostoyae and Heterobasidion abietinum. Due to the different enzymatic pools of these wood‐decay fungi, different strategies in metabolizing the phenols were available. Objective This work explores the changes in phenolic compounds during silver fir wood degradation. Methodology Phenols were analyzed before and after fungus inoculation in silver fir macerated wood after 2, 4 and 6 months. All samples were analyzed using high‐performance liquid chromatography coupled to a hybrid quadrupole‐orbitrap mass spectrometer. Results Thirteen compounds, including simple phenols, alkylphenyl alcohols, hydroxybenzoketones, hydroxycinnamaldehydes, hydroxybenzaldehydes, hydroxyphenylacetic acids, hydroxycinnamic acids, hydroxybenzoic acids and hydroxycoumarins, were detected. Pyrocatechol, coniferyl alcohol, acetovanillone, vanillin, benzoic acid, 4‐hydroxybenzoic acid and vanillic acid contents decreased during the degradation process. Methyl vanillate, ferulic acid and p‐coumaric were initially produced and then degraded. Scopoletin was accumulated. Pyrocatechol, acetovanillone and methyl vanillate were found for the first time in both degrading and non‐degrading wood of silver fir. Conclusions Despite differences in the enzymatic pool, both fungi caused a significant decrease in the amounts of phenolic compounds with the accumulation of the only scopoletin. Principal component analysis revealed an initial differentiation between the degradation activity of the two fungal species during degradation, but similar phenolic contents at the end of wood degradation.


| INTRODUCTION
Phenolic compounds are commonly produced as secondary metabolites from plant and fungal species. In plants they constitute one of the most common and widespread groups of substances which arise biogenetically from the pentose phosphate, shikimate and phenylpropanoid pathways. The plant-associated fungi have symbiotically adopted these pathways into their metabolic cycle and mimic the plants by producing phenols. 1 Plants need phenolic compounds for antifungal activity and resistance to pathogen growth; moreover, both plants and fungi use them for pigmentation, reproduction and many other functions. 2 The fungi Armillaria spp. and Heterobasidion spp. have dual life strategies, being necrotrophic on living trees and subsequently saprotrophic on dead wood. As both fungi are considered white rot fungi, they have a versatile machinery of enzymes to attack directly the lignin barrier. The two fungal genera differ also significantly in symptomatology. 3 Recently, the genomes of a European and a North American Armillaria ostoyae strain were published 4

while the
Heterobasidion abietinum sequence is not described yet. Interestingly, in comparison with other white rot fungi, Armillaria shows an underrepresentation of ligninolytic gene families and an overrepresentation of pectinolytic gene families. 4 Accordingly, recent studies considered Armillaria spp. as white rot fungus, based on the presence of genes encoding lignin-decaying enzymes in their genomes. [4][5][6][7] However, previous studies have also shown that Armillaria species primarily decay the cellulose, hemicellulose and pectin components of the plant cell wall, and leave lignin unattacked during early stages of decay. 8 They have been discussed as white rot species, though their response to wood deviates from that of typical white rotters. While we observed an upregulation of a diverse suite of plant cell wall-degrading enzymes, unlike white rotters, they possess and express an atypical wood-decay repertoire in which pectinases and expansins are enriched, whereas lignin-decaying enzymes are generally downregulated. 9 Heterobasidion spp. have more gene families typically involved in lignin degradation or modification, including laccases and peroxidase.
Results for wood block degradation correlated well with the ability of the Heterobasidion spp. to produce laccase in liquid and solid culture conditions, with H. annosum ss. producing ca 5-6 times more laccase than H. parviporum, indicating that great differences exist between Heterobasidion species' abilities to cause wood decay. 10 The analysis of the Heterobasidion irregulare genome revealed a repertoire of genes encoding lignocellulose-degrading enzymes, including 179 glycoside hydrolases (GHs), eight manganese peroxidases (MnPs) and 17 multicopper oxidases (MCOs). 11 In a study on the degradation of pine wood during saprotrophic growth of H. annosum ss., the induction of many GHs, MCOs, five MnPs and one oxidoreductase was observed, being specific for wood degradation. A total of 31 predicted GH genes were found upregulated in heartwood, 20 in sapwood and 23 in bark compared to the control. 12 In Heterobasidion parviflorum, a close relative of H. abietinum, it was proved that in addition to transcriptome variation, also variation in the methylome (DNA cytosine methylation) is an important epigenetic modification in the lifestyle transition of this fungus. 13 After wounding and inoculation of the bark of Sitka spruce, different concentrations of cell wall-bound phenolic compounds from the necrotrophic lifestyle of H. annosum were found, including unknown2, unknown3, coniferin, astringin, taxifolin, piceid and isorhapontin, whereas in sapwood the concentrations did not differ following treatment. These results indicate that bark of Sitka spruce has a stronger and earlier response to wounding and pathogen inoculation than sapwood. 14 White rot fungi have been investigated extensively since the mid-1980s for their bioremediation capacities; 15 in fact, they are considered the only organisms able to completely decompose lignin into CO 2 and water. [16][17][18] Lignin, the second largest sink of fixed carbon, after cellulose, 19 is a complex phenolic biopolymer that plays a central role in mechanical support of plant cell walls, water transport and pathogen resistance in plants. 20 The lignin molecule can be composed of three different phenylpropane monomer units (monolignols), namely, para-coumaryl alcohol, coniferyl alcohol and sinapyl alcohol, linked by ether and carbon-carbon type bonds. 21 The composition and amount of this polymer vary depending on the different botanical groups (conifers vs. broadleaves), between different tree species and even between the different woody tissues of the same tree. 22 Conifers (softwood) are known to contain high amounts of lignin consisting mainly of guaiacyl units (90%) derived from coniferyl alcohol; broadleaf trees (hardwoods) and herbaceous species contain similar amounts of both guaiacyl and syringyl units derived from coniferyl alcohol and sinapyl alcohol, respectively. 23,24 Moreover, softwood lignin is estimated to be comprised of about 19% to 26% of phenolic units, whereas hardwood lignin contains about 14 to 18% of phenolic units. 25 Due to the chemically complex structure, lignin polymer is highly resistant to physical, chemical and biological degradation. 26 In order to depolymerize and mineralize the complex lignin molecule, these fungi secrete various combinations of strong extracellular oxidative lignolytic enzymes known as "ligninases". 27 Ligninases include mainly lignin peroxidase, MnP and laccase. 28,29 In particular, lignin peroxidase directly attacks the non-phenolic lignin model compounds, such as veratryl alcohol (VA), by producing intermediate radicals, 30,31 whereas Mn-dependent peroxidase and laccase are able to oxidize the phenolic lignin units to phenoxy radicals, leading to the decomposition of the woody structures. 32,33 Different combinations of these enzymes produced by white rot fungi underlie different mechanisms of lignin degradation with the production of various phenolic compounds. [34][35][36] For example, the degradation of guaiacylβ-coniferyl ether in softwood lignin (e.g. pine and spruce lignins) leads to the formation of coniferyl alcohol, coniferylaldehyde, ferulic acid, several low-molecular weight aromatic acids and aldehydes, including vanillin and vanillic acid. [37][38][39][40] Other common degradation products of hardwood are syringic acid, syringaldehyde, protocatechuic acid and gallic acid. 41 Many studies have extensively investigated the role and activity of ligninolytic enzymes produced by white rot fungi, such as Phanerochaete chrysosporium, during lignin depolymerization. [42][43][44][45] Several studies have also described different strategies of fungi in metabolizing the phenolic compounds during wood degradation 24,46,47 ; however, only a preliminary study has focused on silver fir (Abies alba Mill.) wood which evaluated the distribution and variation of extractable total phenols and tannins in the logs of four conifers after 1 year on the ground. 48 Silver fir, one of the most valuable conifer wood species in Europe, is widely distributed in Central and Southern European forests and therefore of significant ecological and economic value. 49 Among the main opportunistic pathogens that cause root and butt rot on silver fir are A. ostoyae 50 and H. abietinum, 51,52 which was previously called Heterobasidion annosum F-group. 53 The infections caused by these fungi can spread from tree to tree through root connections. When a host tree dies or is cut, both pathogenic fungi act as strong saprophytic organisms 54,55 ; consequently, these fungi are of major importance in the decay of living trees and of dead silver fir wood.
Various chromatographic techniques have been used to identify and quantify phenolic compounds of lignin, including gas chromatography (GC), liquid chromatography (LC), size exclusion chromatography (SEC), capillary electrophoresis (CE) and two-dimensional chromatography. 56 However, among these analytical techniques, the coupling of high-performance LC and mass spectrometry (HPLC-MS) proved to be a powerful technique for the analysis of low-molecular weight compounds such as phenols with high selectivity and sensitivity. 57,58 Considering the already well-described lignin degradation realized by rot fungi 59 and also in vitro degradation, 60,61 this work aimed to study how the different enzymatic pools of the two rot fungi influenced the trend of phenolic compounds at different wood degradation periods (2, 4 and 6 months) on macerated silver fir wood.
Phenolic profiles were explored before and after fungal inoculations in the laboratory using HPLC coupled to a hybrid quadrupole-orbitrap mass spectrometer (LC-Q-Orbitrap).

| Evaluation of sample preparation
The extraction procedure is an important step for the quantification of wood extractives, such as phenolic compounds. The extraction yield of phenolics is affected by several factors, such as the solvents used with varying polarities, the sample-to-solvent ratio, extraction time, temperature and the characteristics of the sample. 63

| Fungal strains, culture conditions and sample preparation
Two different species of white rot fungi, A. ostoyae and H. abietinum, obtained from the fungal culture collection of the Pathology Lab of the Edmund Mach Foundation (TN), were used. In order to promote mycelium growth, both fungal species were axenically cultivated on 20 g L À1 of sterile PDA plates for about 20 days at room temperature.
A total of 12 Petri dishes were prepared (six replicates for each fungal species). After fungal growth, the mycelium of A. ostoyae and H. abietinum was scratched off from the plates and cut into small pieces using a sterile scalpel blade and then inoculated axenically onto 3 g of sterile macerated silver fir wood in 18 sterile glass vials of 10 cm 3 (Sartorius AG, Goettingen, Germany), in nine replicates per each fungal species, out of which three were harvested per time point.
Macerated wood was prepared by milling sapwood of silver fir with an M20 mechanical mill (IKA-WERKE, Staufen, Germany) to obtain wood chips of about 3 mm in diameter. Silver fir sapwood was previously taken from a living tree located in the "Abeti Soprani" forest in the Molise Region (Italy) using a chainsaw. To promote fungal growth in macerated silver fir wood, 5 mL of sterile ultra-pure water was added in each glass vial. 66 In addition to the vials with fungal inoculum, three other vials were prepared as controls (t0), containing only sterile macerated wood, which were stored at À20 C. The inoculated vials were stored at room temperature and after two (t1), four (t2) and six (t3) months, six of them (three for each fungus) were put in the

| LC-HRMS analysis
The analysis of phenolic compounds was carried out according to the method described by Barnaba et al. (2018). In particular, the identification and quantification of these compounds was performed using a

| Method validation
The characteristics of the method were studied using the 13 pure

| Statistical analysis
Statistical analysis was performed using XLSTAT (version 2020, Addinsoft, France). Significant differences between the concentrations of phenolic compounds measured during different times of degradation were determined using the Kruskal-Wallis test (p < 0.05).
Principal component analysis (PCA) was carried out to evaluate the relationships between phenolic compounds at different times of silver fir wood degradation and the activities of two fungal species.

| Evaluation of sample preparation
The evaluation of the best conditions for sample preparation was conducted on a sample of silver fir sapwood taken from a living tree  (Table 2).

| Method validation
Phenol quantification was performed on precursor ions detected in the extracted ion chromatograms (EICs) corresponding to the deprotonated molecules [M-H] À . Due to confirmed sample compounds, accuracy-mass tolerance was set at <5 ppm and RT and dd-MS/MS spectra were compared with those collected from available standards (Figure 1).
The method characteristics including LOQ, linearity range, precision and accuracy determined for each phenolic compound are shown in Table 3. Accuracy, evaluated in terms of recovery (%), was between 40% and 77% in the samples supplemented with 0.

| Phenolic compounds in silver fir wood
Several phenolic compounds were identified in silver fir wood samples. As reported in Table S1, among simple phenols, a low content of pyrocatechol (from 0.29 to 0.44 mg kg À1 ) was detected. To the best of our knowledge, this compound was found for the first time in fresh silver fir wood, while other studies showed the production of pyrocatechol due to the action of ligninolytic enzymes of white rot fungi during lignin degradation. 69

| Effect of fungal activity on phenolic components
The changes in the phenolic profiles of silver fir wood, in relation to the activity of the two different fungal species belonging to the Armillaria and Heterobasidion genera, were evaluated during 6 months of wood degradation. The phenol content was analyzed at 2, 4 and 6 months. Table S1 summarizes the phenolic compound content, while the trends of these compounds at different times of degradation can be observed in Figure 3(a, b).
Specifically, considering phenols already available in macerated wood, a similar decay pattern was observed for coniferyl alcohol, vanillin, benzoic acid, 4-hydroxybenzoic acid and pyrocatechol for both rot fungi.
In particular, vanillin, benzoic acid, 4-hydroxybenzoic acid and Benzoic acid and 4-hydroxybenzoic acid were already identified as lignin degradation products, [78][79][80] and several studies reported that benzoic might arise after the oxidative cleavage of the α and β carbons of the alkyl side chain by MnP. 31,81 As regards pyrocatechol, studies conducted on the white rot fungus P. chrysosporium highlighted the production of this compound due to the oxidation of β-O-4 linkages of lignin by lignin peroxidase. 69,70 However, other studies indicated a decrease of pyrocatechol levels after the action of laccase with the subsequent conversion to phenoxyl radicals through oxidation processes. 19,82 Regarding coniferyl alcohol, the decreasing trend during the degradation process appeared much slower for both fungi and the consumption of phenols was not complete. Coniferyl alcohol showed a concentration significantly lower after 6 months (t3) from fungal inoculation compared to the other time points (Table S1).
This alcohol is reported as one of the most abundant Note: RT = retention time; LOQ = limit of quantification.*Linearity ranges and LOQs are defined without considering sample dilution. monolignols of softwood lignin polymer 23 and several studies described its degradation by fungi and bacteria with the consequent production of ferulic acid, coniferylaldehyde and vanillic acid. 76,83,84 Among the other phenols present in wood at t0, acetovanillone levels showed a decreasing trend for both fungal species, but for During wood degradation, the formation of coniferaldehyde, pcoumaric acid and scopoletin was observed; these compounds were initially totally absent in the macerated silver fir wood. At 2 months (t1) after the inoculation of both fungal species, the coniferylaldehyde concentration exceeded 1 mg kg À1 and subsequently decreased until this compound was almost completely degraded at 6 months (t3).
Several studies reported the production of this compound after the enzymatic degradation of coniferyl alcohol by white rot fungi. 84,90 Moreover, Falconnier et al. 76

| Phenolic profiles and degradative fungal activity
PCA was applied for the content of each phenolic compound quantified in macerated silver fir wood samples, in order to evaluate the correlations between phenolic profiles and the activities of two fungal species at different times of wood degradation (Figure 4). PCA, with F I G U R E 3 (a, b) Box plots with phenolic compound content (mg kg À1 and μg kg À1 for isoacetosyringone) at times t0, t1, t2 and t3 (different times of macerated silver fir wood degradation) by Armillaria ostoyae and Heterobasidion abietinum PC1 and PC2 collectively accounting for 76% of total variance, revealed a good differentiation between the activities of the two fungal species during the first period (at t1 and t2) of silver fir wood degradation. In particular, PC1 clearly separates the data based on