Fungi associated with decomposing deadwood in a natural beech-dominated forest

Fungal Ecology 23 (2016) 109e122

Contents lists available at ScienceDirect

Fungal Ecology journal homepage: www.elsevier.com/locate/funeco

Fungi associated with decomposing deadwood in a natural beechdominated forest  a, Vojte ch Tla skal a, Anna Davidova  a, Ve ra Merhautova  a, Petr Baldrian a, *, Petra Zr
ustova s Vrska b Toma a b

 1083, 14220 Praha 4, Czech Republic  ska Laboratory of Environmental Microbiology, Institute of Microbiology of the CAS, Víden  25/27, 602 00 Brno, Czech Republic Department of Forest Ecology, The Silva Tarouca Research Institute for Landscape and Ornamental Gardening, Lidicka

a r t i c l e i n f o

a b s t r a c t

Article history: Received 13 January 2016 Received in revised form 23 June 2016 Accepted 1 July 2016 Available online 10 August 2016

Deadwood represents a specific habitat of particular importance in natural, unmanaged forests where wood is not harvested. Here, we characterized the basic wood chemistry, enzyme activity, fungal biomass content and community composition of Fagus sylvatica, Abies alba and Picea abies coarse woody debris decomposing for <5, 5e15, 16e38 and > 38 years in a natural temperate forest. The results indicate that coarse deadwood represents a highly diverse substratum in terms of the quality, fungal biomass content and, in particular, the composition of fungal communities whose properties change with time. Because sequences recovered from individual logs were typically dominated by one or few fungal species, which were rarely tree species-specific, the community assembly appeared to show a high level of stochasticity. Among the estimated variables, nitrogen content that increased with decay length was the most important candidate driver of fungal biomass content, community composition and enzyme activity. © 2016 Elsevier Ltd and British Mycological Society. All rights reserved.

Corresponding editor: L. Boddy. Keywords: Deadwood Decomposition White-rot Brown-rot Extracellular enzymes Succession Community assembly Mixed natural forest

1. Introduction Deadwood represents a specific habitat whose abundance varies greatly among forests. Whereas deadwood volume is typically low in managed forests where most wood is harvested, it can be very high in natural forests, where it can represent a C stock of a coml parable or even greater size than that of standing tree biomass (Kra et al., 2010a; Stokland et al., 2012). In the natural forests of Europe, deadwood volume typically ranges in hundreds of m3 ha
1, reaching up to 1200 m3 ha
1, compared to the 2e65 m3 ha
1 stock volume typical of managed forests (Stokland et al., 2012). Reflecting physical and chemical properties, such as impermeability, high lignin content and low N concentrations, wood is resistant to the rapid penetration by microorganisms and represents a nutrient pool with a slow turnover. Deadwood decomposition is dominated by fungi, especially basidiomycetes and xylariaceous ascomycetes

* Corresponding author. E-mail address: [email protected] (P. Baldrian). http://dx.doi.org/10.1016/j.funeco.2016.07.001 1754-5048/© 2016 Elsevier Ltd and British Mycological Society. All rights reserved.

(Rayner and Boddy, 1988) that, due to their unique ability to efficiently decompose impermeable wood biopolymers, colonize wood and use the nutrients within it for their growth (Eichlerov a et al., 2015). In addition to its importance in C turnover, deadwood also represents a specific habitat for bacteria living under the strong skova  et al., 2009; Hoppe et al., 2015). influence of fungi (Vala The climate, soil properties and sun exposure are important determinants of fungal community composition in deadwood at large spatial scales (Seibold et al., 2015), but different drivers may apply in local environments that include the species-specific properties of wood. Because many fungal taxa colonize fresh wood, the assembly and development of fungal communities seems to be rather stochastic. The identity of primary colonizers together with interspecific interactions of fungi largely determines the establishment of later arriving species, a phenomenon referred to as the priority effect (Boddy, 2000; Fukami et al., 2010; Lindner et al., 2011; Hiscox et al., 2015). As a result of the initial colonization, dead wood decomposition may follow different paths, such as brown-rot or white-rot decay, resulting in profound differences in the chemistry and decomposition rates (Baldrian, 2008; van der

110

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

Wal et al., 2015). Despite this theoretical stochasticity, the current view of fungal community assembly on dead wood is that the tree species and decay stages are major drivers of fungal community composition (Jonsson et al., 2008; Kuffer et al., 2008). Because fungal community composition in dead wood was traditionally revealed by surveying macroscopic fruitbodies, it was not possible to assess the factors affecting the development of the entire community. Advances in molecular analysis based on fungal DNA and RNA offer an alternative approach to investigating the entire fungal community. The results of fruit body surveys and molecular analyses of the same logs were found to be generally consistent when surveying those fungi forming large fruiting bodies, although mycelial occurrence was not confined to those decomposition stages where the fruit bodies appeared (Ovaskainen et al., 2013). In addition, molecular analyses typically discover additional taxa that fail to fruit. Wood decomposition typically takes tens of years to complete and is characterized by the successive development of fungal communities with an initial dominance of decomposers (Rayner and Boddy, 1988), but also the support of ectomycorrhizal (ECM) fungi during late decay (Rajala et al., 2011). Decomposition represents an intricate interplay between nutrient depletion, chemical changes, development of fungal communities, which includes a network of feedbacks. The long residence times of dead wood make it difficult to experimentally capture the complete fungal community development in dead wood. Thus, controlled experiments are typically short term, and rather than time, environmental surveys generally rely on the visual classification of dead wood into decay stages or wood density measurements to describe the progress of decomposition (Lindner et al., 2011; Rajala et al., 2011). Here, we employed a unique dataset of surveys of living and dead trees in the mixed natural forest of Central Europe to classify decomposing coarse woody debris (CWD, tree trunks) into decay length classes instead of decay stages. Moreover, the availability of a complete database of dead wood with diameters >10 cm made it possible to randomly select CWD before the field survey in a way that fairly represents the diversity observed in the field. This made the dataset highly representative for the description of the fate of dead wood in a mixed natural forest. The aims of this work were to: (1) describe the changes in wood chemistry during decay and the production of fungal biomass at a defined timescale, (2) describe the composition of fungal communities in individual CWD and identify the drivers of community assembly; and (3) identify the factors affecting the activity of enzymes involved in wood decomposition. We hypothesized that, similar to litter and soil, CWD tree species largely determines the fungal community composition. Peak fungal biomass and enzyme production should be expected during early decomposition when wood is fully colonized, but nutrients are less limiting, such as observed in the case of another complex lignocellulose substrate,  plant litter (Snajdr et al., 2011a; Urbanov a et al., 2014). Due to stochastic assembly, high levels of variation in community composition among logs of the same species were expected, but the abundance of individual fungal decomposers was expected to be driven by the CWD age category, where they find the most suitable conditions for their existence. With respect to enzyme activity, it was expected that this reflects the type of decay, decay stage and the composition of fungal community in a CWD. 2. Materials and methods 2.1. Study area and deadwood data collection  Hory mountains, The study area was located in the Novohradske specifically in the 25 ha Zofin ForestGEO® Dynamics Plot (www.

forestgeo.si.edu), which is the part of 42-ha core zone of the  Zofínský prales National Nature Reserve in the Czech Republic (48 390 5700 N, 14 420 2400 E). This core zone of the forest reserve had never been managed, and stands under protection since 1838. It thus represents a rare fragment of virgin forest. The reserve is situated along an altitudinal gradient of 735e830 m a.s.l.; gentle NW slopes predominate. Bedrock is almost homogenous and consists of finely to medium-grainy porphyritic and biotite granite. Annual average rainfall is 866 mm and annual average temperature is 6.2  C (Anderson-Teixeira et al., 2015). At present, the reserve is covered by a mixed forest. Fagus sylvatica predominates in all diameter classes (51.5% of total living wood volume), followed by Picea abies (42.8%). Abies alba represents 4.8% of standing volume and the remaining tree species (Ulmus glabra, Acer pseudoplatanus, Acer platanoides, Sorbus aucuparia) together represent 1% of standing volume; the living tree volume is calculated at 690 m3 ha
1 (Kr al et al., 2014). The volume of coarse woody debris l (CWD) is 102e310 m3 ha
1 with an average of 208 m3 ha
1 (Kra  et al., 2010a; Samonil et al., 2013). The representation of F. sylvatica, P. abies and A. alba is much more even in dead wood, representing respectively 23.6%, 43.7% and 31.4% of the total voll et al., 2014). The positions (coordinates X, Y, Z) of all trees ume (Kra with diameter at breast height (DBH) 10 cm, and selected tree parameters (tree species, DBH, tree status e live/dead, standing/ lying, snag, breakage, windthrow, stump etc.) were recorded repeatedly in 1975, 1997, 2008 and 2013, and a stem-position (incl. lying deadwood) map resulted from each census (Kr al et al., 2010a). Each CWD was classified into one of three decay stages in each census. The three decay stages were defined as follows: ‘‘H” e hard: relatively healthy and hard wood (having hard woody biomass was the main distinctive feature), tree species was recognizable, and the stem still had bark (although not necessarily); ‘‘T” e touchwood: the wood was not compact along the entire stem length, with the core or outer mantle subjected to rot, tree species was recognizable. Touchwood was a relatively widely defined class: in practice everything that did not belong to the ‘hard’ or ‘disintegrated’ classes; ‘‘D” e disintegrated: the wood was at a stage of advanced rot, and the species could no longer be identified. Kicking the stem resulted in stem breakage, and ‘‘little graves” with patchy vegetal et al., 2014). The information tion were frequently observed (Kra on deadwood and the records about the corresponding trees when alive allowed us to track the fate of individual CWD and to randomly select those with appropriate properties. 2.2. Study design and sample collection For the present study, we preselected CWD (tree trunks) belonging to three dominant tree taxa (F. sylvatica, P. abies, A. alba) with initial DBH ranges at the time when first recorded fallen between 30 and 100 cm. This selection comprised the bulk of the deadwood volume in the ecosystem. CWD with DBH <30 cm that rapidly decomposed and those trunks with DBH >100 cm that could not be representatively sampled, were excluded. Trees that decayed while standing were also excluded because the length of decomposition was unclear. Preselection resulted in a total of 739 stems, of which 255 stems were beech, 323 stems were spruce and 161 stems were fir trees. The CWD was sampled in a manner that ensured even coverage of all tree species and all classes of decay length (CWD first recorded as decaying in 1975, 1997, 2008 and 2013). Within each group, trees were randomly selected considering the equal representation of DBH size classes between 30 and 100 cm. The overview of selected trees is shown in Table 1. The CWD types were designated as FS, AA and PA, indicating the CWD tree species (“CWD species”) and decay lengths of <5, 5e15, 16e38 and > 38 y. The distance between logs typically ranged in tens of

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122 Table 1  Coarse woody debris sampled in the Zofín natural forest.

Sampled CWD (all CWD) Recorded as CWD in 1975 First recorded as CWD in 1997 First recorded as CWD in 2008 First recorded as CWD in 2013

Fagus sylvatica

Picea abies

Abies alba

39 (250) 5 (5) 12 (60) 13 (176) 9 (9)

43 10 10 12 11

36 (160) 7 (7) 13 (74) 13 (76) 3 (3)

(323) (36) (167) (106) (14)

meters. Four CWD samples were obtained from each selected log in October 2013 using an electric drill with a bit diameter of 8 mm. The length of each CWD unit (or the sum of the lengths of its fragments) was measured, and samples were collected at 1/8, 3/8, 5/8 and 7/8 of the CWD length. Drilling was performed vertically from the middle of the upper surface up to a depth of 40 cm. The drill bit was sterilized between drillings, and sawdust was collected in batches of two adjacent drill holes in sterile plastic bags and frozen within a few hours after drilling.

111

the manufacturer's instructions. Briefly, cells were lysed using SL1 lysis buffer. Enhancer SX was added prior to lysis. The samples were homogenized using FastPrep-24 (MP Biomedicals, Santa Anna, USA) at 5 m s
1 for 2  30 s. In the last step, DNA was eluted from the columns using 50 ml of elution buffer SE (5 mM Tris/HCl, pH 8.5). Two replicate extractions per sample were performed, combining the sawdust obtained from two adjacent drill holes. After extraction, DNA from both extractions was combined to represent the whole CWD. For the microbial community analysis, PCR amplification of the fungal ITS2 region was performed using barcoded gITS7 and ITS4 (Ihrmark et al., 2012) in three PCR reactions per sample as  ca kova  et al., 2016). Triplicate PCR redescribed previously (Zif actions contained 2.5 ml of 10  buffer for DyNAzyme DNA Polymerase, 0.75 ml of BSA (20 mg ml
1), 1 ml of each primer (0.01 mM), 0.5 ml of PCR Nucleotide Mix (10 mM each), 0.75 ml polymerase (2 U ml
1 DyNAZyme II DNA polymerase 1: 24 Pfu DNA polymerase) and 1 ml of template DNA. Cycling conditions were 94  C for 5 min, 35 cycles of 94  C for 1 min, 62  C for 1 min, and 72  C for 1 min, and a final extension at 72  C for 10 min. Sequencing of amplicons was performed in-house on Illumina MiSeq (2Х250 base pair-end run).

2.3. Sample processing, chemical composition and enzyme assays 2.5. Sequence data processing and analysis In the laboratory, the sawdust material was weighed, followed by freeze-drying and milling using an Ultra Centrifugal Mill ZM 200 (Retsch, Germany), and the resulting fine sawdust was used for subsequent analyses. Dry mass content was measured as a loss of mass during freeze-drying, and the pH was measured in distilled water (1:10). The wood C and N contents were measured in an external laboratory (Research Institute for Soil and Water Consertrovský vation, Prague, Czech Republic) as described previously (Ve and Baldrian, 2015). C was measured using sulfochromic oxidation (ISO 14235), and nitrogen content was estimated by sulphuric acid mineralisation with the addition of selenium and sodium sulphate and conversion to ammonium ions (ISO 11261), which were measured by the segmented flow analyser (SFA) Skalar. Total ergosterol was extracted using 10% KOH in methanol and analysed  by HPLC (Snajdr et al., 2008). Enzyme activity analysis was performed as previously described   and Baldrian, 2011). Freeze-dried samples were extracted (Stursov a at 4  C for 2 h on an orbital shaker (100 rpm), and 12 ml of 50 mM acetate buffer, pH 5, were used for extraction from 250 mg of each sample. Substrates conjugated with fluorescent 4methylumbellyferol were incubated with sample extracts and fluorescence was measured using an excitation wavelength of 355 nm and emission wavelength of 460 nm to infer the activity of b-glucosidase, b-xylosidase, b-galactosidase, cellobiohydrolase (exocellulase), a-glucosidase, N-acetylglucosaminidase, phosphomonoesterase (phosphatase) and esterase (lipase). The activities of endocellulase and endoxylanase were measured using azo-dyed carboxymethylcellulose and xylan (Megazyme, Ireland). 150 ml of the sample and 150 ml of the substrate were mixed and subsequently incubated. The dye released after incubation was quantified at 595 nm (Baldrian, 2009). The activities of laccase and manganese peroxidase were assayed as the oxidation of 2,2-azinobis-3-ethylbenzothiazoline-6sulfonic acid measured at a wavelength of 420 nm, and the oxidative coupling of 3emethyle2ebenzotiazolinon hydrazone and 3,3edimetylaminobenzoic acid was assessed at 595 nm (Baldrian, 2009). 2.4. Extraction and analysis of environmental DNA Total genomic DNA was extracted from 200 mg of material using the NucleoSpin Soil Kit (Macherey-Nagel, Germany) according to

The amplicon sequencing data were processed using the pipetrovský and Baldrian, 2013). Briefly, pair-end line SEED 1.2.1 (Ve reads were merged using fastq-join (Aronesty, 2013). The ITS2 region was extracted using ITS Extractor 1.0.8 (Nilsson et al., 2010) before processing. Chimeric sequences were detected using Usearch 7.0.1090 (Edgar, 2010) and deleted, and sequences were clustered using UPARSE implemented within Usearch (Edgar, 2013) at a 97% similarity level. Consensus sequences were constructed for each cluster, and the closest hits at the species level were identified using BLASTn against UNITE (Koljalg et al., 2013) and GenBank. Where the best hit showed lower similarity than 97% with 95% coverage, the best genus-level hit was identified. The species-level analyses were performed on a dataset where OTUs belonging to the same species were combined and all other OTUs were combined into the genus of the best hit and designated “sp.”. Sequences identified as nonfungal were discarded. Sequence data have been deposited in the MG RAST public database ((Meyer et al., 2008), data set number 270336). To assign putative ecological functions to the fungal OTUs, the fungal genera of the best hit were classified into ecophysiological categories (e.g., white-rot, brown-rot, saprotroph, arbuscular mycorrhiza, ectomycorrhiza) based on the published literature. The definition of categories was the same as in (Tedersoo et al., 2014). Fungal OTUs not assigned to a genus with known ecophysiology and those assigned to genera with unclear ecophysiology remained unassigned. 2.6. Statistics Diversity estimates (Shannon Wiener Index, OTU richness, Evenness) were calculated for a dataset containing 1959 randomly trovský and selected sequences from each sample in SEED 1.2.1. (Ve Baldrian, 2013). Statistica 7 (Statsoft, USA) or PAST 3.03 (http://folk. uio.no/ohammer/past/) were used for statistical analysis. BrayCurtis distance was used as a metric of fungal community similarity between samples. Differences in other variables were tested using the Mann-Whitney U test, which assumes the measurements on a rank-order scale but does not assume the normality of the data. Mantel tests with 99,999 permutations were used to examine the correlations between data matrices, Bray-Curtis similarities were used as a measure of similarity in the fungal community

112

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

composition and Euclidean distances were used for all other variables. Two-dimensional non-metric multidimensional scaling (NMDS) ordination analysis on Bray-Curtis distances was performed in R with package vegan (Oksanen et al., 2013; R Core Team, 2013). Variables were fitted to the ordination diagram as vectors with 999 permutations and included pH, C and N content, ergosterol content, time of decay and relative abundances of sequences of fungi belonging to various ecological classes. Multilevel pattern analysis, as implemented in the indicspecies package of R (De Caceres and Legendre, 2009), was used to identify indicator species, i.e., species significantly associated with certain treatment. In all cases, differences at P < 0.05 were considered statistically significant. 3. Results 3.1. The chemistry of decomposing wood and the fungal biomass content The pH values ranged from 3.2 to 6.0 in the coarse woody debris analysed. The highest mean value of approximately 5.1 was recorded for young FS CWD, whereas old AA CWD showed the lowest mean pH of approximately 3.6. Within each CWD species, the pH decreased with increasing decay length (Fig. 1). The nitrogen content was highly variable, spanning 0.1e1.8 mg g
1, with the highest mean of 0.77 mg g
1 in FS > 38 y. In F. sylvatica, the mean values increased with increasing decay length, and in P. abies, older CWD showed a higher N content than young ones. In A. alba, N content did not show clear temporal development, except for low values at 5e15 y (Fig. 1). For all CWD species, the mean observed decay stages significantly increased from CWD <5 y to those 5e15 y and to older CWD, but the differences in observed decay stages were not significant for CWD 16e38 and > 38 y. Fungal biomass content, expressed as ergosterol content, ranged from 4 to 340 mg g
1. Typically, CWD decaying for 5e15 y showed the lowest ergosterol content, whereas the ergosterol content in

older CWD was higher (Fig. 1). Ergosterol content was significantly correlated with CWD N content (R2 ¼ 0.316, p < 0.00001). 3.2. Fungal community associated with coarse woody debris Highly diverse community of fungi was associated with the CWD, but there was neither a difference in diversity among CWD species nor a clear trend in diversity with increasing decay length (Fig. 1). Two-way PERMANOVA revealed that the CWD species, length of decay and their interaction all significantly affected fungal community composition (P < 0.0001 in all cases). Based on the NMDS, however, the effect of time seemed to be more obvious than that of the CWD species and tended to group samples along the x axis. Among environmental variables, most significant NMDS ordination was found for N, pH, and decay length (Fig. 2A). According to Mantel tests, tree species, decay length, N content and pH were each significant predictors of fungal community composition (P < 0.00001, R > 0.098). The combinations of pH and N content (P < 0.00001, R ¼ 0.1715) and CWD species and decay length (P < 0.00001, R ¼ 0.1435) were the best composite predictors of the fungal community structure. Within CWD species, the composition of fungal communities reflected the length of decay (Fig. 3), with significant effects in all three CWD species. Typically, most fungal sequences retrieved from CWD belonged to Basidiomycota (67.8%) and Ascomycota (30.4%), followed by Mortierellomycotina (1.28%) and Mucoromycotina (0.45%). Sequences of Chytridiomycota, Glomeromycota, Entomphthoromycota and Kicxellomycotina were also retrieved, but these sequences comprised less than 0.05% and did not exceed 0.1% of sequences in any CWD type. The predominance of Basidiomycota was recorded in AA >38 y and PA >38 y, representing 86% and 79% of all sequences, and Ascomycota were frequently observed in FS < 5 y (45%), FS 5e15 y (41%) and PA 16e38 y (42%). Among all sequences, 75% were identified to species level, whereas the other sequences were grouped into genera based on best hits (typically >94% sequence similarity). At this level, the

 Fig. 1. Chemistry of the coarse woody debris of Fagus sylvatica, Abies alba and Picea abies in the Zofín natural forest, fungal biomass content and diversity of fungal communities associated with the CWD. Different letters indicate significant differences between the classes of decay length within each CWD tree species (Mann-Whitney U test, P < 0.05).

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

113

 Fig. 2. Two-dimensional NMDS of fungal communities in the coarse woody debris in the Zofín natural forest. The dataset contained all fungal species with relative sequence abundances of 10% in at least one CWD and those with abundances of >0.5% in at least three CWD units. (A) individual samples (CWD) and environmental and ecological variables. (B) the contributions of individual species; the sizes of circles indicate mean relative sequence abundance in the entire dataset. Results of two-dimensional NMDS ordination are indicated with significant values in bold.

composition of fungal communities was highly diverse. As many as 115 fungal taxa represented more than 10% of all sequences at least in one sample, and 108 additional taxa were recorded with an abundance of >0.5% in at least three CWDs (Supplementary Table 1). Due to this high diversity, the relative abundance of dominant fungal species in the entire dataset was relatively low: the most abundant species was Ganoderma applanatum, with 5.3% of all sequences, followed by Hyphodontia aspera (5.2%), Resinicium furfuraceum (4.7%), Megacollybia marginata (3.6%) and Fomitopsis pinicola (3.1%; Supplementary Table 1). Among the 11 most

abundant species with sequence abundances >1%, nine species were white-rot fungi, F. pinicola was a brown-rot fungus, and Kretzschmeria deusta (1.7%) was a xylariaceous wood decayer. Even the most abundant taxa, however, were typically absent from most CWD. Tylospora fibrillosa (0.8%) and Tomentella sublilacina (0.45%) were the most abundant ectomycorrhizal fungi, and yeasts of the genus Candida were also frequently recorded (0.6%). The dominant fungal species represented relatively little of the entire dataset, reflecting that fungi are rarely shared among multiple CWD units. Within a CWD unit, the sequence pool was

114

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

 Fig. 3. Composition of fungal communities associated with the coarse woody debris in the Zofín natural forest on the level of fungal genera (A), orders (B) and ecological groups (C). For each of the groups, the data represent means of relative sequence abundances.

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

typically dominated with sequences of one or a very few fungi. The most dominant fungal species in a CWD unit represented between 8.9% and 89.6% of all sequences, with an average of 39.3% of all sequences. In 25 (21%) and 6 CWD units (5%), the dominant fungal species comprised more than 50% and 75% of sequences, respectively. For example, Bjerkandera adusta represented 89.6% of all sequences in a F. sylvatica CWD unit, whereas Resinicium furfuraceum represented 85.0% of all sequences in an A. alba CWD unit, and Megacollybia marginata represented 82.5% of all sequences in a F. sylvatica CWD unit (Table 2). An average of 58.4% of all sequences

115

within a single log were represented by dominant fungal species with a relative abundance >10% (typically 1e5 per CWD). The diversity of the dominant fungal taxa was very high, with as many as 57 species dominating in at least one CWD unit. Among these, 37 species were white-rot fungi, 3 species were brown-rot fungi, 15 species were other saprotrophs (including two yeasts), and 2 were ectomycorrhizal species. Among species that dominated in >2 CWD, all except two were dominant on multiple CWD species; Megacollybia marginata, Hyphodontia aspera and Mycena laevigata were dominant on all three CWD species, indicating the low tree

Table 2  List of fungal species dominating individual coarse woody debris in the Zofín natural forest. The data are based on the relative abundance of ITS2 sequences. Species

Ecology

CWD with dominance1

Fagusa

Abiesa

Piceaa

CWD >10%b

Max. abundance (%)c

Megacollybia marginata Fomitopsis pinicola Resinicium furfuraceum Hyphodontia aspera Resinicium bicolor Ganoderma applanatum Heterobasidion annosum Mycena laevigata Fomes fomentarius Hyphodontia sp. Ischnoderma benzoinum Kretzschmaria deusta Trichaptum abietinum Skeletocutis odora Acephala sp. Hyphodontia alutaria Eutypa spinosa Aphanobasidium sp. Rigidoporus crocatus Vesiculomyces sp. Peniophorella sp. Hyphoderma sp. Lycoperdon pyriforme Bjerkandera adusta Resupinatus applicatus Mycena arcangeliana Porodaedalea chrysoloma Phellinus hartigii Hyphodontia breviseta Mycena vitilis Mycena galericulata Xylaria carpophila Mycena niveipes Ascocoryne cylichnium Phlebia centrifuga Ossicaulis lachnopus Hypholoma fasciculare Antrodia serialis Capronia sp. Hyphoderma guttuliferum Galerina fallax Inocybe napipes Sistotremastrum sp. Trametes versicolor Phlebia sp. Arachnopeziza sp. Phlebia unica Pholiota scamba Candida sp. Mycena leptocephala Trichoderma saturnisporum Pleurotus sapidus Ascocoryne sarcoides Xerocomus badius Hericium americanum Chloridium sp. Trichosporon porosum

White-rot Brown-rot White-rot White-rot White-rot White-rot White-rot White-rot White-rot White-rot White-rot Saprotroph White-rot White-rot Saprotroph White-rot Saprotroph Saprotroph White-rot White-rot White-rot White-rot Saprotroph White-rot Saprotroph White-rot White-rot White-rot White-rot White-rot White-rot Saprotroph White-rot Saprotroph White-rot Brown-rot White-rot Brown-rot Saprotroph White-rot White-rot Ectomycorrhiza White-rot White-rot White-rot Saprotroph White-rot White-rot Yeast White-rot Saprotroph White-rot Saprotroph Ectomycorrhiza White-rot Saprotroph Yeast

7 7 6 6 6 6 6 5 4 3 3 3 2 2 2 2 2 2 2 2 2 2 2 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

4 2

2

1 5 3 1 3

9 14 12 10 10 15 9 6 6 4 3 5 2 2 2 3 2 2 2 2 3 3 2 1 1 1 2 1 2 3 2 1 2 2 3 1 1 1 2 1 1 2 2 3 1 1 1 1 3 1 1 3 3 1 1 1 1

82.5 48.4 85.0 79.8 66.7 66.4 50.4 64.7 68.0 64.0 55.7 49.0 44.5 44.0 41.7 40.8 37.9 33.8 33.8 26.1 23.9 21.8 14.7 89.6 61.0 53.9 48.1 43.2 40.3 39.5 35.4 35.2 33.1 32.0 31.7 30.4 30.3 29.6 29.4 29.0 27.1 26.2 23.9 22.7 22.5 21.5 21.0 21.0 20.1 20.1 19.9 19.5 19.3 18.0 17.5 17.3 16.9

a b c

Number of CWD where the sequences of this taxon were most abundant. Number of CWD where this taxon represented >10% of all sequences. Maximal recorded relative abundance of sequences.

4 3 1 3

3 1 3 3 2 1 2 3

4 3 1 1

3 1

2 1

2 2 2 2 2 2 1 1 2 1

1 1

1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1 1

116

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

species specificity of the dominant fungi (Table 2). The dominant fungi also significantly contributed to differences in the total fungal communities in CWD, as revealed by NMDS (Fig. 2B). Among all fungal sequences, distinct ecology was assigned to 96.5% of them based on their species- or genus-level identification. Most sequences belonged to white-rot fungi (52%), followed by other saprotrophs (32%), ectomycorrhizal fungi (4.7%) and brownrot fungi (3.8%). While white-rot fungi and other saprotrophs were generally present in all samples, brown-rot and ectomycorrhizal fungi-dominated CWD were more rare and showed specific clustering in the NMDS (Figs. 2 and 4). CWD dominated by whiterot fungi was most common, although some CWD was dominated by other saprotrophs (24%), and one CWD unit was dominated by ectomycorrhizal fungi. Facultative yeasts accounted for 3.4% of all sequences (maximum 34% in a single CWD) and yeasts accounted for 1.7% of all sequences (maximum 26%; Fig. 3C). The pattern of fungal ecology distribution among CWD types was indistinct in most cases. Although brown-rot fungi tended to show low abundance in old CWD, the differences in their abundance among CWD age classes were not statistically significant. Ectomycorrhizal sequences tended to be more frequent in CWD after >16 y of decay, but significant differences were only found in F. sylvatica and A. alba between 5 and 15y and >38 y (Fig. 4). Ten indicator taxa were found for F. sylvatica, five for A. alba and eight for P. abies. Moreover, 16 additional indicator species were shared by the two coniferous species, including the highly abundant species Resinicium furfuraceum. F. sylvatica and P. abies shared six indicator taxa, but none of these species were typical wood decomposers (Table 3). Relatively few indicator species were identified for the individual decay lengths of F. sylvatica and A. alba, indicating little common trend in community development over time. Several additional indicator taxa were proposed for P. abies, likely indicating more predictable community development. Most of the indicator taxa, however, were not the typical dominant

white- and brown-rot fungi, and advanced decay stages were rather indicated by the presence of yeasts, Mortierellomycotina and the ectomycorrhizal species Tylospora fibrillosa (Table 3). Generally, there was little overlap between dominant and indicator taxa, and the abundance of most identified indicator taxa was low (Tables 2 and 3, Supplementary Table 1). 3.3. Enzyme activity in coarse woody debris The activity of all twelve enzymes determined in the CWD varied considerably among samples. According to Mantel tests, significant correlations between enzyme activity and fungal community composition (R ¼ 0.1129, P < 0.01), decay length (R ¼ 0.1732, P ¼ 0.00003), ergosterol content (R ¼ 0.2736, P ¼ 0.00002) and, in particular, N content (R ¼ 0.3687, P < 0.00001) were observed. The effects of tree species and pH on enzyme activity were not significant, and N content was also a better predictor of enzyme activity than any combination of variables. Considering individual enzymes, the activities of endocellulase, b-glucosidase, phosphomonoesterase and esterase were detected in >95% of CWD, whereas the activities of b-galactosidase, b-xylosidase, N-acetylglucosaminidase, and laccase were present in >85% of CWD. The activities of other enzymes, such as a-glucosidase (80% of CWDs) and Mn-peroxidase (77% of CWDs), were less frequently recorded. The activity of individual enzymes showed significant positive correlations with ergosterol content (all enzymes except Mn-peroxidase and laccase), followed by N content (all except Mnperoxidase, laccase and endocellulase); the activity of the latter two enzymes was positively correlated with wood pH (P < 0.05). Not surprisingly, the strongest correlation (R ¼ 0.5948) was observed between ergosterol content and the activity of N-acetylglucosaminidase, which participates in chitin decomposition. Within individual CWD types, the activities of certain enzymes showed significant differences among decay length classes. This

 Fig. 4. Ecological groups of fungi within individual coarse woody debris in the Zofín natural forest. The relative abundance of fungal sequences belonging to each ecological group was plotted at the position of the corresponding sample in NMDS. The size of the circle indicates relative abundance (the smallest dots indicate zero).

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

117

Table 3  Indicator taxa of coarse woody debris in the Zofín natural forest. Indicator species discriminating between CWD tree species and between length of decay within a single tree species; * - P < 0.05,** - P < 0.01,*** - P < 0.001. Fagus sylvatica

Fagus sylvatica <5 y

Candida sp. Hypocrea sp. Pluteus cervinus Zignoella pulviscula

* *** *** ***

Aureobasidium pullulans Ramularia endophylla

Eutypa sp. Trichoderma sp. Jobellisia sp. Eutypa spinosa

** ** *** *

Ascocoryne sarcoides Lophiostoma sp.

Pseudoclathrosphaerina sp. Tubaria sp.

* *

Sugiyamaella sp. Anomoloma sp.

Abies alba

Picea abies <5 y ** *

Orbilia sp. Stereum sanguinolentum Entomocorticium dendroctoni Leucostoma kunzei

** ** ** **

* *

Phanerochaete sordida Skeletocutis odora

* *

Fagus sylvatica 5e15 y

Fagus sylvatica 16e38 y

Picea abies 5e15 y * *

Fagus sylvatica > 38 y

Hericium americanum Gymnopilus sp.

*** *

Pseudoclathrosphaerina sp. Peniophorella sp.

Aphanobasidium sp. Vesiculomyces sp.

* *

Fagus sylvatica > 15 y

Cortinarius sp.

*

Zopfiella sp. Eutypa sp. Hyaloscypha albohyalina

* * **

Arachnopeziza variepilosa Zalerion arboricola

*** ***

Ceramothyrium sp.

*

Melanchlenus eumetabolus Phaeomoniella sp. Phlebia sp.

*** * *

Phialocephala scopiformis Entomocorticium dendroctoni Amylostereum areolatum

*** * *

Hyphodontia aspera Capronia pilosella Eutypa sp. Cordana sp.

Lachnum sp.

*

Abies alba 5e15 y

Sebacina sp. Typhula intermedia Calocera sp. Mycena leptocephala Hyaloscypha albohyalina

** * ** ** *

Ascocoryne cylichnium Hericium americanum Helicoon sp. Ischnoderma benzoinum

Picea abies

* *

Picea abies 16e38 y Botryobasidium sp. Pseudaegerita sp.

** **

Kretzschmaria deusta Xeromphalina campanella Galerina fallax

* * *

Tylospora fibrillosa Abies alba < 5 y

Sugiyamaella sp.

** ** * *

Helicodendron websteri Phaeomoniella sp. Naevala minutissima Hypoxylon fragiforme Zalerion arboricola

*** *** ** * ***

Melanchlenus eumetabolus

**

Phialocephala scopiformis Ascocoryne sarcoides Hyaloscypha aureliella Helicoon sp.

*** ** * *

Picea abies > 15 y *

*** ** ** * ** ** ** ** *** * * * * * ** *

effect was observed in F. sylvatica for b-xylosidase, esterase and phosphomonoesterase, which were highest after 16e38 y, and bgalactosidase and N-acetylglucosaminidase, which were high after 5e38 y. For A. alba, the enzyme activities were typically different, showing low activity between 5 and 15 y and high activity between 16 and 38 y. In P. abies, the activity of multiple enzymes was high after 16e38 and > 38 y of decay (Fig. 6).

**

Picea abies < 16 y * ** ** *

Abies alba > 15 y

Hyphodiscus sp. Hyaloscypha sp. Heterobasidion annosum Hyphodontia sp. Ceramothyrium sp. Pseudaegerita sp. Resinicium furfuraceum Flagellospora sp. Sugiyamaella paludigena Heterodermia sp. Mortierella humilis Mycena niveipes Anomoloma sp. Helicoon sp. Hyaloscypha aureliella Mycena silvae nigrae

** ** * *

Picea abies > 38 y

Fagus sylvatica þ Picea abies

Abies alba þ Picea abies

Elaphomyces muricatus Oidiodendron chlamydosporicum Xylaria carpophila Amphinema byssoides

Arachnopeziza sp.

***

Mortierella gemmifera Mortierella pulchella Ceramothyrium sp. Mortierella humilis Coniochaeta sp. Troposporella sp. Meliniomyces variabilis Cryptococcus terricola Botryobasidium subcoronatum Metulocladosporiella sp. Anomoloma sp. Mortierella turficola

*** ** *** ** ** ** *** * * ** ** *

4. Discussion Although high variability in deadwood chemistry was observed, there was a general trend of increasing N content and decreasing pH during decomposition. Of these two factors, N content appeared to be more important for deadwood fungi, with a significant influence on the fungal community composition, biomass content and enzyme activity. Contrary to expectations and in contrast to leaf litter, where

118

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

 Fig. 5. Relative abundance of fungal ecological groups in coarse woody debris of Fagus sylvatica, Abies alba and Picea abies in the Zofín natural forest. Each circle indicates one sample and circle areas correspond to relative abundances of sequences of given group.

fungal biomass reaches maximal values during early decomposition when the availability of easily decomposable substrates is high   et al., 2014), the fungal biomass in (Snajdr et al., 2011a; Urbanova deadwood was highest during the late stages of decomposition (Fig. 1). Deadwood colonization was likely not the limiting factor because high fungal biomass was observed in multiple samples of the youngest CWD. This supports the assumption that low N represents a more important limitation for fungal growth than the recalcitrance of wood biopolymers. The fungal biomass in deadwood was considerable: if we assume that fungal biomass contains 3.8 mg g
1 ergosterol (Baldrian et al., 2013), then the CWD in the present study would contain, on average, 1.8% fungal biomass and up to 5e9% fungal biomass in the most biomass-rich CWDs. These values were comparable to the estimates of Fomes fomentarius biomass in deadwood near a fruit body containing 5e15% fungal trovský et al., 2011). Considering that the dead wood biomass (Ve density was approximately 0.253 t m
3, the estimated mean for decomposing P. abies logs (Teodosiu and Bouriaud, 2012), and the estimated 208 m3 ha
1 deadwood content for the studied  ecosystem (Kr al et al., 2010b; Samonil et al., 2013), the total fungal biomass was estimated as 0.95 t ha
1 in 52 t ha
1 dead wood. This finding is comparable to the amount of fungal biomass in the soil or litter of temperate and boreal forests (Baldrian et al., 2013; Ekblad et al., 2013). However, the fungal biomass in dead wood is much more concentrated and distributed largely unevenly across the ecosystem. In the present study, we have tried to describe the fungal community composition in dead wood, which is a challenging task. Fungal sequence abundance may be an improper proxy of species abundance due to multiple reasons, such as the variation of rDNA trovský copy numbers (Baldrian et al., 2013; Ihrmark et al., 2012; Ve et al., 2016). However, fungi found as abundant on P. abies wood typically also produced large fruit bodies which demonstrated the suitability of DNA-based quantification at least for important fruit

body-forming decomposers. (Ovaskainen et al., 2013). It has been proposed that RNA and not DNA might be the most suitable marker of active fungi on dead wood due to the potential persistence of DNA in dead wood (Rajala et al., 2011). Considering the turnover of fungal DNA in decomposing plant litter and soil, which was found   et al., 2014; Vorískova  and Baldrian, to be rather short (Stursov a 2013), it seems that DNA is sufficient to provide information about community development at time frames of multiple years in the highly metabolically active substrates of decomposing wood. Both combinations of pH and N, and of decay length and CWD tree species, significantly affected fungal community composition in the present study; however, it was impossible to determine which of these factors was more important, because these factors were also significantly correlated. That initial assumption was that CWD species represent one of the decisive factors in the fungal community development, as past analyses of fungal fruit bodies indicated several tree-species specific fungi across European forests (Kuffer et al., 2008). However, the present study showed that these results are likely context dependent: Amphinema byssoides specific for A. alba according to the above mentioned study was detected on P. abies and F. sylvatica but not A. abies CWD in the present study. Additionally, the distinction between coniferous and angiospermassociated fungi (Kuffer et al., 2008) was not observed in the present study. In contrast to the high level of association of fungal  communities with specific tree taxa in litter and soil (Urbanova et al., 2015), the differences in wood chemistry had much less effect on the fungal community, as clearly indicated by the fact that several fungal species became dominant on the dead wood of multiple tree species (Table 2). In the past, classification of individual fungal species into early, mid and late decomposers typically applied and seemed to be backed by fruit body survey data (Jonsson et al., 2008). Here, associations of fungal wood decomposers, forming large fruit bodies, with certain decay stages are questioned because most of these

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

119

 Fig. 6. Distribution of extracellular enzyme activities in coarse woody debris of Fagus sylvatica, Abies alba and Picea abies in the Zofín natural forest. Different letters indicate significant differences between the classes of decay length within each CWD tree species (Mann-Whitney U test, P < 0.05).

fungi occurred on dead wood of various ages (Supplementary Table 1) and did not represent suitable indicators of decomposition length (Table 3). This was despite the fact that the observed decay stages corresponded with decomposition length, except for the two oldest CWD groups. Apparently, there is not such a strict association between species composition and succession stage, as a   et al., 2015; observed in the rapidly decaying tree litter (Han ckova  and Baldrian, 2013). Vorískova Observations of decaying logs of P. abies in boreal forests indi et al., 2012; cated increase of diversity with decay stage (Kubartova Rajala et al., 2012; Rajala et al., 2015) but this was not confirmed here. Moreover, considering that the bulk of the fungal community

in a decomposing CWD includes highly abundant taxa (Table 2), the importance of species diversity for the fate of dead wood seems to be minor. In previous surveys of unmanaged boreal P. abies forests, wood density was considered a proxy of decomposition length. The C/N ratio decreased and the lignin and moisture content increased during succession, with high abundances of ascomycetes in early stages, brown-rot fungi in the intermediate stages and ECM fungi in the late stages of decomposition in CWD with high N content (Rajala et al., 2012; Rajala et al., 2015). The results of the present study support the observation of increasing ECM share with time, although high ECM abundance was also recorded in individual

120

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

CWD units of the youngest class (Fig. 5). ECM fungi may establish in dead wood early and independently on the decay stage, as indicated by the observations of their presence in P. abies logs across all  et al., 2012). Brown-rot fungi were rarely decay stages (Kubartova observed in the present study, and the occurrence of these fungi showed a rather random pattern, likely reflecting the stochastic colonization of the CWD. The dead mycelium of fungi may represent an important substrate in dead wood, and when well defined, myceliumdecomposing fungi would represent an interesting ecological guild to be followed on dead wood. Recently, it was reported that dead mycelia in forest soil are decomposed by specific fungal taxa  et al., 2016). Among the genera found to be associated (Brabcova with this decomposing fungal biomass, Mortierella and Mucor were the most frequent in CWD, representing 1.28% and 0.28% of fungal sequences in this study, respectively. It is interesting to note that several Mortierella species proved to be indicator species of the late decay of P. abies CWD (Table 3). In total, the genera involved in  et al. (2016) mycelium decomposition in soil according to Brabcova made up 2.27% of all fungal sequences in the present study. Dead or even living mycelia apparently represent important resources for many wood decomposers and other saprotrophs that readily utilize  them (Boddy, 2000; Lindahl and Finlay, 2006; Snajdr et al., 2011b). There was an extremely wide variety of fungi in the beechdominated natural forest that dominated CWD decomposition, and the dominance of each species was restricted to few CWD units (Table 2). Thus, the general patterns of community composition were impossible to propose, resulting in a high variability in fungal community composition among individual CWD units (Fig. 2A), which was also observed in other recent molecular surveys  et al., 2015; Ottosson et al., 2015; van der Wal et al., (Kubartova 2015). Together with the observation and experimental evidence that certain taxa tend to be supported or excluded from CWD precolonized by specific fungi (Hiscox et al., 2015; Ottosson et al., 2014), it seems that the initial establishment, which is probably highly stochastic, may set each individual log on a highly specific track of development in time. This may be further complicated by the fact that individual logs most likely do not develop under the influence of one community but rather hosts patches of fungi associated with various microhabitats within its volume (Boddy,  et al., 2012) that may behave more or less inde2000; Kubartova pendently. When drilling dead wood, it was often apparent by the colour of the sawdust that the wood properties varied within a single CWD unit. The diversity of dead wood development scenarios under natural conditions is likely even more diverse than reported here because the fate of dead wood also reflects the mode of stem breakage, and the conditions in trees decomposing as standing snags (that were intentionally excluded from our study) are also distinct and lead to the development of a specific fungal community (Ottosson et al., 2015). Moreover, although CWD represents the bulk of all the dead wood in the ecosystem, studies of CWD do not provide sufficient information on the diversity of wood-associated fungi because many species are restricted to fine woody debris, €ssler where the community assembly rules are likely different (Ba et al., 2010; Juutilainen et al., 2011). Recently, CO2 emissions were demonstrated to be tree-species dependent in the initial decomposition phase (Kahl et al., 2015) indicating various rates of initial wood decomposition among trees. In the present study, the enzyme activities in dead wood exhibited high variation but did not significantly differ among tree species. The fact that enzyme activity was most closely related to the N content indicates that nutrient availability is the most likely driver of enzyme production. A strong correlation between ergosterol content and enzyme activity supports the generally accepted view

that the enzymes decomposing dead wood are largely of fungal origin. Different nutritional guilds of fungi are known to be distinct in their enzyme production. In an experiment with F. sylvatica wood, rates of lignin removal and production of ligninolytic enzymes differed after inoculation among (1) classic white-rot fungi (e. g., Phlebia radiata), (2) nonspecific wood-rotters (e. g., Agrocybe aegerita), (3) white-rotters of leaf litter (e.g., Stropharia rugosoannulata) or (4) soft-rotters of wood (e. g., Xylaria polymorpha; Liers et al. (2011)). Specifically, Mn-oxidizing peroxidases were produced by the first three groups which resulted in lignin depolymerization (Liers et al., 2011). Profound differences in enzyme production were also reported between brown-rot and white-rot fungi, with the latter being the exclusive producers of ligninolytic  et al., 2015). peroxidases and laccase (Eichlerova Unfortunately, measurements of enzyme activities in dead wood samples from the field are so far rare. In a study that compared enzyme production in dead wood close to Fomes fomentarius fruit bodies, multiple hydrolytic and oxidative enzymes were recorded including cellulases, hemicellulases, laccase and Mn-peroxidase and there was a significant difference in the activity of certain enzymes between the wood of F. sylvatica and Betula trovský et al., 2011). Recently, pendula colonized by the fungus (Ve oxidative enzymes, laccases and peroxidases, were surveyed in the dead wood of P. abies, F. sylvatica and Pinus sylvestris (Arnstadt et al., 2016). The pH and lignin fragment content in wood seemed to be the best predictors of enzyme activity; however, only a limited set of wood properties was examined. In the present study, no difference in enzyme activities was recorded for CWD dominated by brown-rot versus white-rot fungi in P. abies, where a sufficient number of samples (6 and 29, respectively) were available. Neither the total activity, nor the relative contribution of individual enzymes differed significantly. This may indicate that the drivers of enzyme production are complex, and the presence/absence of fungi with specific ecology are not sufficient to predict it. Recent results indicate that the ability to decompose cellulose and other polysaccharides is also common in bacteria, for example in plant litter pez-Monde jar et al., 2016). Bacteria thus likely also contribute to (Lo enzyme production in dead wood, but the extent of this contribution is unclear. This work indicates that coarse dead wood represents a highly diverse substrate in terms of quality, fungal biomass content and especially fungal community composition, whose properties change with time. The fact that individual logs are typically dominated by one or a very few fungal species that are rarely tree species-specific indicates stochastic assembly as a probable cause of this diversity. Among the estimated variables, N content decreasing with decay length seemed to be the most important candidate driver of fungal biomass content, community composition and enzyme activity. Acknowledgements This work was supported by the Czech Science Foundation (1327454S) and the research concept of the Institute of Microbiology of the CAS (RVO61388971). Appendix A. Supplementary data Supplementary data related to this article can be found at http:// dx.doi.org/10.1016/j.funeco.2016.07.001. References Anderson-Teixeira, K.J., Davies, S.J., Bennett, A.C., Gonzalez-Akre, E.B., Muller-

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122 Landau, H.C., Wright, S.J., et al., 2015. CTFS-ForestGEO: a worldwide network monitoring forests in an era of global change. Global Change Biol. 21, 528e549. €ssler, C., et al., 2016. PatArnstadt, T., Hoppe, B., Kahl, T., Kellner, H., Krüger, D., Ba terns of laccase and peroxidases in coarse woody debris of Fagus sylvatica, Picea abies and Pinus sylvestris and their relation to different wood parameters. Eur. J. For. Res. 35, 109e124. Aronesty, E., 2013. Comparison of sequencing utility programs. Open Bioinf. J. 7, 1e8. Baldrian, P., 2008. Wood-inhabiting ligninolytic basidiomycetes in soils: ecology and constraints for applicability in bioremediation. Fungal Ecol. 1, 4e12. Baldrian, P., 2009. Microbial enzyme-catalyzed processes in soils and their analysis. Plant Soil Environ. 55, 370e378.  trovský, T., Cajthaml, T., Dobia sov nkova , M., Snajdr, Baldrian, P., Ve a, P., Petra J., et al., 2013. Estimation of fungal biomass in forest litter and soil. Fungal Ecol. 6, 1e11. €ssler, C., M€ Ba aller, J., Dziock, F., Brandl, R., 2010. Effects of resource availability and climate on the diversity of wood-decaying fungi. J. Ecol. 98, 822e832. Boddy, L., 2000. Interspecific combative interactions between wood-decaying basidiomycetes. FEMS Microbiol. Ecol. 31, 185e194. , V., Nov , M., Davidova , A., Baldrian, P., 2016. Dead fungal mycelium in Brabcova akova forest soil represents a decomposition hotspot and a habitat for a specific microbial community. New Phytol. 210, 1369e1381. De Caceres, M., Legendre, P., 2009. Associations between species and groups of sites: indices and statistical inference. Ecology 90, 3566e3574. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460e2461. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods 10, 996e998.   , I., Homolka, L., Zif , P., Baldrian, P., 2015. Eichlerova c akov a, L., Lis a, L., Dobi asova Enzymatic systems involved in decomposition reflects the ecology and taxonomy of saprotrophic fungi. Fungal Ecol. 13, 10e22. Ekblad, A., Wallander, H., Godbold, D.L., Cruz, C., Johnson, D., Baldrian, P., et al., 2013. The production and turnover of extramatrical mycelium of ectomycorrhizal fungi in forest soils: role in carbon cycling. Plant Soil 366, 1e27. Fukami, T., Dickie, I.A., Wilkie, J.P., Paulus, B.C., Park, D., Roberts, A., et al., 2010. Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. Ecol. Lett. 13, 675e684.    Han a ckov a, Z., Koukol, O., Stursov a, M., Kolarík, M., Baldrian, P., 2015. Fungal succession in the needle litter of a montane Picea abies forest investigated through strain isolation and molecular fingerprinting. Fungal Ecol. 13, 157e166. Hiscox, J., Savoury, M., Muller, C.T., Lindahl, B.D., Rogers, H.J., Boddy, L., 2015. Priority effects during fungal community establishment in beech wood. ISME J. 9, 2246e2260. Hoppe, B., Krüger, D., Kahl, T., Arnstadt, T., Buscat, F., Bauhus, J., et al., 2015. A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies. Sci. Rep. 5, 9456. Ihrmark, K., Bodeker, I.T.M., Cruz-Martinez, K., Friberg, H., Kubartova, A., Schenck, J., et al., 2012. New primers to amplify the fungal ITS2 region - evaluation by 454sequencing of artificial and natural communities. FEMS Microbiol. Ecol. 82, 666e677. Jonsson, M.T., Edman, M., Jonsson, B.G., 2008. Colonization and extinction patterns of wood-decaying fungi in a boreal old-growth Picea abies forest. J. Ecol. 96, 1065e1075. Juutilainen, K., Halme, P., Kotiranta, H., Monkkonen, M., 2011. Size matters in studies of dead wood and wood-inhabiting fungi. Fungal Ecol. 4, 342e349. Kahl, T., Baber, K., Otto, P., Wirth, C., Bauhus, J., 2015. Drivers of CO2 emission rates from dead wood logs of 13 tree species in the initial decomposition phase. Forests 6, 2484e2504. Koljalg, U., Nilsson, R.H., Abarenkov, K., Tedersoo, L., Taylor, A.F.S., Bahram, M., et al., 2013. Towards a unified paradigm for sequence-based identification of fungi. Mol. Ecol. 22, 5271e5277. l, K., Janík, D., Vrska, T., Adam, D., Hort, L., Unar, P., et al., 2010a. Local variability Kra of stand structural features in beech dominated natural forests of Central Europe: implications for sampling. For. Ecol. Manag. 260, 2196e2203.  l, K., Vrska, T., Hort, L., Adam, D., Samonil, Kra P., 2010b. Developmental phases in a temperate natural spruce-fir-beech forest: determination by a supervised classification method. Eur. J. For. Res. 129, 339e351.  l, K., Valtera, M., Janík, D., Samonil, Kra P., Vrska, T., 2014. Spatial variability of general stand characteristics in central European beech-dominated natural stands - effects of scale. For. Ecol. Manag. 328, 353e364. Kubartov a, A., Ottosson, E., Dahlberg, A., Stenlid, J., 2012. Patterns of fungal communities among and within decaying logs, revealed by 454 sequencing. Mol. Ecol. 21, 4514e4532. Kubartov a, A., Ottosson, E., Stenlid, J., 2015. Linking fungal communities to wood density loss after 12 years of log decay. FEMS Microbiol. Ecol. 91, fiv032. Kuffer, N., Gillet, F., Senn-Irlet, B., Aragno, M., Job, D., 2008. Ecological determinants of fungal diversity on deadwood in European forests. Fungal Divers 30, 83e95. Liers, C., Arnstadt, T., Ullrich, R., Hofrichter, M., 2011. Patterns of lignin degradation and oxidative enzyme secretion by different wood- and litter-colonizing basidiomycetes and ascomycetes grown on beech-wood. FEMS Microbiol. Ecol. 78, 91e102. Lindahl, B.D., Finlay, R.D., 2006. Activities of chitinolytic enzymes during primary and secondary colonization of wood by basidiomycetous fungi. New Phytol. 169, 389e397. Lindner, D.L., Vasaitis, R., Kubartova, A., Allmer, J., Johannesson, H., Banik, M.T., et al.,

121

2011. Initial fungal colonizer affects mass loss and fungal community development in Picea abies logs 6 yr after inoculation. Fungal Ecol. 4, 449e460. pez-Monde jar, R., Zühlke, D., Becher, D., Riedel, K., Baldrian, P., 2016. Cellulose Lo and hemicellulose decomposition by forest soil bacteria proceeds by the action of structurally variable enzymatic systems. Sci. Rep. 6, 25279. Meyer, F., Paarmann, D., D'Souza, M., Olson, R., Glass, E.M., Kubal, M., et al., 2008. The metagenomics RAST server - a public resource for the automatic phylogenetic and functional analysis of metagenomes. BMC Bioinforma. 9, 386. Nilsson, R.H., Veldre, V., Hartmann, M., Unterseher, M., Amend, A., Bergsten, J., et al., 2010. An open source software package for automated extraction of ITS1 and ITS2 from fungal ITS sequences for use in high-throughput community assays and molecular ecology. Fungal Ecol. 3, 284e287. Oksanen, J., Blanchet, F.G., Kindt, R., Legendre, P., Minchin, P.R., O'Hara, R.B., Simpson, G.L., Solymos, P., Henry, M., Stevens, H., Wagner, H., 2013. Vegan: Community Ecology Package. Version 2.0e7. n, J., Dahlberg, A., Edman, M., Jonsson, M., Larsson, K.H., et al., Ottosson, E., Norde 2014. Species associations during the succession of wood-inhabiting fungal communities. Fungal Ecol. 11, 17e28. Ottosson, E., Kubartova, A., Edman, M., Jonsson, M., Lindhe, A., Stenlid, J., et al., 2015. Diverse ecological roles within fungal communities in decomposing logs of Picea abies. FEMS Microbiol. Ecol. 91, fiv012. n, B., et al., Ovaskainen, O., Schigel, D., Ali-Kovero, H., Auvinen, P., Paulin, L., Norde 2013. Combining high-throughput sequencing with fruit body surveys reveals contrasting life-history strategies in fungi. ISME J. 7, 1696e1709. R Core Team, 2013. R: A Language and Environment for Statistical Computing. Version 3.0.0. Rajala, T., Peltoniemi, M., Hantula, J., Makipaa, R., Pennanen, T., 2011. RNA reveals a succession of active fungi during the decay of Norway spruce logs. Fungal Ecol. 4, 437e448. €a €, R., 2012. Fungal community dyRajala, T., Peltoniemi, M., Pennanen, T., M€ akipa namics in relation to substrate quality of decaying Norway spruce (Picea abies L. Karst.) logs in boreal forests. FEMS Microbiol. Ecol. 81, 494e505. €kipa €€ Rajala, T., Tuomivirta, T., Pennanen, T., Ma a, R., 2015. Habitat models of woodinhabiting fungi along a decay gradient of Norway spruce logs. Fungal Ecol. 18, 48e55. Rayner, A.D.M., Boddy, L., 1988. Fungal Decomposition of Wood: its Biology and Ecology. Wiley, Chichester ; New York. €ssler, C., Brandl, R., Gossner, M.M., Thorn, S., Ulyshen, M.D., et al., 2015. Seibold, S., Ba Experimental studies of dead-wood biodiversity d A review identifying global gaps in knowledge. Biol. Conserv. 191, 139e149. Stokland, J.N., Siitonen, J., Jonsson, B.G., 2012. Biodiversity in Dead Wood. Cambridge Univ Press, Cambridge.  , P., Vasí l, K., et al., 2013. Samonil, P., Dole zelova ckov a, I., Adam, D., Valtera, M., Kra Individual-based approach to the detection of disturbance history through spatial scales in a natural beech-dominated forest. J. Veg. Sci. 24, 1167e1184.  skova , V., Merhautova , V., Herinkova , J., Cajthaml, T., Baldrian, P., 2008. Snajdr, J., Vala Spatial variability of enzyme activities and microbial biomass in the upper layers of Quercus petraea forest soil. Soil Biol. Biochem. 40, 2068e2075.  skova , V., Merhautova , V., Petra nkov Snajdr, J., Cajthaml, T., Vala a, M., Spetz, P., et al., 2011a. Transformation of Quercus petraea litter: successive changes in litter chemistry are reflected in differential enzyme activity and changes in the microbial community composition. FEMS Microbiol. Ecol. 75, 291e303.  sova , P., Ve trovský, T., Vala skova , V., Alawi, A., Boddy, L., et al., 2011b. Snajdr, J., Dobia Saprotrophic basidiomycete mycelia and their interspecific interactions affect the spatial distribution of extracellular enzymes in soil. FEMS Microbiol. Ecol. 78, 80e90.  , M., Baldrian, P., 2011. Effects of soil properties and management on the Stursova activity of soil organic matter transforming enzymes and the quantification of soil-bound and free activity. Plant Soil 338, 99e110.    , M., Snajdr,
, H., Baldrian, P., 2014. Stursov a J., Cajthaml, T., B arta, J., Santr u ckova When the forest dies: the response of forest soil fungi to a bark beetle-induced tree dieback. ISME J. 8, 1920e1931. Tedersoo, L., Bahram, M., Polme, S., Koljalg, U., Yorou, N.S., Wijesundera, R., et al., 2014. Global diversity and geography of soil fungi. Science 346, 1078-þ. Teodosiu, M., Bouriaud, O.B., 2012. Deadwood specific density and its influential factors: a case study from a pure Norway spruce old-growth forest in the Eastern Carpathians. For. Ecol. Manag. 283, 77e85.  , V., Merhautova , V., Dobi , P., Cajthaml, T., Urbanov a, M., Snajdr, J., Brabcova asova et al., 2014. Litter decomposition along a primary post-mining chronosequence. Biol. Fertil. Soils 50, 827e837.  Urbanov a, M., Snajdr, J., Baldrian, P., 2015. Composition of fungal and bacterial communities in forest litter and soil is largely determined by dominant trees. Soil Biol. Biochem. 84, 53e64. skova , V., de Boer, W., Gunnewiek, P.J.A.K., Pospísek, M., Baldrian, P., 2009. Vala Phylogenetic composition and properties of bacteria coexisting with the fungus Hypholoma fasciculare in decaying wood. ISME J. 3, 1218e1221. van der Wal, A., Ottosson, E., de Boer, W., 2015. Neglected role of fungal community composition in explaining variation in wood decay rates. Ecology 96, 124e133.  trovský, T., Vorískova , J., Snajdr, Ve J., Gabriel, J., Baldrian, P., 2011. Ecology of coarse wood decomposition by the saprotrophic fungus. Fomes Fomentarius. Biodegrad. 22, 709e718. trovský, T., Baldrian, P., 2013. Analysis of soil fungal communities by amplicon Ve pyrosequencing: current approaches to data analysis and the introduction of the pipeline SEED. Biol. Fertil. Soils 49, 1027e1037.

122

P. Baldrian et al. / Fungal Ecology 23 (2016) 109e122

trovský, T., Baldrian, P., 2015. An in-depth analysis of actinobacterial commuVe nities shows their high diversity in grassland soils along a gradient of mixed heavy metal contamination. Biol. Fertil. Soils 51, 827e837.   trovský, T., Kolarík, M., Zif Ve c akov a, L., Zelenka, T., Baldrian, P., 2016. The rpb2 gene represents a viable alternative molecular marker for the analysis of environmental fungal communities. Mol. Ecol. Res. 16, 388e401.

, J., Baldrian, P., 2013. Fungal community on decomposing leaf litter unVorískova dergoes rapid successional changes. ISME J. 7, 477e486.   kova , L., Ve trovský, T., Howe, A., Baldrian, P., 2016. Microbial activity in forest Zif ca soil reflects the changes in ecosystem properties between summer and winter. Environ. Microbiol. 18, 288e301.