Microbial Diversity and Community Structure in Deadwood of Fagus

Microbial diversity and community structure in deadwood of Fagus

sylvatica L. and Picea abies (L.) H. Karst

Inaugural-Dissertation zur Erlangung der Doktorwürde der Fakultät für Umwelt und Natürliche Ressourcen der Albert-Ludwigs-Universität Freiburg i. Brsg.

vorgelegt von

Dipl.-Forstwirt Björn Hoppe

Freiburg im Breisgau

Dekan: Prof. Dr. Tim Freytag

Referent: Prof. Dr. François Buscot

Korreferent: Prof. Dr. Siegfried Fink

Disputationsdatum: 03.02.2015 ii

Zusammenfassung

Totholz wird im Rahmen forstwirtschaftlicher Aktivitäten zunehmend größere Bedeu- tung zuteil. Die Forstwirtschaft hat unlängst realisiert, dass die Förderung und der Er- halt natürlicher Totholzvorkommen von immenser Wichtigkeit für Ökosystemdienst- leistungen sind, nicht zuletzt, weil Totholz ein Reservoir für biologische und funktionel- le Diversität darstellt.

Das Hauptziel dieser umfassenden Studie bestand zum einen in der Untersuchung des

Einflusses unterschiedlicher Waldbewirtschaftungsstrategien auf die mikrobielle Diver- sität an Totholz zweier in Deutschland forstlich relevanter Baumarten Fagus sylvatica and Picea abies. Zum anderen sollte der Zusammenhang zwischen baumartspezifischen physikalischen und chemischen Parametern und den damit verbundenen Veränderungen der mikrobiellen Diversität aufgeklärt werden. Ein nicht unerheblicher methodischer

Fokus lag hierbei in der Verknüpfung moderner molekularbiologischer Techniken

(„Next-generation sequencing“) mit klassischer, auf Observation basierender Feldmyko- logie. Die vorliegende Arbeit umfasst drei unabhängige, sich aber ergänzende Kapitel.

Kapitel 2 umfasst eine Studie, die die Diversität und Artenzusammensetzung von Pilz- gemeinschaften an und im Totholz der beiden oben genannten Baumarten beschreibt.

Des Weiteren wurde untersucht, inwieweit sich holzphysikalische und -chemische Ei- genschaften und pilzliche Diversität bedingen und ob eine erhöhte Diversität zu erhöh- ter Enzymaktivität und somit schnelleren Holzabbauraten führt. Hierfür wurden mole- kularbiologische Methoden (454-Pyrosequenzierung) zur Identifikation der Pilze mit

Enzymassays und umweltanalytischen Verfahren (Massenspektrographie) zur Elemen- tenanalyse kombiniert. Diese Arbeit stellt die erste umfassende Gegenüberstellung von

iii

Zusammenfassung

Totholz-(„ökosystemen“) der zwei in Deutschland forstwirtschaftlich relevantesten

Laub- und Nadelbaumarten dar. Das Ergebnis der Studie stellt klar, dass Pilze spezifisch auf Totholz der entsprechenden Baumarten vorkommen. Dies lässt sich entschei- dend von der festgestellten Spezifität der physikalischen und chemischen Substrateigen- schaften ableiten. Darüber hinaus beeinflusst die ausgeübte Waldbewirtschaftung (Na- turwald versus Altersklassenbewirtschaftung) die Zusammensetzung der Artengemein- schaft. Beide Baumarten weisen auch unterschiedliche Sukzessionsmuster1 („dynamics“) auf. So tritt z.B. der Weißfäulepilz Resinicium bicolor dominant an Fichten- stämmen aller Zersetzungsstufen auf, wobei Vertreter der Xylariaceae das initiale Zer- setzungsstadium der Buchen dominieren und im weiteren Verlauf von Polyporaceae wie Fomes formentarius abgelöst werden.

In Kapitel 3 ist die Untersuchung zur Diversität von nifH-Genen im Totholz beschrieben. Stickstoffverfügbarkeit im Totholz ist stark beschränkt und seit den 1960er Jahren wird vermutet, dass Pilze, die genau den Stickstoff zur Fruktifikation benötigen, Asso- ziationen mit Bakterien eingehen, die unter großer Energieaufwendung in der Lage sind, atmosphärischen Stickstoff zu spalten/ binden und somit biologisch verfügbar machen.

Das hierbei untersuchte nifH-Gen, welches für das Enzym Dinitrogenase-Reduktase kodiert, fungiert als etabliertes Markergen für stickstofffixierende Bakterien in ver- schiedensten Umweltkompartimenten und Ökosystemen. Unsere auf „clone library sequencing“ basierenden Untersuchungen konnten zeigen, dass die von uns identifizierten nifH-Gene spezifisch im Totholz vorkommen und bisher nicht in anderen Substraten verschiedener Ökosysteme (z.B. Meere, Flüsse, Seen, Böden, Ölfelder u.v.m.) detektiert wurden. Außerdem ließ sich die Artengemeinschaft Artenzusammensetzung auch zwi-

1 Von Sukzession im eigentlichen Sinne, kann nicht gesprochen werden, da keine Zeitreihen untersucht wurden. iv

Zusammenfassung schen den beiden untersuchten Baumarten auftrennen und differenzieren. Darüber hinaus zeigen die Ergebnisse einen signifikanten Zusammenhang zwischen Diversität der nifH-Gene und fruktifizierender Pilze. Zusätzlich konnte durch aufwendige Netzwerka- nalysen bestätigt werden, dass Interaktionen (sogenannte „Co-occurrence patterns“) zwischen bestimmten Stickstofffixierenden Bakterien und Pilzen nicht zufälliger Natur sind, sondern determiniert sind, also bestimmten Abhängigkeiten unterliegen.

Während in Kapitel 2 die pilzliche Diversität an Totholz beschrieben wird, präsentiert

Kapitel 4 die Ergebnisse zur gesamtbakteriellen Diversität. Auch hierbei wurde 454- pyrosequenciert, um Bakterien in den jeweiligen Totholzstämmen verschiedener Zerset- zungsstadien zu identifizieren. Alphaproteobacteria, Acidobacteria and Actinobacteria waren die dominanten taxonomischen Gruppen in beiden Totholzsubstraten. Interessan- terweise war ein signifikanter Anstieg von Bakterien der Ordnung Rhizobiales an Fa- gus-Stämmen der Zersetzungsstufe 3 zu verzeichnen (25% der Gesamtdiversität). Ein

Großteil der Vertreter dieser Ordnung „besitzt“ das nifH-Gen, welches die Expression der Dinitrogenase-Reduktase ermöglicht. Des Weiteren veranschaulichen die Ergebnis- se, dass es auch im Falle der Artenzusammensetzung der Bakterien ähnliche und starke

Korrelationen zu den holzphysikalischen und –chemischen Parametern gibt, wie sie in

Kapitel 2 für Pilze identifiziert wurden. Außerdem konnte ein negativer Einfluss inten- siver Waldbewirtschaftung auf die bakterielle Diversität nachgewiesen werden.

Insgesamt tragen alle Ergebniskapitel zum Nachweis einer bisher unbekannten Vielfalt von Pilzen und Bakterien bei, die potentiell am Totholzabbau beteiligt sind. Dass diese

Gemeinschaften nicht unabhängig voneinander leben, sondern untereinander interagie- ren und eine Dynamik im Zusammenhang mit Baumart, Zersetzungsstufe, aber auch mit externen Variablen wie Waldbewirtschaftung aufweisen, zeigt, dass sie integraler und funktioneller Bestandteil der Biodiversität im Wald sind, und genau wie auffälligere v

Zusammenfassung

Gruppen der Waldbiozönose (z.B. Pflanzen, Insekten und Vertebraten) bei Untersu- chungen zur Waldfunktionen und -dienstleistungen, zu berücksichtigen sind. Die neuen

Sequenzierungsmethoden ermöglichen es erst seit kurzer Zeit, diese Blackbox rational zu erforschen.

Summary

Deadwood is gaining greater importance in the context of forestry activities. The forestry sector meanwhile has realized that the promotion and preservation of natural deadwood occurrences is of immense importance for ecosystem services, since it serves as a reservoir for biological and functional diversity. An important objective of this comprehensive study/ thesis was to explore the impact of different forest management intensities on the microbial diversity in deadwood of the two in Germany silviculturally important tree species Fagus sylvatica and Picea abies. Furthermore, the relationship between tree species-specific physico-chemical properties and the associated changes of microbial diversity and species composition should be elucidated. A significant methodological focus was on the combination of modern molecular techniques ("Next- generation sequencing") together with classical field (-observation) mycology. This recent thesis comprises three independent but complementary chapters.

Chapter 2 encompasses a study on the diversity and structure of fungal communities in the deadwood of the two above mentioned tree species Fagus sylvatica and Picea abies.

This study further investigates the relation between physico-chemical properties and fungal diversity and whether an increase in diversity refers to higher lignin-modifying enzyme activities and increased wood decomposition rates, respectively. The study combines 454 pyrosequencing (to identify the fungal community), enzyme assays and chemical elements analyses (mass spectrography). The work presents the first comparison of the deadwood (-ecosystems) of these two economically relevant coniferous and deciduous tree species. The results revealed that the fungal communities strongly correspond to the specific deadwood substrate, which occurred to be independently from the

vii

Summary surrounding habitat. This fact clearly relates to the specific substrate properties. The results further revealed that forest management type (age class managed beech and spruce forests versus extensively managed beech forests) significantly impacts the fungal community structure on the according forest plots. Both tree species also displayed distinct fungal successional patterns (dynamics). White-rot causing Resinicium bicolor, for example, occurred to be dominant on Picea abies deadwood logs of all decay stages, whereas members of the Xylariaceae were dominating the initially decayed logs of

Fagus sylvatica and then were substituted by Polyporaceae (e.g. Fomes formentarius).

Chapter 3 describes the investigations of the distribution of nifH genes in the same deadwood logs described in chapter 2. N-availability is highly restricted in deadwood and since the 1960s, it is assumed that fungi meet their N requirements for forming fruiting bodies by associating with bacteria which are capable to fix atmospheric dinitrogen and make it biologically available. The investigated nifH gene which encodes for dinitrogenase reductase hereby serves as the marker gene for diazotrophs and has been detected in various environments and ecosystems. The results of the presented study revealed that the nifH genes that were detected in deadwood are highly specific to this substrate and have not been found in other substrates (e.g. oceans, rivers, lakes, soils and oil spills). The study further showed that the nifH OTU community structure significantly varied between deadwood logs of Fagus and Picea. The results also indicate a significant correlation between nifH OTU richness and sporocarps on the investigated logs. Network analyses based on non-random species co-occurrence patterns revealed interactions among fungi and N-fixing bacteria in the deadwood and strongly indicate the occurrence of at least commensal relationships between these taxa.

viii

Summary

Chapter 4 presents the results of the first investigation of the total bacterial diversity in decaying deadwood under natural conditions. Alphaproteobacteria, Acidobacteria and

Actinobacteria were the dominant taxonomic groups on both tree species. There were no differences in bacterial OTU richness between the tree species but richness tended to increase with progressing wood decay. Interestingly, bacteria from the order Rhizobiales became more abundant during the intermediate and advanced stages of decay, accounting for up to 25% of the entire diversity in such logs. The most dominant OTU was taxonomically assigned to the genus Methylovirgula, which was before isolated from woodblocks of Fagus sylvatica and has been shown to also possess the nifH gene. In line to the results on fungi (chapter 2) this study also demonstrates that the bacterial community structure was influenced by a range of deadwood species' physico-chemical properties including decay stage, water content, pH, and C and N availability. Further- more, intensive forest management also negatively impacted bacterial diversity.

Altogether, the results of the three presented chapters of this thesis substantially contribute to the identification of rather unknown fungal and bacterial diversity, that are part of deadwood decomposition processes. The fact that these communities are not independent from each other and rather interact depending on e.g. tree species, decay stage and forest management practices, demonstrates that they have to be considered as an integral and functional part of biodiversity in forest ecosystems. By applying novel molecular sequencing techniques, it is possible to further explore this “blackbox”.

Publications and contributions of co-authors

Chapters 2, 3 and 4 have been published in peer-reviewed journals.

Chapter 2 – Hoppe et al. was published in August 2015 (online August 11th 2015) as a research article in Fungal Diversity.

(http://link.springer.com/article/10.1007/s13225-015-0341-x)

Hoppe B†, Purahong W†, Wubet T†, Kahl T, Bauhus J, Arnstadt T, Hofrichter M,

Buscot F, Krüger D (2015) Linking molecular wood-inhabiting fungal diversity and

community dynamics to ecosystem functions and processes in Central European for-

ests. Fungal Diversity, doi: 10.1007/s13225-015-0341-x

BH, WP and TW have equally contributed to this manuscript. BH and WP wrote the

manuscript, performed statistics and interpreted results. TW provided and performed

bioinformatic pipeline for subsequent analysis of 454 data. TK, JB, FB and DK have

made substantial contributions to conception and design of this study. TK and TA

provided substantial data on lignin-modifying enzymes and wood physico-chemical

properties. All authors participated in improving the article.

Publications and contributions of co-authors

Chapter 3 – Hoppe et al. (2014) was published in February 2014 (online February 5th

2014) as a research article in PLoS ONE.

(http://www.plosone.org/article/info%3Adoi%2F10.1371%2Fjournal.pone.008814)

Hoppe B, Kahl T, Karasch P,Wubet T, Bauhus J, Buscot, Krüger D (2014) Network

analysis reveals ecological links between N-fixing bacteria and wood-decaying fun-

gi. PLoS ONE, e88141. doi:10.1371/journal.pone.0088141

BH performed statistics, interpreted data and wrote the manuscript. DK and BH sub-

stantially contributed to conception and design of this “side”-study and developed the

idea to perform an experiment on nifH genes in deadwood. DK with help of BH per-

formed phylogenetic analysis and wrote that part for “Material and Methods”. TK,

JB, FB substantially contributed to the broad conceptional design of the study. TK

provided substantial data on wood physico-chemical properties. PK sampled and de-

termined sporocarps. TW critically helped with rebuttal. All authors substantially

improved the article.

Chapter 4 – Hoppe et al. was published in April 2015 (online April 8th 2015) as a research article in Nature Scientific Reports.

(http://www.nature.com/srep/2015/150329/srep09456/full/srep09456.html)

Hoppe B, Krüger D, Kahl T, Arnstadt T, Buscot F, Bauhus J, Wubet T A pyrose-

quenced insight into sprawling bacterial diversity and community dynamics in de-

Publications and contributions of co-authors caying logs of tree species Fagus sylvatica and Picea abies. Scientific Reports, 5, doi:10.1038/srep09456

BH performed statistics, interpreted data and wrote the manuscript. TW provided and performed bioinformatic pipeline for subsequent analysis of 454 data. DK and BH contributed to conception and design of this study investigating bacteria in deadwood. TK, JB, DK, FB substantially contributed to the broad design of the study. TK and TA provided substantial data on wood physico-chemical properties. All authors substantially improved the article.

xii

Contents

Zusammenfassung ...... iii

Summary ...... vii

Publications and contributions of co-authors ...... x

Contents ...... xiii

List of Figures ...... xvi

List of Tables ...... xxi

List of abbreviations ...... xxiv

1 Introduction ...... 1 1.1 Deadwood in forest ecosystems ...... 1 1.2 Wood decay by fungi ...... 3 1.3 The role of fungi in wood-decaying processes ...... 6 1.4 The role of bacteria in wood-decaying processes ...... 7 1.5 Approaches to assess microbial diversity ...... 10 1.6 The impact land use on biodiversity ...... 13 1.7 Research objectives and hypotheses ...... 15

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests ...... 19 2.1 Introduction ...... 19 2.2 Material and methods ...... 22 2.2.1 Experimental design, deadwood selection and sampling ...... 22 2.2.2 Wood physico-chemical properties and lignin-modifying enzyme assays ...... 23 2.2.3 DNA isolation, PCR and pyrosequencing ...... 24 2.2.4 Bioinformatic analysis ...... 26 2.2.5 Statistical analysis ...... 26 2.3 Results ...... 28 2.3.1 Wood physico-chemical properties in different decay classes ...... 28 2.3.2 Wood-inhabiting fungal richness and community structure ...... 29 2.3.3 Pyrosequencing and community sampling statistics ...... 33

xiii

Contents

2.3.4 WIF dynamics on deadwood of different tree species ...... 34 2.3.5 Factors correlating to fungal community structure in the two deadwood species ...... 35 2.3.6 Relationships between wood-inhabiting fungal richness, taxonomy and lignin-modifying enzyme activities ...... 37 2.3.7 Relationships between wood-inhabiting fungal richness, taxonomic identity and wood decomposition rates ...... 38 2.4 Discussion ...... 39 2.4.1 Fungal diversity and community composition ...... 40 2.4.2 WIF dynamics during decomposition processes and corresponding factors ...... 40 2.4.3 The role of WIF in ecosystem functions and processes ...... 43 2.5 Conclusion ...... 47 2.6 References ...... 48 2.7 Supplemental information ...... 53

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi ...... 70 3.1 Introduction ...... 70 3.2 Material and methods ...... 72 3.2.1 Experimental design ...... 72 3.2.2 Deadwood ...... 72 3.2.3 Fungal sporocarp inventories ...... 74 3.2.4 DNA isolation ...... 74 3.2.5 PCR, cloning and initial sequence analysis ...... 75 3.2.6 Compilation of a nifH Database ...... 76 3.2.7 Protein alignment and phylogenetic analyses ...... 77 3.2.8 Statistical analysis ...... 79 3.3 Results ...... 81 3.3.1 PCR clone library analysis and clustering ...... 81 3.3.2 Protein phylogenetic analysis of the functionality of nifH genes from deadwood ...... 83 3.3.3 nifH sequence diversity ...... 88 3.3.4 Fungal diversity and influencing factors ...... 89 3.3.5 Co-occurrence patterns ...... 92 3.4 Discussion ...... 95 3.4.1 Characterization and diversity of nifH sequences in deadwood ...... 95 3.4.2 Correlations of nifH community structure with environmental settings ...... 97 3.4.3 Interrelation of fungal fructification and N-fixing bacteria through N- availability ...... 98 3.5 References ...... 102 xiv

Contents

3.6 Supplemental information ...... 107

4 A pyrosequenced insight into sprawling bacterial diversity and community dynamics in decaying logs of tree species Fagus sylvatica and Picea abies ...... 119 4.1 Introduction ...... 119 4.2 Material and methods ...... 122 4.2.1 Experimental design, deadwood selection and physico-chemical properties .... 122 4.2.2 DNA Isolation, PCR and Pyrosequencing ...... 123 4.2.3 Bioinformatic analysis ...... 124 4.2.4 Statistical analyses ...... 125 4.3 Results ...... 126 4.3.1 Wood properties ...... 126 4.3.2 Sequence analyses ...... 127 4.3.3 Bacterial 16S rRNA diversity and richness ...... 127 4.3.4 Effect of forest management on bacterial richness ...... 131 4.3.5 Bacterial community structure and variation with progressing wood decay ..... 132 4.4 Discussion ...... 136 4.5 References ...... 142 4.6 Supplemental information ...... 146

5 Synthesis and discussion ...... 156 5.1 General rationale and achievement ...... 156 5.2 The potential and limits of Next-generation sequencing as the method of choice ...... 157 5.3 Impact of land use intensity on deadwood volume and species richness ...... 163 5.4 Does the substrate control community structure and diversity? ...... 167 5.5 The role of N-fixation in deadwood decomposition ...... 171 5.6 References ...... 175

Acknowledgements ...... 183

List of Figures

Fig. 1.1: Wood disc sampled on a Fagus sylvatica log of decay class 3 (CWD_ID 08815). It displays the small scale heterogeneity and potential microhabits and niches that are occupied by various fungal species...... 2 Fig. 1.2: Illustration of a cell wall model and its five layers: ML = middle lamella, PW = primary wall, S1 = outer secondary wall, S2 = middle secondary wall, S3 = inner secondary wall...... 3 Fig. 1.3: Overview of three German Biodiversity Exploratories. In detail: topographic map of the “Schwäbische Alb”, including all grassland and forest at the EP and VIP level...... 15 Fig. 1.4: Overview of the working packages of this thesis: (1) How forest management practices and wood physico-chemical properties change forest stand structure and correspond to fungal and bacterial communities and (2) how do enzyme activities influence wood d ecay by fungi and N- fixation by bacteria. (Modified from Krüger et al. (unpublished)) ...... 17 Fig. 2.1: Decay rates, mass losses, and wood physico-chemical properties of Fagus and Picea deadwood logs of different decay classes. The figure shows the relative deviation from the sample mean in each case. Mean values of all parameters for deadwood logs of bot both species are shown in the middle. The differences between the two deadwood species and also among different decay classes for individual deadwood species were analyzed by t-tests and one-way analysis of variance, incorporated in the form of Fisher's Least Significant Difference (ns = not significant, * P < 0.05, ** P < 0.01, *** P < 0.001). The four FASY (Fagus sylvatica) and PIAB (Picea abies) decay classes are shown in different color shades...... 29 Fig. 2.2: Patterns of cOTUs on Fagus and Picea deadwood displayed as sunburst charts (beatexcel.com). The outer rings show the cumulative relative abundances of different fungal OTUs (cOTUs) at the species and genus levels with at least 95% and 90% probability o of correct identification, respectively, based on secondary BLAST re-analysis. The middle rings display the cumulative relative abundances at the family level. The inner rings display the cumulative relative abundances of different concordant fungal phyla. Only cOTUs that accounted for > 1% of the total abundances and identified at least to genus are labeled...... 32 Fig. 2.3: 3D-Nonmetric multidimensional scaling (NMDS) ordination of fungal community structure in Fagus (green) and Picea (red) deadwood using the plot3d and ordigl functions in R. The NMDS ordination (stress = 0.16) was fitted to wood physico-chemical and anthropogenic factors (a) and also abundances of different fungal families (b) by using the envfit command in “vegan”. ANOSIM revealed significant separation of fungal community structure according to tree species (R = 0.60, P < 0.001, 999 permutations)...... 36

xvi

List of Figures

Fig. S2.4: Number of sequence reads and number of detected OTUs (rarefaction curves calculated as individual rarefaction in PAST) of all samples used in the study. Four decay classes of (a) Fagus (FASY) and (b) Picea (PIAB) differentiated by line colors...... 64 Fig. S2.5: Relationships between fungal OTUs richness and decay classes (as determined by density (a) and remaining mass (b)), green = Fagus sylvatica, red = Picea abies...... 64 Fig. S2.6: Heatmap displaying cOTU abundances and presences per sample. This figure is a companion to Fig. 2.2 of the main text and contains only the cOTUs that appear labeled in that figure. The rows are sorted by the number of sequences on the family level, which also only includes the cOTUs appearing in this table...... 65 Fig. S2.7: Comparison of sharedness and species (= cOTU) distribution. Green colors indicate Fagus, red colors Picea. (a) Table of 74 species that are shared between Fagus and Picea made as MS Excel in-cell dot chart. The left and right columns are filled by a heatmap corresponding to overall cOTU size in number of sequences while the bars are relative to cOTU size within the tree species. Up to 150 sequences this dot chart is size- correct. (b) Area-correct Venn diagram for sharedness of genera (Pan- Omics Research, Pacific Northwest National Laboratory)...... 66 Fig. S2.8: Heatmap (green-white) of mean abundances of highly detected fungal families and cOTUs in different decay classes (DC, labeled FASY_1 through FASY_4) in Fagus sylvatica deadwood logs...... 67 Fig. S2.9: Heatmap (red-white) of mean abundances of highly detected fungal families and cOTUs in different decay classes (DC, labeled PIAB_1 through PIAB_4) in Picea abies deadwood logs...... 68 Fig. S2.10: Some examples: Fagus sylvatica logs showing pseudosclerotial plates (PSPs) due to the colonization by Xylariaceae...... 69 Fig. 3.1: Rank abundance chart displaying the distribution of the 12 most abundant nifH OTUs derived from the deadwood dataset within the compiled nifH dataset comprising 26,383 sequences. Only the 200 largest OTUs are shown due to space limitations. Colored bars indicate deadwood tree species (green, Fagus sylvatica; red, Picea abies). The inserted table lists the best BLASTn hit reference sequences in NCBI Genbank for the same 12 most abundant wood-derived OTUs from our study...... 82 Fig. 3.2: Distribution of the 7,730 nifH OTUs according to the environments where they have been detected, and whether described as originating from an isolate in GenBank. 168 of the 176 OTUs derived from this deadwood study have been exclusively identified in wood samples. The integrated heatmap displays proportions of rare sequence types (singletons)...... 83 Fig. 3.3: Phylogenetic tree of nifH protein sequences. 50% majority rule consensus tree of 13,500 PhyloBayes post burn-in trees, unrooted. Black values at internodes = Bayesian Posterior Probability (if > 0.5). Pink values = MEGA5 Maximum Parsimony (MP) bootstrap support (if > 50). Green values = GARLI Maximum Likelihood bootstrap support (if > 50). Terminal triangles represent monophyletic clades with MOTUs solely of one tree species origin, collapsed but keeping the internal distance xvii

List of Figures

(substitutions per site, see scale bar), in light pink = 50-79 MP bootstrap support, dark pink = 80-100 MP bootstrap support. Green color indicates MOTUs solely from Fagus origin, red color Picea origin and dark blue color mixed origin (with bars showing ratio of [green] vs. [red]). Terminal labels with sequences from this study: MOTU ID (SMOTU = singleton MOTU), total number of sequences, FASY = from Fagus, PIAB = from Picea, followed by number of sequences in the same order, then forest management type(s) (AC.Conif = managed spruce forests, AC.Decid = managed beech forests, Extensiv = extensively managed beech forests) and number of sequences in same order. Terminal labels with sequences from other sources: near BLAST hit, summary of ecological data of sequences in that MOTU. MOTUs that contain nucleotide sequences flagged as potential chimeras appear in italics and with the term PotChim (only present in Fig. 3, Supergrade). The width of visible terminal branches represents the number of sequences (size correct up to 10 sequences). To the right, amino acid sequence logos and Kyte-Doolittle hydophobicity alignments for labeled nodes on the tree. The small tree shape (based on screenshot from Archaeopteryx v.0.972 ) shows the position within the complete phylogenetic tree...... 85-87 Fig. 3.4: nifH community structure A: Multivariate regression tree of nifH OTU community composition estimated from sequences of the clone library obtained from deadwood of Picea abies and Fagus sylvatica. Analyses were conducted for different decay classes, based on the remaining mass per deadwood log after decay using k-means cluster analysis. B: Principal Component Analysis biplots of the group means of the multivariate regression tree. The larger circles (per color) represent the multivariate group means, the individual logs are denoted by smaller circles, with matching colors and designation to Fig. 3.4A. The identity of selected OTUs with characteristic discriminatory loading is specified. Each OTU label is located at its weighted mean from the group means. Intersect correlation is given in brackets...... 88 Fig. 3.5: Interrelations between sporocarp richness and remaining mass after decay in%, nifH OTU richness and log-transformed nitrogen content per density unit (N (g/cm³)) (A, B, C). The figure displays interrelations separately per deadwood species...... 92 Fig. 3.6: Network organized around 23 nifH OTUs and 27 fungal species (abbreviations according to the legend). Fungi (sporocarps) and nifH OTUs serve as connected nodes, solid lines display co-occurrence patterns (Z-Score < -1.96) and dotted lines avoidance patterns. Edge widths display significance levels from thinnest = 0.049 to thickest = 0.0017. Differently shaped and colored nodes/ hubs display taxonomic differences on phylum level and their ecological role in wood decay. Subnetworks are grouped by tree species, and colored background circles indicate affiliations of included taxa to substrate deadwood species (green = Fagus affiliated, red = Picea affiliated, blue = unaffiliated “Generalists”)...... 94 Fig. S3.7: Sampling scheme visualized using Treemap v. 3.1.0. (Macrofocus, Zurich, Switzerland) in squarified layout. Items are grouped by management type. Treemap cell size is proportional to mass loss in% xviii

List of Figures

(smaller cells = less decayed logs) Colors represent tree species. (red = Picea abies, green = Fagus sylvatica). Numbers indicate the ID of the deadwood item...... 114 Fig. S3.8: Splitstree reticulogram. The three major parts of the phylogeny (compare phylogenetic tree in Fig. 3.3) are labeled here...... 115 Fig. S3.9: Bargraphs including standard errors displaying nifH OTU richness (A) and nitrogen content per density unit (B) within deadwood tree species...... 115 Fig. S3.10: Scatterplot displaying number (= richness) of fruiting fungal species per deadwood log. Solid vertical lines display mean values of richness, dashed vertical lines median richness per tree species (green = Fagus sylvatica, red = Picea abies). Heatmapped bars to the left and right display density probability as calculated by kernel density estimation using the denstrip package in R (Jackson CH (2008) Displaying uncertainty with shading. Am Stat 62: 340-347.) ...... 116 Fig. S3.11: Relative abundances (left) of Basidiomycota and Ascomycota on deadwood logs of Fagus sylvatica and Picea abies and mean number of sporocarps per tree species (right)...... 116 Fig. S3.12: Interrelation of nifH OTU richness and remaining mass after decay in% (A) and water content in% (B) and water content in% and remaining mass after decay in% (C) on logs of Fagus sylvatica...... 116 Fig. S3.13: Non-random sporocarp – nifH OTU community assembly assessed by C-score distribution and Checkerboard index for observed and expected/ randomized species occurrences...... 117 Fig. S3.14: Boxplots including median, upper and under quartiles and whiskers displaying the interrelation of deadwood species and log transformed carbon content per density unit (A) and log transformed wood density (B). .. 117 Fig. S3.15: Effects of remaining mass after decay in% on log-transformed nitrogen content per density unit (g*cm³) and N concentration in g*g-1. Interrelations are displayed separately per wood species...... 118 Fig. 4.1: Relative abundances of phylogenetic groups (bacterial phyla including proteobacterial classes) in deadwood from two species (Fagus sylvatica = FASY, Picea abies = PIAB) in different stages of decay (decay classes 1- 4). OTUs that could not be taxonomically assigned at the phylum/subphylum level are reported as “Others” and comprise 0.006% of all sequences. The category “other” also includes all OTUs with <1.5% relative sequence abundance...... 129 Fig. 4.2: Relative abundances of three dominant phylogenetic groups (bacterial orders) in deadwood of the two studied tree species (Fagus sylvatica = FASY, Picea abies = PIAB) in different stages of decay (decay classes 1- 4)...... 131 Fig. 4.3: Two-dimensional non-metric multidimensional scaling (NMDS) ordination plots of bacterial community structure across the different tree species at each stage of decay (FASY1-4, PIAB1-4). Plots show centroids within a single decay stage, bars represent one SD along both NMDS axes. Statistical significances (R2 and P-values) are based on Goodness-of-fit

xix

List of Figures

statistics for environmental variables and bacterial order abundances per sample...... 135 Fig. S4.4: Sampling scheme visualized using Treemap v. 3.1.0. (Macrofocus, Zurich, Switzerland) in squarified layout. Items are grouped by management type and corresponding forest plot. Treemap cell size is proportional to remaining mass in% (smaller cells = less decayed logs) Colors represent tree species. (red = Picea abies, green = Fagus sylvatica). Numbers per cell indicate the Tree_ID, Plot_ID, length of deadwood log in (m) and the number of wood subsamples taken per log...... 147 Fig. S4.5: Bacterial richness of the two deadwood species Fagus sylvatica and Picea abies distinguished upon the 4 different decay classes (e.g. FASY1 = Fagus sylvatica, decay class 1) at a genetic cutoff of 3%. Richness is displayed as the mean number of observed sequences per tree decay classes. Differences of bacterial species richness between decay classes and tree species were analyzed by employing one-way analysis of variance and Fisher’s Least Significant Difference (LSD) post hoc test...... 151 Fig. S4.6: Correlation between bacterial OTU richness and remaining mass in%. Green triangles represent data on FASY (Fagus sylvatica) and red diamonds on PIAB (Picea abies)...... 151 Fig. S4.7: Relative abundances of four selected bacterial OTUs (methano- and methylotrophic genera) per wood tree species (Fagus sylvatica = FASY, Picea abies = PIAB) and connected decay classes (1-4)...... 152 Fig. S4.8: Mean bacterial OTU richness according to differently managed forest types (including all decay classes). (A) Fagus sylvatica logs in each of the three differently managed forest plots and (B) vice versa of Picea abies logs. Fisher’s Least Significant Difference (LSD) post hoc were calculated separately for (A) and (B)...... 152 Fig. S4.9: Correlation between bacterial OTU richness and the two land use intensity indices ForMI (light grey squares) and SMI (dark grey diamonds)...... 153 Fig. S4.10: Relative abundances and their contribution to total community variation of bacterial families which represent > 0.75 of the sequences of the whole dataset as calculated by SIMPER. Both, Fagus (FASY) and Picea (PIAB) were separated in four wood decay classes. Heatmaps (white = zero, red = maximum abundance) were overlaid separately for Fagus and Picea...... 153 Fig. S4.11: Kronafile SK1: Interactive web-visualization of taxonomic information on all deadwood logs using Krona (http://sourceforge.net/projects/krona/; Ondov et al. 2011)...... 154 Fig. S4.12: Kronafile SK2: Interactive web-visualization of taxonomic information on Fagus sylvatica deadwood logs using Krona (http://sourceforge.net/projects/krona/; Ondov et al. 2011)...... 154 Fig. S4.13: Kronafile SK3: Interactive web-visualization of taxonomic information on Picea abies deadwood logs using Krona (http://sourceforge.net/projects/krona/; Ondov et al. 2011)...... 155 Fig. 5.1: Comparison of correlations between sporocarps richness (fungi)/ ARISA- based OTU richness (arisa) as the chosen method (meth) and

List of Tables

environmental data: (upper left - a) exploratory (exp = Alb – Schwäbische Alb, Hai - Hainich, Sch – Schorfheide), (upper right – b) tree species (Fasy – Fagus sylvatica, Pcab – Picea abies, Pisy – Pinus sylvestris, (lower left - c) remaining mass (mass.perc), and (lower right - d) volume of the log (v.log)...... 161 Fig. 5.2: Workflow of an SIP- (stable isotope propping) microcosm experiement. 15 Environmental wood samples are incubated with a N2-labelled substrate, and microorganisms that actively assimilate this substrate will incorporate 15 15 the N2 into their DNA. The “heavy” N2-labelled nucleic acids can then be separated from the “light” unlabelled nucleic acids by an ultracentrifugation. Microbial member then will be identified in the different fractions by applying molecular tools...... 174

List of Tables

Table 2.1:Goodness-of-fit statistics (R2) for parameters fitted to the nonmetric multidimensional scaling (NMDS) ordination of fungal community structure. The significance estimates were based on 999 permutations. Significant factors (Bonferroni corrected P < 0.05) are indicated in bold. Marginally significant variables (Bonferroni corrected P < 0.10) are indicated in italics...... 37 Table S2.2: Sampling design: Distribution of deadwood logs according to tree species and the respective 9 forest plots which were assigned to three different forest management types...... 55 Table S2.3: List of MID tags for deadwood logs (samples) and additional characterization of sampling design. For further details regarding remaining mass of the respective deadwood log, compare Fig. S1 in Hoppe et al. (2014)...... 56 Table S2.4: Decay rate, remaining mass and wood physico-chemical parameters of Fagus (FASY) and Picea (PIAB) deadwood logs in different decay classes (DC 1 through 4)...... 58 Table S2.5: Mean ligninolytic enzyme activities and macro/ micro nutrients of Fagus and Picea deadwood logs...... 58 Table S2.6: Total observed OTU richness and estimated OTU richness (Chao1 and ACE) of Fagus and Picea deadwood logs in different decay classes (DC)...... 59 Table S2.7: Summary of fungal sequences, OTU statistics and fungal phyla distribution from MOTHUR analysis...... 59 Table S2.8: Relationships between fungal richness, abundances of fungal families and cOTUs and activities of ligninolytic enzymes revealed by non- parametric Kendall-tau correlation analysis. Significant correlation (P < 0.05) are displayed in bold. Marginal significant correlations (P < 0.1) are displayed in italics...... 60

xxi

List of Tables

Table S2.9: Relationships between fungal richness, abundances of fungal families and cOTUs and wood decomposition rates as revealed by non-parametric Kendall-tau correlation analysis. Significant correlation (P < 0.05) are displayed in bold. Marginal significant correlations (P < 0.1) are displayed in italics...... 61 Table S2.10: Fungal cOTUs and their potential roles in deadwood decomposition. FASY= Fagus sylvatica, PIAB = Picea abies...... 63 Table 3.1: Results of perMANOVA analysis of Bray-Curtis dissimilarities in nifH OTU community structure in relation to tree species, decay class (based on remaining mass after decay) and management type and their interactions, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseudo-F = F value by permutation, boldface indicates statistical significance at P < 0.05, P-values based on 999 permutations (lowest P- value possible is 0.001)...... 89 Table 3.2: ANOVA table of effects of the indicated factors on fungal fructification ability. Complete model summary representing R2, F, P statistics. Abbreviations of the depicted ANOVA table Df = degrees of freedom, SS = sum of squares, MS = mean sum of squares. The summary model is as follows: R², F, and p were 0.5208, 8.476 and <0.001 (significant), respectively. Boldface indicates statistical significance...... 91 Table S3.3: Sequence percentage identity of OTUs taxonomically assigned through BLASTn against GenBank (uncultured/ environmental sample sequences excluded). 19 OTUs were assigned to Rhizobiales at a 95% similarity threshold (65 OTUs at ≥ 90%). A total of 80 OTUs were identified to genus level...... 107 Table S3.4: List of sporocarps identified on the respective deadwood trees A) Fagus sylvatica B) Picea abies...... 108 Table 4.1: Results of perMANOVA analysis of the Bray-Curtis dissimilarities for bacterial OTU community structure in relation to tree species, decay class (assigned based on the remaining mass of the log in question), management regime, and their interaction, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseudo-F = F value by permutation. Boldface indicates statistical significance (P < 0.05); P- values are based on 999 permutations (i.e. the lowest possible P-value is 0.001)...... 133 Table 4.2: Goodness-of-fit statistics (R2) for parameters fitted to the non-metric multidimensional scaling (NMDS) ordination of bacterial community structure. The significance estimates were based on 999 permutations. Significant factors (Bonferroni corrected P < 0.05) are indicated in bold. Marginally significant variables (Bonferroni corrected P < 0.10) are indicated in italics...... 134 Table S4.3: Sampling design: Distribution of deadwood logs according to tree species and the respective 9 forest plots which were assigned to three different forest management types...... 147 Table S4.4: Mean values of wood properties and standard error for each tree species related decay class and ANOVA P values. Differences in C/N, concentrations per wood density and relative wood moisture between xxii

List of Tables

decay classes were analyzed by employing one-way analysis of variance and Tukey pair-wise comparisons. Significant ANOVA P values are shown in bold (P < 0.05). Different letters indicate differences among decay classes. (P < 0.05)...... 148 Table S4.5: Correlations among assessed wood physico-chemical properties. Significant values (P < 0.05) are given in bold...... 148 Table S4.6: Mean relative abundances of dominant bacterial phyla (including proteobacterial subphyla) at different decay stages. Differences of bacterial abundances between decay classes and tree species were analyzed by employing one-way analysis of variance and Fisher’s Least Significant Difference (LSD) post hoc test...... 149 Table 4.7: Pearson rank correlation table of selected dominant bacterial phyla (including proteobacterial subphyla) with wood physico-chemical properties. Significant correlations (P < 0.05), are displayed in bold. Calculations were performed separately for the respective tree species...... 149 Table 4.8: Pearson Rank correlation table of selected methylo- and methanotrophic bacteria with wood physico-chemical properties. Significant correlations (P < 0.05) are displayed in bold...... 149 Table 4.9: One-way analysis of similarity (ANOSIM) based on each two distance measures using relative abundance and presence/ absence data for different wood-inhabiting bacterial communities on two different tree species (Fagus sylvativca and Picea abies)...... 150 Table 5.1: Results of a simplified perMANOVA (interactions are excluded) analysis of Bray-Curtis dissimilarities in ARISA OTU (left) and sporocarp community structure in relation to tree species, exploratory decay class and and volume of the deadwood log, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseudo-F = F value by permutation, boldface indicates statistical significance at P < 0.05, P- values based on 999 permutations (lowest P-value possible is 0.001)...... 160

xxiii

List of abbreviations

AFLP Amplified fragment length polymorphism

(A)RISA (Automated) ribosomal intergenic spacer analysis

C Carbon

CWD Coarse woody debris

DGGE Denaturing gradient gel electrophoresis

N Nitrogen

RAPD Random amplified polymorphic DNA

OTU Operational taxonomic unit

(MOTU) molecular OTU (as used in phylogentic tree of chapter 3)

SIP Stable Isotope probing

T-RFLP Terminal restriction fragment length polymorphism

WIF Wood-inhabiting fungi

xxiv

1 Introduction

1.1 Deadwood in forest ecosystems

Deadwood is an important structural component, microhabitat and nutrient resource in all forest ecosystems (Harmon et al. 1986). There are various reasons why trees are dying, which in turn determine substrate qualities and hence the initiation and progress of wood decomposition. Tree mortality is driven by competition for growing space among trees, environmental stresses, pathogens, as well as intrinsically caused senescence of the organisms, or occurs suddenly due to ecosystem disturbances such as fires, windbreaks, snow load or simply for anthropogenic reasons such as forest management practices. According to Stokland et al. (2012) these mortality causes can be classified as biotic versus abiotic or as allogenic versus autogenic. However these classifications fail to account for the complex interactions among trees themselves. The physical and chemical qualities of deadwood determine the microhabitat and its spatial structure.

Together with local conditions related to surrounding land cover, geology, hydrology and wind conditions (habitat filters), these wood properties determine which organisms arrive and thrive on and into wood (Bässler et al. 2014; Boddy & Heilmann-Clausen

2008; Mouillot et al. 2013). The volume of deadwood in natural forests depends on site productivity (biomass production) and resulting deadwood input rate, the decomposition rate and the disturbances affecting the input rate (Harmon et al. 1986). In general the average deadwood volume is determined by a relation between input rate to decay rate

(Siitonen 2001). Deadwood volumes can make up to 145 m³*ha-1 in Fennoscandian

Norway spruce old growths stands (Siitonen 2001). Studies from the largest primal Eu-

1 Introduction ropean beech forest (Uholka-Shyrokyi Luh) in the Ukrainian Carpathians revealed mean deadwood amounts of 162 m³*ha-1 (Hobi 2013). The volume of deadwood in managed forests is of course lower and typically less than 10% of comparable natural forests

(Stokland et al. 2012). Especially large diameter deadwood is reduced or lacking, due to the timber extraction. However, it is exactly this large diameter deadwood, which is of tremendous importance in nature conservation issues. The diameter does not directly impact saproxylic species, but influences bark thickness and surface-to-volume

ratio (Lachat et al. 2013).

The log dimension and proportion also influence many biological factors involved in decomposition. In general, larger logs do also offer more ecological niches and micro- habitats (Fig. 1.1), and for example, strong correlations between beetle body sizes and log diameter have been described (Gossner et al. 2013). In contemporary forest management (also in commercial forests) different thresholds of deadwood volumes have

1 Introduction been established to ensure nature conservation aspects of various organism groups. For example polypore fungus Antrodiella citronella has been shown to spread from two relict habitats to the whole area of the Nationalpark “Bayrischer Wald”, after an increase of the total deadwood volume > 134 m³*ha-1 (Bässler & Müller 2010).

1.2 Wood decay by fungi

Among the organisms depending on deadwood, fungi serve as main decomposers, due to their capability to break down the cell wall that includes highly complex biopolymers. The process of microbial wood decay is related to the degradation of polysaccharides and the biopolymers cellulose, hemicellulose and lignin, which are the components of plant cell walls (Fig. 1.2). Based upon the type of chemical and structural components which are firstly degraded, one distinguishes between three types of fungal- caused wood rot: brown rot, white rot and soft rot.

Fig. 1.2: Illustration of a cell wall model and its five layers: ML = middle lamella, PW = primary wall, S1 = outer secondary wall, S2 = middle secondary wall, S3 = inner secondary wall. Figure was kindly provided by Prof. Dr. F. Schwarze (Schwarze et al. 2000).

1 Introduction

Brown-rot is exclusively caused by fungi of the phylum Basidiomycota and, with some exceptions, restricted to coniferous tree species (Hibbett & Donoghue 2001). Among these basidiomycetes, the Polyporaceae contribute most (Schwarze et al. 2000). How- ever, only 6-7% of the wood-inhabiting fungi are known to cause brown-rot (Baldrian

2008; Schwarze et al. 2000). Brown-rot fungi degrade the celluloses and hemicelluloses by initially releasing oxalic acids, which sequester iron that is bound in the cell wall and then enter into the S2 layer (Stokland et al. 2012). Here, a mechanism relying on the production of hydroxyl radicals, which are generated via the Fenton reaction2, takes place (Arantes et al. 2012; Hatakka & Hammel 2011; Jellison et al. 1997). These highly reactive hydroxyl radicals than tackle and break up the cellulose and hemicellulose molecules, which leaves the lignin behind and results in the characteristic brownish, brittle cubicles.

White-rot on the other hand is caused by various fungi, also including Xylariceae of the

Ascomycota, that degrade all cell wall components (including lignin) by secreting a sophisticated set of enzymes. Lignin, which makes up to 40% of the tissue dry weight in woody plants, is a branched biopolymer (Baldrian 2008) that contains (in ratios depending on the tree species) three different aromatic components (monolignols): p-cumaryl- alcohol, coniferyl and synapyl alcohol. Two forms of white-rot are distinguished: simultaneous white-rot and selective white-rot. In so called “simultaneous white-rot”, which is commonly observed in wood of deciduous tree species, celluloses, hemicelluloses and lignin are degraded at comparable rates. Fungal hyphae enter the S3 layer of the cell wall via the lumen and secrete enzymes in their direct vicinity (Schwarze et al. 2000).

2 Fenton reactions/ reagents are based on reactions between peroxides (typically H2O2) and iron ions in an acid medium. It is a chain mechanism that contains two reaction phases where firstly hydroxyl radicals 2+ 3+ 3+ are formed due to the oxidation of Fe to Fe . And afterwards reactions between Fe and H2O2 or HO2 take place, which form a cycle that sustains itself by regenerating Fe2+ (Arantes et al. 2012). 4

1 Introduction

During “selective white-rot”, lignin is degraded preferentially via direct attacks in the secondary wall and the middle lamella (Stokland et al. 2012). Celluloses and hemicellulose remain for the present unchanged, which results in the appearance of lighter/ white zones.

Of corresponding importance for the degradation of lignin are the extracellular oxidative enzymes laccase (Lac, EC 1.10.3.2), manganese peroxidase (MnP, EC 1.11.1.13), manganese independent peroxidases (MiP) lignin peroxidase (LiP; EC 1.11.1.14) and versatile peroxidase (VP; EC 1.11.1.16), that are capable of oxidizing the recalcitrant lignin polymer. Lignocellulose in woody material and leaf litter is ubiquitous in forest ecosystems and fungi have to overcome this lignin barrier to gain access to the energy- rich polysaccharides (Kellner et al. 2014). The expression of these enzymes has been addressed in various studies (Christ et al. 2011; Kellner et al. 2009; Kellner et al. 2010;

Kellner et al. 2014; Theuerl & Buscot 2010). Transcriptomic studies targeting the expression of the mentioned wood attacking fungal enzymes now enable scientists to link and synthesize information on diversity and community structure with ecosystem functions (Edwards et al. 2011).

The third main type of rot that also is known to occur on living trees is the so-called soft rot, which is mainly caused by ascomycetes and broadly attributed to members of the

Xylariceae (e.g. Kretzschmaria deusta). Soft-rot causing fungi mainly degrade celluloses and hemicelluloses of the secondary cell wall, which causes the spongy consisten- cy of the wood (Schwarze et al. 2000; Stokland et al. 2012). Hyphae grow in the lumen of cells, and then produce fine penetration branches growing through the thin S3 layer to gain access to the thick S2 layer. (Deacon 2009).

1 Introduction

1.3 The role of fungi in wood-decaying processes

As mentioned in the previous section, wood-inhabiting fungi play crucial roles in the decomposition of deadwood. They have evolved upon the existence of plants that serve as an important energy source. Fungi enter woody tissue either through wounds or infections of the roots. Through their presence as spores in the atmosphere or even as mycelial filaments in the surrounding soil, fungi are always close in the vicinity of living, dying and deadwood. Hendry et al. (2002) and Baum et al. (2003) verified latent infections of living trees by endophytic fungi that turn to act as saprotrophs at a later stage.

Examples of these fungi are Fomes fomentarius and Hypoxolon fragiforme, both common white- or soft-rot causing agents in wood of Fagus sylvatica. These studies revealed that woody tissue contains diverse fungal infections without being immediately sanitarily attacked. Wood-decaying fungi can be divided into parasitic and saprotrophic taxa. Obligate parasites grow and reproduce only in living plants and rely on living cells nutrient acquisition (biotrophy), while facultative parasites on the other hand live a certain time of their live as parasites and later turn to act as saprotrophs (necrotrophs/ per- thotrophs).

The way fungi colonize and retain “territories” in deadwood depends on various attributes and characteristics. The resulting successional patterns in the course of wood decay depend on dispersal modes, as well as competitive abilities and adaptions to substrate, disturbance and stress. A very detailed summary of ecological strategies and life-history traits of basidiomycetes is given by Boddy and Heilmann-Clausen (2008). They classify fungi into three catagories: Primary fungi are involved in lignocellulose decomposition prior to colonization by others. These early colonizers are tolerant to stress and include root pathogens wound parasites and canker formers. Secondary fungi are involved in

1 Introduction decomposition after early colonization. These fungi are tolerant to stress and strong competitors for growing space. The last category, tertiary fungi, comprises fungi that are not primarily involved in wood decomposition, and enter the wood from soil and litter or act as mycoparasites on other fungi. The concept of life-history traits is based on the idea that fungi are influenced by three environmental aspects and therefore have developed different life strategies. There are a) the “c-selected” – fungal traits that are determined by occurrence of competitors, b) the “s-selected” – comprising fungi that are resistant to certain stresses (e.g. pH, wood moisture) and c) “r-selected” – ruderal strate- gy, encompassing fungi that are able to immediately react to environmental disturbances

(Boddy & Heilmann-Clausen 2008).

A last aspect to be mentioned, concerns fungal succession and community composition patterns. Fungal succession is described as the “sequential occupation of the same site by thalli (normally mycelia) either of different fungi, or of different associations of fungi” (Rayner & Todd 1979). Kreisel (1961) divided succession into three phases, starting with an “initial phase” that is characterized by a few, but constant (substrate-dependent) species. After 13 – 31 months the “optimal phase” sets in, which is determined by a pool of characteristic species. The “final phase”, where wood decay has already progressed, is characterized by high species richness. All three successional stages are characterized by a representative composition or occurrences of specific indicator species (the German equivalent term is “Differentialarten”) and the time since the tree is dead and degraded, but the stage of decay is not considered (Volkenant 2007).

1.4 The role of bacteria in wood-decaying processes

Most studies concerning microbial diversity in deadwood in relation to degradation processes focus on higher fungi exclusively (Heilmann-Clausen & Christensen 2003; 7

1 Introduction

Kubartova et al. 2012; Müller et al. 2007; Rajala et al. 2012). The role of bacteria and their contributions in wood decay is largely unexplored or at least of technical interest only (concerning conservation of treated wood; (Eaton 1994)). Greaves (1971) firstly developed a concept concerning a functional classification of wood-inhabiting bacteria: with (1) bacteria that affect permeability but do not cause losses in material strength, (2) bacteria that attack wood structures, (3) bacteria that act as integral synergistically members of the total microflora and (4) the “passive” bacteria, which may act as antagonists to other bacteria. Bacteria are numerous and ubiquitous in practically all environments and are capable to colonize wood under aerobic and anaerobic conditions

(Clausen 1996). They have been detected in waterlogged logs under oxygen depletion, which would be unfavorable to most fungi (Clausen 1996; Kim et al. 1996). These so- called “erosion bacteria”, that are known since the 1980s (Daniel & Nilsson 1998) are even able to attack the cell wall. Several aspects of bacterial-functional classification have been captured over the last decades but information on taxonomical assignments are still lacking. Nilsson and Björdal (2008) were for example able to isolate erosion bacteria from wood piles, but could only speculate upon their potential taxonomical relationship to Cytophaga sp. (Bacteroidetes), based on their known ability to secret cellulases. Another phenomenon in which bacteria utilize dissolved sugars in the sap- wood but do not cause structural losses (Stokland et al. 2012) was described as “bacterial wetwood” (Schmidt & Liese 1994). A further functional classification refers to tunneling bacteria (Nilsson & Daniel 1983). These bacteria also tolerate a wide range of temperature and humidity conditions (Singh 2012).

A last functional aspect is concerning the interactions between fungi and bacteria, which can range from predatory or competitive to mutualistic/ synergistic (Clausen 1996; de

Boer & van der Wal 2008). A case where fungi impact negatively on bacteria was re- 8

1 Introduction ported by Folman et al. (2008), who observed decreased total numbers (mitospore counts) and abundances of bacteria in wood blocks after these had been colonized by

Hypholoma fasciculare and Resinicium bicolor. Wood-inhabiting fungi are able to re- duce pH by secreting e.g. oxalic acids, which can be unfavorable for certain bacteria (de

Boer et al. 2010). Wood-colonizing bacteria could also serve as nitrogen source for predatory fungi that feed on them and compensate temporary carbohydrate diets (Barron

2003). This proposal may even be extended to potential mutualistic interactions between wood-inhabiting fungi and N-fixing bacteria. Cowling and Merrill (1966) firstly hypothesized that associations with N-fixing bacteria may enable wood-inhabiting fungi to meet their substantial N requirements, which cannot be met by the N-poor wood substrate alone. The actual N-fixation in deadwood has been proven repeatedly since the

1970s by acetylene reduction assays (Seidler et al. 1972; Spano et al. 1982) and Aho et al. (1974) firstly isolated species of the genera Enterobacter and Klebsiella. Brunner and Kimmins (2003) were among the last to measure N-fixation rates in coniferous deadwood of different decay classes. They discovered a general increase with the progress of decay that can be predicted by wood moisture. As first described (de Boer & van der Wal 2008) and later published with accompanying data by Folman et al. (2008) and Valaskova et al. (2009), a fungal-bacterial co-colonization experiment (see above,

Folman et al. 2008), revealed that among the reduced number of bacteria, a large proportion (59%) were affiliated to Alphaproteobacteria and among them 25% to the order

Rhizobiales. Owing to the presence of N-fixing methylotrophic bacteria (Methylocapsa acidiophilia) they further hypothesized that these species rely on the methanol that is released by fungal ligninolytic activities and in turn provide N to the fungi. A complete model of N flow in decomposing wood must probably also account for other sources of

N, such as atmospheric depositions, import through hyphae from surrounding soil, recy- 9

1 Introduction cling of N within hyphal networks and from decomposing sporocarps as well as input through feces and carcasses of animals (Frey-Klett et al. 2011).

1.5 Approaches to assess microbial diversity

Microorganisms are the most diverse and abundant group of organisms on earth and it is estimated that soil contains 4-5 × 1030 microbial cells (excluding viruses) (Singh et al.

2009). But interestingly less than 1% of this tremendous diversity is assumed to grow under laboratory conditions (Singh et al. 2009), which makes it hardly impossible to identify the respective organisms using culture-dependent methods. On the opposite, many surveys on fungal diversity and community composition and assembly histories have been conducted by sporocarp sampling and microscope identifications (Blaser et al. 2013; Müller et al. 2007; Ódor et al. 2006), and generally traditional mycologists know well which substrate a fungus (as sporocarp) requires. The identification of wood- inhabiting fungi has long been restricted to traits of their sexual and sometimes asexual reproductive structures. Especially the higher Ascomycota and Basidiomycota form mushroom sporocarps. With progresses and developments of molecular techniques it became possible to improve the power of ecological field studies, because these new tools enable us to detect vegetative structures of fungal organisms such as spores and mycelia even inside the wood and in the absence of any sporocarp production, just by extracting and analyzing their DNA. Moreth and Schmidt (2000), Jasalavich et al.

(2000), Vainio and Hantula (2000), Adair et al. (2002) and later Raberg et al. (2005) and Allmer et al. (2006) conducted the first molecular studies that screened wood- inhabiting fungal communities in deadwood. Since then there have been many methodological improvements and now a broad spectrum of techniques is available for that purpose. DNA-fingerprints on gels or capillary sequencers detect different microbial 10

1 Introduction organisms by length polymorphisms of the targeted or randomly amplified DNA fragment (RISA, ARISA, T-RFLP, AFLP, RAPD) or by base composition of the amplicon

(DGGE). These techniques are fast and quite easy to apply. Their application is also inexpensive but they all lack information on taxonomic identity, which could be critical for certain scientific questions, as different fungi may not be fully redundant in their ecological function (further details: Anderson and Cairney (2004)). If performed on gels such as in the original RISA or on gradient gels, excising, cloning and sequencing from bands is an optional advantage that can generate additional information. Co-migration of amplicons, the limited number of size bins that may quickly be saturated and size heterogeneity of the target within taxa make such fingerprint methods rather good in detecting shifts in not too complex systems and as a pre-selection before sequencing.

Clone library sequencing has served for a long time as a standard in systematics and ecology and is still useful in environmental investigations, if the number of samples is restricted. Microbiological surveys have undergone a revolution within the last decade due to the development and rising accessibility of high-throughput DNA sequencing

(Hall 2007). Several Next-generation sequencing platforms are available, all able to capture millions of sequence reads per run at high accuracy rates (Mardis 2008). The

454 pyrosequencing platform was launched and published in 2005 by 454 Life Sciences

(Rothberg & Leamon 2008) as a time and cost reducing alternative to capillary electrophoresis sequencing based on the Sanger, dye- dNTP chain interruption method (for comprehensive details check Margulies et al. (2005)). Pyrosequencing on the ITS barcode marker approach avoids the cloning biases (Liang et al. 2011) and allows for the simultaneous multiplexed analysis of large numbers of samples. In spite of technical improvements of the methods themselves, drawbacks to this are still the relatively short sequence reads of the most modern NGS (Next-generation sequencing) metagenomic 11

1 Introduction methods, database limitations, disconnects between active (better addressable by meta- proteomics and rRNA metatranscriptomics) and inactive fungi, intragenomic marker variability, difficult taxonomic resolution between so-called OTUs and named species, and the lack of trait databases that could provide information on ecological functions of a detected fungus. These NGS methods also allow limited human control over bioinformatics routines. They are still relatively costly and not free of generating technical errors as well (e.g. generating chimeras).

What all these molecular methods have in common is that their quantitative information is limited because the usually targeted rDNA marker is a multi-copy gene with some copy number fluctuation and copy variability within genomes. For information on specific activity, methods that incorporate stable isotopes or mRNA of functional genes may be more useful (Pinto-Tomás et al. 2009; Strickland & Rousk 2010).

All molecular techniques have evolved upon a few molecular markers that serve as standards in both microbial environmental and taxonomy-based studies that are now backed-up by vast databases (Silva, RDP, UNITE, GenBank). Ribosomal DNA marker genes and genome regions, including 16S/18S rDNA, 23S/28S rDNA and ITS region surrounding the eukaryotic 5.8S rDNA, have long been used successfully in fungal and bacterial diversity studies (Buée et al. 2009; Nacke et al. 2011; Will et al. 2010; Wubet et al. 2012). The ITS1-5.8S-ITS2 became the most commonly used genetic marker for fungal barcoding (e.g. Seifert (2009)), species identification, molecular systematics and diversity studies (Kubartova et al. 2012; Ovaskainen et al. 2013; Rajala et al. 2012), due to its high variability and taxonomic resolution power. The highly conserved primer binding sites of bacterial 16S rRNA regions (V1-V6) provide amplicons with species- specific nucleotide orders, which are suitable for identification.

1 Introduction

The present thesis contains data that are derived from both, traditional clone library sequencing (chapter 3), and 454 pyrosequencing (chapters 2 and 4).

1.6 The impact land use on biodiversity

Changes in land use are one of the major determinants altering biodiversity (Sala et al

2000) and have been addressed in various studies in the past years (Foley et al. 2005;

Lambin et al. 2001). Potential mechanisms that determine ecosystem responses and biological diversity to such environmental changes are not well known to date and of broad interest. A large portion of biodiversity in forest ecosystems is associated with deadwood (Stokland et al. 2012). Forest management practices, such as forest type con- versions and harvesting are common activities that determine tree species composition, stand density and structure, as well as resulting qualities and quantities of deadwood

(Paillet et al. 2010; Purahong et al. 2014). As a result of these anthropogenic interven- tions, roughly 60% of the German forests are dominated by coniferous species

(~66,000 km²), whereas 75% of the whole German countryside (~266.000 km²) would potentially naturally be covered by European beech (Fagus sylvatica) (BMELV 2011).

Purahong et al. (2014) recently detected significant decreases in fungal species richness and abundances of certain important wood-inhabiting fungi, when forests had been converted from beech to conifers.

The Biodiversity Exploratories were set up in 2006 and are funded by the German Sci- ence Foundation (DFG) “as an initiative to advance biodiversity research in Germany”3 and is a large-scale and long-term open research platform (Fischer et al. 2010). They are a system of not only observational but also sophisticated experimental plots in three geographically distant regions of Germany: (1) UNESCO Biosphere reserve

3 http://www.biodiversity-exploratories.de/1/aims/ 13

1 Introduction

“Schwäbische Alb” in Southwestern Germany, (2) the National Park “Hainich-Dün” in

Central Germany, and (3) the UNESCO Biosphere reserve “Schorfheide-Chorin” in

Northeastern Germany (Fig. 1.3). The main aims of the Biodiversity Exploratories are to gain comprehensive knowledge on the relationships between biodiversity of different taxa and levels and on the impact of land use practices and disturbance on these relationships including the consequences in terms of ecosystem functioning and service providing. Each exploratory encompasses grassland and forest plots of different study intensity levels. 1000 plots per exploratory were initially assigned according to a grid design and land-use types. Among 1000 grid plots per exploratory, 100 experimental plots (EP) were selected for broader environmental observations and manipulations.

Among them 18 very intensively investigated plots (VIP) (9 in forests and 9 in grasslands), were chosen for detailed research, including studies that also often require intensive and expensive analytical work in the laboratory.

This dissertation relies on a subset of samples of 48 deadwood logs of the “Schwäbische

Alb”-exploratory that was gained during a larger sampling campaign (194 deadwood logs) across all 9 forest VIPs in 2009. The mean annual temperature in this region is 6–

7°C and annual precipitation ranges between ca. 700 and 1000 mm (Fischer et al. 2010).

All plots are located at an altitude between 715 and 809 m and represent the following three forest management types: (i) extensively managed/ unmanaged beech forests, where timber harvesting stopped several decades ago, (ii) managed beech forests dominated by Fagus sylvatica and (iii) managed spruce forests dominated by Picea abies.

1 Introduction

Fig. 1.3: Overview of three German Biodiversity Exploratories. In detail: topographic map of the “Schwäbische Alb”, including all grassland and forest at the EP and VIP level.

1.7 Research objectives and hypotheses

This thesis and its investigations were conducted in the context of the DFG-funded

(German Science Foundation) project “FunWood”, which was made up by Dr. Tiemo

Kahl and Prof. Dr. Jürgen Bauhus (University of Freiburg).

The initial intention of this thesis was to study and compare the fungal diversity in deadwood of differently managed (according to the overarching design of the Biodiver- stiy Exploratories) forest plots by applying traditional sporocarp monitoring and DNA- based fingerprint techniques. This combination was thought to efficiently capture the hidden non-observable fungal species that reside inside the deadwood logs as well as the physiologically active, growing ones, which develop sporocarps during the seasons.

I expected an increase in diversity of wood-inhabiting fungi with increasing proportion of beech trees in the investigated forest stands. Conversely, high timber extraction rates were expected to impact negatively on fungal diversity in deadwood. I furthermore hypothesized that fungal diversity leads to a higher variation of wood decay mechanisms and therefore to enhanced and accelerated overall deadwood decomposition.

1 Introduction

Initial cursory sequencing of bacterial 16S rDNA and the observation of enormous sporocarps on some logs led me to the question how these fungi meet their requirements for nitrogen, which is restricted in deadwood with unfavorable C:N ratios that can be as low as 800:1 (own data, Spano et al. (1982)). By conducting a literature survey on N- fixation on deadwood, it became obvious, that this issue has been addressed by various scientists, since the early 1970s. However, recent investigations, especially on the molecular/ gene level were lacking. Furthermore, little is known about the general bacterial diversity and community structure in deadwood. To date, there has been only one study that assessed the 16S bacterial diversity of one decaying Pinaceae species in China

(Zhang et al. 2008).

As mentioned above, with the continuing rapid advancement in molecular environmental mircrobiology, I was able analyze a subset of 48 deadwood logs (24 each of Fagus sylvatica and Picea abies, respectively) using 454 pyrosequencing. By applying this method, I was able to link taxonomically resolved information on resident fungi and bacteria with microbial-mediated ecosystem services and functions.

This thesis is divided into three chapters: Chapter 2 is addressing fungal diversity, community structure and taxonomically-resolved identity in relation to wood physico- chemical properties, lignin-modifying enzyme activity and their consequences on decomposition rates. This study is in line with two recent Fennoscandian investigations

(Kubartova et al. 2012; Ovaskainen et al. 2013) but expands our knowledge due to the exclusive comparison of two very different tree species and because it is the first study of this kind outside the boreal zone. The study further considers community structure and diversity patterns and detects key players causing high enzyme activities and fast wood decay. Chapter 3 reports on a study that details the highly diverse bacterial nifH genes present in deadwood. The study further reveals non-random co-occurrence pat- 16

1 Introduction terns of the nifH genes with sporocarp fungal diversity. Chapter 4 is devoted to the bacterial diversity and community structure from 16S 454 pyrosequencing. It is the first

“deep-inside” approach that gives a descriptive reflection of which bacteria (taxonomically-resolved) are present and how the community structure is related to wood physico- chemical properties. This study was also conducted to corroborate the distinctive patterns of nifH genes.

Fig. 1.4: Overview of the working packages of this thesis: (1) How forest management practices and wood physico-chemical properties change forest stand structure and correspond to fungal and bacterial communities and (2) how do enzyme activities influence wood decay by fungi and N-fixation by bacteria. (Modified from Krüger et al. (unpublished))

An illustrative overview encompassing the three working packages is given in Fig. 1.4.

All three chapters (chapters 2-4) are equating to single manuscripts that have been either published or are currently in revision or submitted at scientific journals. They all are concluded by a broad discussion concerning the respective objectives and hypotheses addressed in each of these single studies. The final synthesis/ discussion at the end of 17

1 Introduction this thesis intends to link main aspects and research objectives of the three manuscripts and therefore partly reconsiders previously detailed points of discussion.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2 Linking molecular wood-inhabiting fungal diversity and

community dynamics to ecosystem functions and

processes in Central European forests

2.1 Introduction

Deadwood is one of the most important organic carbon pools in forest ecosystems

(Floudas et al. 2012). Due to its lignin content of 15-40%, deadwood is rather difficult to decompose and is therefore an important temporal store of carbon and macronutrients

(Kopra & Fyles 2005; Krankina et al. 1999; Sarkanen & Ludwig 1971). Microorgan- isms, mainly fungi, play crucial roles in forest ecosystems. Their diversity and enzymatic activities constitute the basis for the food-webs in wood and litter (Pollierer et al.

2012; Stokland et al. 2012). Filamentous fungi of the phyla Basidiomycota and (to a lesser extent) Ascomycota are particularly important in lignin mineralization. White-rot fungi secrete a set of extracellular oxidative enzymes such as laccase (Lac, EC

1.10.3.2), manganese peroxidase (MnP, EC 1.11.1.13) and/or manganese independent peroxidases (MiP) to oxidize the recalcitrant lignin polymer (Hofrichter et al. 2010). In contrast, brown-rot fungi do not secrete these enzymes. Instead, they oxidize lignin via a mechanism relying on hydroxyl radicals, which are generated via the Fenton reaction

(Hatakka & Hammel 2011; Jellison et al. 1997; Martinez et al. 2005).

The decomposition processes driven by decomposer organisms are complex and influenced by the host tree species and environmental factors (Boddy 2001). Different fungal

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests species have different capacities for wood decomposition (Valmaseda et al. 1990) and have further been reported to dependently related to different tree species (Rajala et al.

2010). Moreover, the identity of the host tree determines several key abiotic substrate factors that influence the interactions between the resident microbiota along with their growth, reproduction, and metabolism (Gadd 2010; Kögel-Knabner 2002). These factors include the chemical properties and cell structure of the wood (Schwarze et al.

2000) as well as its pH and water content.

The existing ecological data on wood-inhabiting fungi (WIF) in Central European forests, including information on their diversity patterns, resource use and the determinants of their community structure, was primarily obtained via sporocarp surveys (Blaser et al. 2013; Heilmann-Clausen & Christensen 2003; Müller et al. 2007), which only record the composition of a portion of the actively reproducing fungal community at a specific point in time. However, the macroscopically observable fungal flora may not be fully representative of the fungal community that is present, especially if sampling is only conducted once or over a short period of time (Halme & Kotiaho 2012). In contrast, high-throughput sequencing allows for detailed analysis of community composition and may thus uncover a hitherto concealed fungal diversity residing in deadwood. This approach has been used to study WIF communities within and among Norway spruce

(Picea abies) logs (Kubartova et al. 2012) and to test the link between fungal life history and population dynamics (Ovaskainen et al. 2013). As expected, these studies revealed much higher fungal species richness than previously published sporocarp surveys and suggested that highly abundant fruiting species may be only weakly representing the fungal community as a whole.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

All of the European high-throughput sequencing based studies on WIF conducted to date focused on deadwood of either gymnosperm tree species, Picea abies, in a single biome: the boreal forest zone of Northern Europe (Kubartova et al. 2012; Ovaskainen et al. 2010; Ovaskainen et al. 2013) or two angiosperm tree species, Fagus sylvatica and

Quercus robur in temperate forests (Hiscox et al. 2015; van der Wal et al. 2015). There- fore, our information on community dynamics of WIF in deadwood of different tree species in the same forest ecosystem at different locations is still limited. Our investigation is also the first to explore the fungal diversity in deadwood of European beech under natural condition using contemporary massively parallel sequencing techniques. In line with recent findings of Ottosson et al. (2015), we also expected deadwood of Euro- pean beech and Norway spruce species to harbor a significant proportion of rare taxa, ones that are not presented through high sequence abundances. Links between fungal richness, taxonomically resolved community composition, and ecosystem processes are rarely studied, especially under natural conditions (van der Wal et al. 2015).

In this study, we aimed to: (i) using 454 pyrosequencing to compare the diversity and community structure of WIF in deadwood of two silviculturally important tree species found in Central European temperate forests - the coniferous Picea abies and the deciduous Fagus sylvatica, (ii) disentangling the ecological and environmental factors that correlate with WIF community structure, and (iii) using the resulting data to link WIF richness and taxonomically-resolved identity to microbial-mediated ecosystem functions (lignin-modifying enzyme activities) and processes (wood decomposition rates).

We hypothesized that the different wood physico-chemical properties of the two tree species could lead to significant differences in fungal richness and community structure.

The different physico-chemical parameters are linked to wood decay and how fungal

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests decomposition of the wood alters their own environment. Specifically during wood decomposition (decreasing of wood density), C/N ratio decreases as a result of N accumulation and at the same time, lignin and wood moisture increase. The degrees how wood physico-chemical properties as well as macro-and micro-nutrient changes may depend greatly on the initial properties of each tree species.

In this study the lignin-modifying enzyme activities (indicator for lignin decomposition) and wood decomposition rates (indicator for decomposition process) were used as a proxy for microbial-mediated ecosystem functions and processes, respectively. Some studies have shown that species richness was found to positively link or promote stability of the ecosystem function in soil systems, however this is still unclear in the case of deadwood-inhabiting fungi under field conditions (van der Heijden et al. 1998; Cole- man and Whitman 2005; Proulx et al. 2010; Eisenhauer et al. 2012). In a recent study, it was shown (under an artificial set-up) that fungal diversity was associated with wood decomposition rates in the intermediate decay stages, as determined by respiration rates

(Valentin et al. 2014). Since the loss of microbial biodiversity could alter ecosystem functioning and stability, we expected positive correlations between fungal richness and the activities of lignin-modifying enzymes and wood decomposition rates.

2.2 Material and methods

2.2.1 Experimental design, deadwood selection and sampling

The study was conducted on forest plots of the German Biodiversity Exploratories

(Fischer et al. 2010) located in the UNESCO Biosphere Reserve “Schwäbische Alb” in southwestern Germany. The plot selection criteria were based on forest history and management regimes, dominant tree species and deadwood status (Fischer et al. 2010;

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Hessenmöller et al. 2011 and Luyssaert et al. 2011). All selected forest plots had, apparently, been subjected neither to clear-cutting procedures nor to a period of agricultural use in the past (Luyssaert et al. 2011). A minimum distance between the outer edges of each plot is 200 m and located at least 100 m from the nearest forest edge. Our survey took place on deadwood logs in 9 intensively investigated 1 ha plots, with three plots representing the following three forest management types, respectively: (i) natural beech forests (unmanaged for 100 years, natural regeneration, uneven-aged forest structure, with mature trees > 100 years), (Hessenmöller et al. 2011), (ii) age-class managed beech forests dominated by Fagus sylvatica (natural regeneration, even-aged forest structure, 50 - 100 years) and (iii) age-class managed spruce forests dominated by Picea abies (planted forest, even-aged forest structure, 80 – 100 years) (Tables S2.2, S2.3). In

April 2009, a set of 48 logs, equally representing the two tree species (P. abies and F. sylvatica) located on the forest floor were randomly selected and their properties

(length, diameter, tree species, etc.) were characterized. Our selection assured that

Fagus logs were present in Picea-dominated plots and vice versa. In June 2009, 3-7 wood samples were taken from each log (according to its size) using a cordless Makita

BDF451 drill (Makita, Anja, Japan) equipped with a 2 x 42 cm wood auger as described in Hoppe et al. (2014, 2015) and Purahong et al. (2014a, b). Sporocarp data were available (Hoppe et al. 2014) and used as corroborative evidence for the presence of particular fungi that were detected as OTUs in the sequencing analysis.

2.2.2 Wood physico-chemical properties and lignin-modifying enzyme assays

The concentrations of C and N in wood samples were determined by total combustion using a Truspec elemental analyzer (Leco, St. Joseph, MI, USA). Klason lignin content was determined gravimetrically as the dry mass of solids remaining after sequential hy-

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests drolysis with sulfuric acid (72% w/w); in a second step, acid soluble lignin was measured by UV-photometry in 4% H2SO4 (Effland 1977; Liers et al. 2011). Total lignin was obtained by summing acid insoluble Klason lignin and acid soluble lignin (Raiskila et al. 2007). The wood samples’ pH values and contents of nutrient ions and lignin- modifying enzymes were measured in aqueous extracts. The extractions were performed using 10 ml distilled water per 1g dry mass of wood for 120 min on a rotary shaker (120 rpm). Macronutrients (Mg, K, Ca, Fe) and micronutrients (Cu, Mn, Zn, Ni) were determined using inductively coupled plasma (ICP) optical emission spectrometry (ICP-

OES) and mass spectrometry (ICP-MS), according to the instrument manufacturers’ specifications. Three oxidative extracellular oxidoreductases important for lignin degradation (laccase - Lac, manganese peroxidase – MnP, manganese-independent peroxidases - MiP) (Hatakka & Hammel 2011) were measured as described by Hahn et al.

(2013). Nutrient ion and lignin-modifying enzyme analyses were conducted in triplicate and in duplicate, respectively, on the same subsamples.

Deadwood logs were assigned to four decay classes based on remaining mass (%) data by k-means cluster analysis as described in Hoppe et al. (2014) and Kahl et al. (2012).

2.2.3 DNA isolation, PCR and pyrosequencing

Total community DNA was isolated from 1 g of each homogenized wood sample using a modified CTAB-protocol (Doyle & Doyle 1987) as described in Hoppe et al. (2014).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

All DNA extracts from the wood samples of each log were pooled into a composite extract prior to PCR. Fungal ITS rDNA amplicon libraries were produced as described in

Wubet et al. (2012). Briefly we used fusion primers designed with pyrosequencing primer B, a barcode and the fungal specific primer ITS1-F (Gardes & Bruns 1996) as a forward primer and pyrosequencing primer A and the universal eukaryotic primer ITS4

(White et al. 1990) as a reverse primer to amplify the fungal nuclear ribosomal internal transcribed spacer (nrITS) rDNA. We used a set of 10nt MID-barcodes provided by

Roche Applied Science (Mannheim, Germany). Each composite DNA extract for the amplicon libraries was amplified separately by PCR in triplicate 50 µl reaction mixtures containing 25 µl 2x GoTaq Green Mastermix (Promega, Madison, WI, USA), 25 µM of each primer and approximately 20 ng template DNA. Amplification was performed using a touchdown PCR program with denaturation at 95°C for 5 min followed by 10 cycles of denaturation at 94°C for 30 sec, annealing at 60–50°C for 45 sec (-1°C per cycle), and extension at 72°C for 2 min, followed by 30 cycles of 94°C for 30 sec, 50°C for 45 sec and 72°C for 2 min, with a final 10 min extension step at 72°C (Lentendu et al. 2014). The PCR products were separated on a 1.5% agarose gel and equimolar volumes of the amplified products of the expected size from the three positive replicate amplicons per sample were homogenized. The pooled products were gel purified using a Qiagen Gel Extraction Kit (Qiagen, Hilden, Germany). The purified DNA was quantified using a fluorescence spectrophotometer (Cary Eclipse, Agilent Technologies,

Waldbronn, Germany). An equimolar mixture of each library was subjected to unidirectional pyrosequencing from the ITS4 end of the amplicons, using a 454 Titanium amplicon sequencing kit and a Genome Sequencer FLX 454 System (454 Life Sciences/

Roche Applied Science) at the UFZ Department of Soil Ecology.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2.2.4 Bioinformatic analysis

We performed multiple levels of sequence quality filtering. The fungal ITS sequences were extracted based on 100% barcode similarity. Sequences were clipped of barcodes and trimmed to a minimum length of 300nt to best cover the ITS2 part of the nrITS using MOTHUR (Schloss et al. 2009). Sequence reads with an average quality score of <

20, and homo-polymers of > 8 bases were removed. Unique good quality sequences from the dataset were filtered and checked for chimeras using the uchime algorithm

(Edgar et al. 2011) as implemented in MOTHUR. To avoid sampling size effects, the number of reads per sample was normalized for each data set by randomly subsampling to the lower number of reads per samples using the subsample script as implemented in

MOTHUR. The sequence dataset was then clustered and assigned to OTUs using CD-

HIT-EST of the CD-HIT package version 4.5.4 (Li & Godzik 2006) at a 97% threshold of pairwise sequence similarity as in Wubet et al. (2012). We used MOTHUR to taxonomically assign representative sequences of the OTUs against the UNITE reference database (as downloaded in May 2013) using the default set-up (Abarenkov et al. 2010).

2.2.5 Statistical analysis

To link WIF richness and taxonomic identity to microbial-mediated enzyme activity and decomposition rates, we defined cumulative OTUs (cOTUs) as species synonyms by aggregating OTUs that were unequivocally given the same name by BLAST reanalysis against GenBank yielding at least 95% (species level) and 90% (genus level) maximum identity scores (Ovaskainen et al. 2013) for the same database species. The Chao1 and

ACE diversity indices were calculated for all OTUs (including 1-3tons) using the estimate function in the R package “vegan” (Oksanen 2013). All multivariate statistics were conducted on proportional abundance data using the WIF dataset excluding 1-3tons.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Analysis of similarities (ANOSIM) and nonmetric multidimensional scaling (NMDS) based on Bray-Curtis distances were conducted using PAST (Hammer et al. 2001) and the “vegan” package in R (Oksanen 2013), respectively, to compare the fungal community structure of Fagus and Picea. The influence of selected wood physico-chemical parameters, fungal family abundances and environmental factors on fungal community structure was investigated by fitting data on each factor to the NMDS ordinations of the fungal communities. The wood physico-chemical parameters considered in these analyses were decay class, concentrations of macronutrients (C, N, K, Ca, Mg, Fe) and micronutrients (Cu, Mn, Ni and Zn), relative wood moisture, wood density, remaining mass and pH. Goodness-of-fit statistics (R2) for environmental variables fitted to the

NMDS ordinations of fungal communities were calculated using the envfit function of

“vegan”, with P values being based on 999 permutations (Oksanen 2013). The P values were Bonferroni-corrected in all cases. We calculated non- parametric Kendall-Tau correlations (τ) (pairwise comparisons) to link the fungal taxonomic groups to lignin- modifying enzyme activities and decomposition rates using PAST. The differences in

OTU richness (observed OTU and cOTU richness and estimated Chao1 and ACE richness) and wood physico-chemical properties among different decay classes were analyzed for differences among means (P < 0.05) by performing one-way analysis of variance (ANOVA) incorporating Shapiro-Wilk’s W test for normality and Levene’s test to check for the equality of group variances. Fisher’s Least Significant Difference (LSD) post hoc test was also performed.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests 2.3 Results

2.3.1 Wood physico-chemical properties in different decay classes

The C/N ratios of the deadwood decreased as it decayed, and were significantly higher in Picea logs (P <0.0001, ranging from 630 ± 48.4 to 423 ± 52.4) than in Fagus logs

(365 ± 14.9 to 194 ± 15.6) (Fig. 2.1 and Table S2.4). This difference can be attributed to the significantly higher C concentrations in Picea deadwood compared to Fagus

(P < 0.0001), which ranged from 49.3% ± 0.31 in decay class 1 to 51.4% ± 0.94 in decay class 4, and also to the significantly higher N concentrations in Fagus logs

(P < 0.0001) compared to Picea. Nitrogen concentrations also increased (P < 0.0001 for

Fagus logs, P < 0.01 for Picea logs) as wood decay progressed (Fig. 2.1). The mean total lignin concentrations in Fagus deadwood were lower than in that of Picea (28.9% and 36.0%, P < 0.0008) and the relative proportion of lignin in the wood increased significantly as it decayed. There were no differences in the mean relative wood moisture, mass loss and decay rate between the two tree species, but these parameters differed significantly between different decay classes within each tree species (Fig. 2.1). For example, relative wood moisture increased in parallel with decay, from 49.8% ± 5.5 for decay class 1 to 155.2% ± 9.1 for decay class 4 in Fagus (P < 0.0001) and from 48.7%

± 11.6 to 163.1% ± 24.6 in Picea (P < 0.0001). The pH of Fagus deadwood was constant across decay classes and significantly higher than that of Picea deadwood, which declined from 4.6 ± 0.1 for decay class 1 to 4.3 ± 0.2 for decay class 4. The allocation of micronutrients and the activities of the lignin-modifying enzymes laccases (Lac), manganese peroxidase (MnP) and manganese independent peroxidases (MiP) are provided in supplementary Table S2.5.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. 2.1: Decay rates, mass losses, and wood physico-chemical properties of Fagus and Picea deadwood logs of different decay classes. The figure shows the relative deviation from the sample mean in each case. Mean values of all parameters for deadwood logs of bot both species are shown in the middle. The differences between the two deadwood species and also among different decay classes for individual deadwood species were analyzed by t-tests and one-way analysis of variance, incorporated in the form of Fisher's Least Significant Difference (ns = not significant, * P < 0.05, ** P < 0.01, *** P < 0.001). The four FASY (Fagus sylvatica) and PIAB (Picea abies) decay classes are shown in different color shades.

2.3.2 Wood-inhabiting fungal richness and community structure

The total observed fungal OTU richness (excluding rare taxa) per sample ranged from

17 to 102 (25-159 including rare taxa) in logs of Fagus sylvatica and from 28 to 102

(38-151 including rare taxa) in Picea abies logs (Fig. S2.4). We did not observe signifi-

cant variation in mean OTU richness between the two tree species (P = 0.32) (Table

S2.6). All four measures of fungal diversity/ richness (total observed, cumulative,

Chao1, and ACE) correlated significantly and positively with decay class, whether it

was quantified in terms of declining wood density (P < 0.05 Fagus sylvatica, P < 0.001

on Picea abies) or remaining wood mass (Fig. S2.5).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

While the distribution of OTUs belonging to the Basidiomycota or Ascomycota was comparatively balanced in terms of their presence or absence, their relative abundances differed significantly. For example, the 201 Basidiomycota OTUs identified in Fagus samples contained 25,053 sequences (63.2% of all sequences) whereas the 265 Asco- mycota OTUs (Table S2.7) only contained 14,091 (35.5%). There was an even more pronounced pattern in Picea, where the 217 identified Ascomycota OTUs contained only 5,614 (13.3%) sequences whereas the Basidiomycota accounted for almost 83%

(34,985) of all sequences clustered into 242 OTUs.

More specifically, 2,487 (6.35%) sequences were assigned to Annulohypoxylon cohaerens, a common European beech saprotroph of the family Xylariaceae, which also represented the dominant family in the Fagus logs, accounting for 5,377 sequences in total (Figs. 2.2 and S2.6). Other important families in the Fagus fungal communities were the Meruliaceae, Polyporaceae, Mycenaceae, Physalacriaceae and Marasmiace- ae, that together accounted for 42% of all identified fungi (Fig. 2.2). OTUs assigned to the white-rot causing fungus Resinicium bicolor, which is listed as Agaricomycetes I.S.

(incertae sedis) in the Index Fungorum (www.index-fungorum.org), were dominant in

Picea deadwood, accounting for 25.3% of all sequences. This fungus was detected in

83.3% of all Picea logs (Fig. S2.6). Bondarzewiaceae, represented by Heterobasidion sp., Stereaceae, Fomitopsidaceae and Mycenaceae also occurred frequently and accounted for 32.64% of the identified sequences/ OTUs. Unidentified species of the Hel- otiaceae and an OTU identified as the potential soft-rot agent Phialophora sp. of the family Herpotrichiellaceae were the most abundant Ascoymcota in Picea but only accounted for 2% and 1.9% respectively of all identified sequences.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

We also examined the sharedness of WIF communities between tree species at species

(cOTU) and genus level (Fig. S2.7ab). Among the 160 cOTUs with the same genus and species epithets, 74 (22 Ascomycota, 50 Basidiomycota, and 2 zygomycetes) were found in both, Picea and Fagus logs (Fig. S2.7a). Genus-level comparisons revealed that 35 genera were unique to Fagus and, 34 to Picea, while 58 genera were shared

(Fig. S2.7b).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

tive tive relative abundances of of of correct identification, respectively, based on secondary

deadwood displayed as sunburst charts (beatexcel.com). The outer rings show the cumulative relative abundances

Picea and

. Only cOTUs that accounted for > 1% totalof the abundances and identified at least to are genus labeled. Fagus terns terns of cOTUs on -analysis. The middle rings display the cumulative relative abundances at the family level. The inner rings display the cumula : : Pat 2 . 2

Fig. Fig. of different fungal OTUs (cOTUs) at speciesthe and genus levels at least with 95% and 90% probability o BLAST re different concordant fungal phyla 32

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2.3.3 Pyrosequencing and community sampling statistics

In total, 139,352 reads were obtained from 454 pyrosequencing of 48 deadwood samples. Sequences were initially quality checked, trimmed, normalized per sample and screened for potential chimeras (841 chimeras were removed). CD-HIT clustering of the remaining 86,935 sequences yielded 2,386 OTUs at a 97% cutoff, of which 1,090 appeared as singletons and 305 as doubletons. Singletons, doubletons, and tripletons were removed from the final dataset, which was considered as having the “rare” taxa excluded. We performed a Mantel test on Bray-Curtis dissimilarities to assess the correlations between the whole matrix and a matrix excluding the rare taxa as just stated. This indicated that the removal of rare taxa from the community composition had no effect

(R = 0.99, P = 0.0001). In total, 81,803 sequences clustering into 779 OTUs were retained for further statistical analysis after removing sequences that could not be taxonomically assigned to fungi. By applying a species-level sequence divergence threshold of 3%, we were able to taxonomically assign 93.1% of the filtered OTUs at the phylum level. Basidiomycota accounted for 338 of the filtered OTUs, Ascomycota for 375, and zygomycetes for 9. Chytridiomycota, Entomophthoromycota, and Glomeromycota were represented by 1 OTU each. Further, 614 (78.8%), 552 (70.9%) and 434 (55.7%) of the filtered OTUs were classified at the order, family and genus levels, respectively. The remaining 6.9% (54 OTUs) were grouped as unknown fungal OTUs.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2.3.4 WIF dynamics on deadwood of different tree species

Different WIF dynamics were clearly observed in the Fagus and Picea logs (Figs. S2.8,

S2.9). WIF communities in Fagus logs were highly dynamic with respect to wood decay, with no cOTU being dominant in all decay classes. The dominant fungal families and cOTUs in Fagus logs of decay classes 1-4 were Xylariaceae (mainly Annulohy- poxylon cohaerens) and Polyporaceae (Trametes versicolor) in decay class 1, Polypo- raceae (Fomes fomentarius) and Meruliaceae (Phlebia livida) in decay class 2, Meru- liaceae (Ceriporiopsis gilvescens), Polyporaceae (Fomes fomentarius) and Maras- miaceae (Megacollybia platyphylla) in decay class 3, and Mycenaceae (Mycena alnetorum), Marasmiaceae (Megacollybia platyphylla, Marasmius alliaceus) and Polypo- raceae (Trametes versicolor) in decay class 4. In contrast, all of the Picea decay classes were dominated by a single cOTU assigned to Agaricomycetes I.S. (Resinicium bicolor); the mean abundances of this cOTU in decay classes 1, 2, 3, and 4 were 26.1%,

36.6%, 27.6%, 16.5%, respectively. Some fungal families and cOTUs were also co- dominant in different decay classes: Stereaceae (mainly Amylostereum areolatum) in decay class 1, Fomitopsidaceae (Fomitopsis pinicola) in decay class 3, and Bondar- zewiaceae (Heterobasidion sp.) in decay classes 3 and 4. The WIF dynamics of both deadwood species are described at greater detail in the supporting information.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2.3.5 Factors correlating to fungal community structure in the two deadwood

species

NMDS analysis clearly separated the WIF communities of the two tree species (Fig.

2.3ab; Table 2.1). Wood physico-chemical parameters correlated significantly with the fungal community structure (Fig. 2.3a). Factors that correlated significantly to the variation in the WIF community structure in both tree species were the decay class, relative wood moisture, pH, remaining mass, wood volume, wood density, C/N ratio and the concentrations of total lignin, C, N, Mg, Fe and Zn (P = 0.024-0.0001). At the individual tree species level, WIF community structure correlated significantly with decay class, relative wood moisture, remaining mass, wood density and total lignin (Table 2.1). In addition, wood volume and C and N concentration were important in Fagus deadwood, while the concentrations of total lignin and Mg were contributing significantly in shaping the fungal community structure in Picea deadwood. Fungal families that correlated significantly with WIF community structure in Fagus and Picea logs are displayed in

Fig. 2.3b.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests n- (red) deadwood deadwood (red)

anthropogenic factors Picea

nd (green) a (green)

chemical and chemical Fagus - physico o wood

egan”. ANOSIM revealed significant separation of fu of separation significant revealed ANOSIM egan”. command in “v command

envfit < 0.001, 999 permutations). 999 < 0.001,

P = 0.60, =

R Nonmetric multidimensional scaling (NMDS) ordination of fungal community structure in structure in community fungal of ordination (NMDS) scaling multidimensional Nonmetric - : 3D 3 . 2

Fig. fitted was t = (stress 0.16) ordination The R. functions in NMDS theusing and plot3d ordigl the using by (b) families fungal different of abundances (a) also and ( species to tree according structure community gal 36

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table 2.1: Goodness-of-fit statistics (R2) for parameters fitted to the nonmetric multidimensional scaling (NMDS) ordination of fungal community structure. The significance estimates were based on 999 permutations. Significant factors (Bonferroni corrected P < 0.05) are indicated in bold. Marginally significant variables (Bonferroni corrected P < 0.10) are indicated in italics. Fagus vs Picea Fagus Picea Parameter R2 P R2 P R2 P Fungal OTU richness 0.1946 0.024 0.541 0.003 0.4758 0.004 Decay class 0.4461 0.001 0.4809 0.005 0.3634 0.022 Remaining mass 0.497 0.001 0.546 0.003 0.3413 0.028 Volume 0.2393 0.008 0.4975 0.004 0.1381 0.384 Density 0.6227 0.001 0.5472 0.003 0.3098 0.047 Relative wood moisture 0.3094 0.002 0.5331 0.001 0.4357 0.01 Total lignin 0.3788 0.001 0.1847 0.256 0.7645 0.001 pH 0.533 0.001 0.0853 0.603 0.3384 0.039 Decay rate 0.0728 0.378 0.2857 0.079 0.1189 0.499 Laccase (Lac) 0.0502 0.527 0.1092 0.487 0.0169 0.987 Manganese independent peroxidase (MiP) 0.1795 0.021 0.3451 0.024 0.0852 0.679 Manganese peroxidase (MnP) 0.0729 0.328 0.0815 0.76 0.1917 0.221 C/N 0.3765 0.001 0.2555 0.119 0.0933 0.588 C 0.2836 0.001 0.1083 0.523 0.6272 0.001 N 0.1941 0.024 0.2216 0.161 0.1828 0.23 C (g/cm³) 0.6 0.001 0.5435 0.003 0.2658 0.094 N (g/cm³) 0.6134 0.001 0.6111 0.002 0.1445 0.38 Mn 0.1507 0.057 0.0234 0.967 0.0497 0.802 Mg 0.3115 0.002 0.2087 0.196 0.3407 0.027 Ca 0.0736 0.316 0.151 0.332 0.2589 0.112 K 0.0684 0.362 0.2733 0.099 0.2357 0.127 Fe 0.1876 0.016 0.1352 0.418 0.2755 0.065 Cu 0.1143 0.145 0.1753 0.248 0.2928 0.061 Zn 0.3138 0.001 0.0863 0.613 0.0801 0.64 Ni 0.0718 0.366 0.153 0.357 0.1858 0.23

2.3.6 Relationships between wood-inhabiting fungal richness, taxonomy and

lignin-modifying enzyme activities

Correlations between WIF richness (in terms of total observed, cumulative, or estimated

OTUs), WIF abundance (at both the family and cOTU levels), and the activities of lignin-modifying enzymes are presented in Table S2.8. There were no positive correlations between any actual richness of OTU/ cOTU as well as the estimated richness of total

OTU and the activities of lignin-modifying enzyme activities. Interestingly, however, there were some significant positive correlations between the abundances of certain

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests fungal families and the measured activity of lignin-modifying enzymes. In Fagus logs, the abundance of the Schizoporaceae and Xylariaceae correlated positively with Lac and MiP activity, respectively (P < 0.05). In Picea logs, the abundance of Bondar- zewiaceae correlated positively with MiP and MnP activity (P = 0.002 - 0.009). We hence found many significant correlations between fungal cOTU abundance and the activity of potential lignin-modifying enzymes (Table S2.8). In Fagus deadwood, the strongest significant positive correlations were found between Mycena alnetorum and

Lac (τ = 0.43, P = 0.003), Hypoxylon rubiginosum and MiP (τ = 0.42, P = 0.004), and

Marasmius alliaceus and MnP (τ = 0.31, P = 0.036). In Picea deadwood, the strongest significant positive correlations were found between Armillaria gallica and Lac

(τ = 0.32, P = 0.028), Heterobasidion sp. and MiP (τ = 0.37, P = 0.010), and Hetero- basidion sp. and MnP (τ = 0.45, P = 0.002). In addition, the abundances of certain families and cOTUs exhibited significant negative correlations with lignin-modifying enzyme activity (Table S2.8).

2.3.7 Relationships between wood-inhabiting fungal richness, taxonomic identity

and wood decomposition rates

There were no significant correlations between total OTU richness (in terms of total observed, cumulative or estimated OTU richness) and wood decomposition rates (Table

S2.9). However, the abundances of specific fungal taxonomic groups and cOTUs correlated significantly with wood decomposition rates. In Fagus deadwood, the abundance of the Xylariaceae family correlated significantly and negatively with decomposition rates (τ = -0.44, P = 0.003). The abundances of three individual species also correlated negatively with decomposition rates: Hypoxylon fragiforme and Xylaria hypoxylon

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

(both Xylariaceae), and Neobulgaria pura (Leotiaceae). On the other hand, Mycena purpureofusca, Phialocephala dimorphospora, Trametes versicolor correlated positively with decomposition rates (τ = 0.31 – 0.41, P = 0.034 – 0.005; Table S2.9). In Picea deadwood, the abundances of the Schizoporaceae were positively correlated with decomposition rates (τ = 0.43, P = 0.005). Individual species whose abundances correlated positively with decomposition rates were Botryobasidium botryosum, Hyphodontia alutacea, Hyphodontia alutaria and Mycena alnetorum (τ = 0.32 – 0.45, P = 0.036 – 0.003;

Table S2.9).

2.4 Discussion

This work builds on earlier studies that used high-throughput sequencing to investigate fungal community structures in deadwood (Kubartova et al. 2012; Ovaskainen et al.

2013) by providing the first comparison of two morphologically different tree species often occurring in close proximity in temperate European forests. By assessing a very comprehensive dataset on physico-chemical wood properties, we were able to identify key factors that correlate with fungal community structure. Moreover, by linking fungal richness and community composition to enzyme activities and decomposition rates, we demonstrated that ecosystem processes are controlled by complex mechanisms such as assembly histories and competition scenarios.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

2.4.1 Fungal diversity and community composition

We found no significant differences between the two deadwood species in terms of total or mean OTU richness. The mean fungal OTU richness tended to increase with increasing decay class. This is inconsistent with the findings of fructification pattern studies, which indicated that the abundance of fruiting bodies was highest at intermediate stages of decay (Heilmann-Clausen 2001; Hoppe et al. 2014). However, the fact that fungal

OTU richness increased with losses of mass and density (i.e. as the decay class increased) is in agreement with the results of previous studies on spruce deadwood that used molecular techniques (Kubartova et al. 2012; Rajala et al. 2012). The discrepancy between these findings may be due to the fact that many fungi tend to reside as vegetative mycelia in deadwood and therefore do not develop fruiting bodies (Kubartova et al.

2012).

Only a small proportion of the sequences obtained from deadwood of Picea abies were assignable to Ascomycota, though the ratio of taxonomically assigned OTUs was balanced to Basidiomycota. This result was in line to recent findings on fungal diversity in

Norway spruce (Ottosson et al. 2015) and reflects that large parts of biological diversity are only accessible via molecular based techniques (Hibbett et al. 2011).

2.4.2 WIF dynamics during decomposition processes and corresponding factors

Fungal community structure differed significantly between the two tree species as indicated by the NMDS analysis. The community structure was also more dependent on the species origin of deadwood than on the surrounding forest type. More specifically, fungal communities in Fagus logs in beech stands were more similar to fungal communi-

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests ties in Fagus logs in spruce forests than to fungal communities in Picea deadwood in beech forests. The same was true for Picea logs in spruce and beech stands. This further demonstrates that substrate type has a greater impact on WIF community structure than forest types. Nevertheless, the impact of tree species should be further tested in the future with more tree species, especially those tree species that have similar wood physico-chemical properties. Our results indicate that the physico-chemical properties of the wood (decay class, relative wood moisture, remaining mass, wood density, C/N ratio, total lignin) correspond significantly to community structure in both tree species. This is consistent with the report of Rajala et al. (2012) who investigated this aspect in 500

Norway spruce logs.

Deadwood of the two tree species also differed substantially with respect to the way in which their WIF communities changed as the wood decayed. In Picea logs, Resinicium bicolor was dominant in all decay classes; other fungi such as Amylostereum areolatum,

Heterobasidion sp. and Fomitopsis pinicola were co-dominant in specific decay classes but much less abundant. Conversely, in Fagus logs, no single fungus was dominant in all decay classes. The dominant species changed from Annulohypoxylon cohaerens and

Trametes versicolor in decay class 1 to Fomes fomentarius in decay classes 2 and 3 and then Mycena alnetorum and Trametes versicolor in decay class 4. It may be surprising that Polyporaceae were still dominant in the decay class 4 since species of this family require substrates with high energy contents to produce fruiting bodies and are therefore consequently rarely found in wood at later stages of the decomposition when using methods that focus on fruiting bodies (Bader et al. 1995; Lindblad 1998). This may be due to the presence of vegetative mycelia and/ or DNA residues in the wood (Kubartova et al. 2012), both of which would be captured by our methodology. However, the

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests change in the dominant polypore species from Fomes fomentarius in decay class 3 to

Trametes versicolor in decay class 4 indicates that this phenomenon may not be entirely related to DNA residues.

The fungal community dynamics in Picea deadwood observed in this study were distinctly different compared to those reported for boreal forest ecosystems in Fen- noscandia (Kubartova et al. 2012; Ovaskainen et al. 2013; Rajala et al. 2011; Rajala et al. 2012). First, the ascomycetes that were reported to be most dominant during the early stages of decay in boreal forests (Rajala et al. 2012) were largely absent in the Cen- tral European forests, where basidiomycetes were most abundant across all stages of decay. Secondly, different fungi were dominant at different stages of decay in the boreal environment (Kubartova et al. 2012; Ovaskainen et al. 2013; Rajala et al. 2012), whereas in our study Resinicium bicolor was most abundant in all decay classes. Third, species such as Hyphodontia alutaria, Ascocoryne cylichnium, Heterobasidion parviporum and Fomitopsis pinicola were classified as early colonizers in boreal forests (Kubartova et al. 2012) but were common in all decay classes in our study and even dominant in

Picea logs of decay class 4. Finally, ectomycorrhizal (ECM) fungi that were already detected during the early stages of decay in previous studies and became strongly dominant during the final stages of decomposition in boreal forests (Rajala et al. 2012) were largely absent in our study. We did not observe any increase in the abundance of ECM species in the more mineralized wood of decay class 4. Only 4 ECM cOTUs were detected in a single Picea log in the later stages of decay: Lactarius sp. (42 sequences),

Laccaria amethystina (22 sequences), Russula fellea (9 sequences) and Xerocomus pru- inatus (4 sequences). Together, these species accounted for only 0.18% of all sequences detected in Picea abies. The relatively low abundance of ECM in Picea logs from tem-

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests perate forests as compared to boreal forests could be due to differences in N-availability in the two forest ecosystems (Nasholm et al. 1998). We assume that ECM fungi in temperate forests preferentially acquire N from the soil, whereas in the N-limited boreal environment it is worthwhile for ECM fungi to acquire nitrogen by attacking deadwood

(especially highly decayed deadwood) to avoid the competition with forest floor vegetation (Rajala et al. 2011).

2.4.3 The role of WIF in ecosystem functions and processes

Both the activities of lignin-modifying enzymes and the decomposition rate were related to the abundances of particular fungal families and cumulative OTUs. Laboratory scale studies demonstrated that members of different fungal taxa that were detected in this study (such as Armillaria sp., Fomes fomentarius, Trametes versicolor) can efficiently produce similar amounts of lignin-modifying enzymes (Baldrian 2006) and cause similarly high levels of mass loss (Valmaseda et al. 1990). The co-occurrence of these fungi in deadwood suggests that there is some functional redundancy within the studied WIF communities. Several authors have reported increased Lac and MnP activities due to two-species-interactions (Baldrian 2004; Freitag & Morrell 1992; Snajdr et al. 2011;

White & Boddy 1992). Chi et al. (2007) found that some combinations of two fungi can accelerate the decay of wood due to increases in MnP production relative to that observed in equivalent cases featuring only a single fungal species. Whether enzymes are actually secreted due to fungus-fungus interactions or for other reasons may depend on the community structure and the state of degradation of the colonized wood. However, recent studies (Dickie et al. 2012; Dowson et al. 1988a, b; Fukami et al. 2010;

Fukasawa et al. 2009) have demonstrated a high degree of interaction among coexisting fungal species, suggesting that WIF may invest more energy into competing 43

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests with one-another than on producing wood-degrading enzymes under natural field conditions. Coates and Rayner (1985) also found that interaction reduced the rate of wood decay. Therefore, as demonstrated in this work, the high species richness of deadwood- resident fungal communities need not be associated with any increase in the production of wood degrading enzymes or wood decomposition rate. Nevertheless, we identified some fungal OTUs that are known to be active producers of lignin-modifying enzymes and strong wood decomposers that cause a white-rot with high mass loss (Table S2.10).

Trametes versicolor is among the most important decomposers, occurring in all decay classes of Fagus deadwood. We further show that different wood types (deciduous vs. coniferous) directly relate to decomposer (fungal) community structure and dynamics.

The decomposition rates on the two deadwood species were not significantly different.

This may be due to functional redundancy in their WIF communities and demonstrates that the fungi in each community are adapted to their host tree species. Fungi in coniferous wood have to deal with larger amounts of extractives as well as more recalcitrant and condensed lignin than is encountered in Fagus wood (Blanchette 1991; Higuchi

2006). In respect to anatomy, coniferous woods have simpler structures than deciduous woods (Fengel & Wegener 1983). Such factors will affect fungal substrate preferences.

Brown-rot fungi prefer coniferous wood while most white-rot fungi colonize both coniferous and deciduous wood (Hibbett & Donoghue 2001). The preference of brown-rot fungi for coniferous wood was apparent in our data: the three most abundant brown-rot fungi Fomitopsis pinicola, Dacrymyces stillatus and Antrodia sinuosa were found to be much more abundant in Picea than in Fagus. Interestingly, we found more sequences for white-rot fungi in Picea wood than we did for brown-rot fungi. The mean enzyme activities of MiP and MnP in Picea were only marginally lower than those observed for

Fagus, further demonstrating the presence of white-rot species in the studied Picea 44

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests deadwood. These observations are consistent with those of Olsson et al. (2011) and

Rajala et al. (2012), who also found white-rot fungi to be more abundant than brown-rot fungi on Picea abies.

The relationships between fungal cOTU richness, family abundances and the activities of lignin-modifying enzymes and decomposition rates were very different in Fagus and

Picea logs. The fungal communities in both deadwood species were composed of different cOTUs that were positively correlated and/ or expected to play roles with all lignin-modifying enzyme activities and wood decomposition rates. Interestingly, there were some fungi whose abundances did not correlate positively with lignin-modifying enzyme activity and/ or wood decomposition rates in this field study even though they secreted high titers of lignin-modifying enzymes and caused high mass losses under laboratory conditions. This could be due to the succession of the studied communities

(i.e. priority effects) and the interspecific interactions among different fungal species, as discussed above. In addition, ITS is of variable copy number and may not directly correlate to biomass, which could distort the relationship between a fungus’s ‘abundance’ in the dataset and the observed enzymatic activity.

Our results also revealed significantly negative correlations between the abundances of

Xylariaceae species and decomposition rates on Fagus logs. Different members of the

Xylariaceae were abundant in many Fagus logs of early to intermediate decay classes, and logs harboring these species often exhibited low rates of decomposition. This could be related to the ability of fungi of this family (which cannot produce MnP) to impede deadwood colonization by secondary saprotrophic basidiomycetes (Fukasawa et al.

2009). Xylariaceae were shown to act very defensively against saprotrophic basidiomy-

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests cetes on 2% malt agar and were not displaced by saprotrophic basidiomycetes in twigs over an incubation period of 6 months (Fukasawa et al. 2009). Xylariaceae can form pseudosclerotial plates (PSPs) to delineate decay columns (demarcation lines) within the wood and protect them from attacks by competing saprotrophs. These decay columns can persist even after several years of decomposition (Fukasawa et al. 2009;

Purahong & Hyde 2011). We observed similar recalcitrant dense matrices of melanized hyphae (PSPs) in Fagus logs that were highly dominated by Xylariaceae (Fig. S2.10).

Trametes versicolor was also highly abundant in the Fagus deadwood at various stages of decay, and its abundance correlated negatively with that of Xylariaceae members.

Interestingly, some logs in the early stages of decay that had been colonized by

Trametes versicolor rather than Xylariaceae exhibited very high rates of wood decomposition. This demonstrates the importance of priority effects and interspecific interactions among different fungal species (Hiscox et al. 2015). Trametes versicolor has been shown to secrete large quantities of different lignin-modifying enzymes and yields high wood decomposition rates under laboratory condition (Valmaseda et al. 1990). We assume that the properties and species origin of deadwood affects both the dynamics of the fungal community and the interactions among different fungal species. Many fungal

OTUs were present in both Fagus and Picea deadwood but the dominance patterns and temporal dynamics of the communities in each case differed substantially (Figs. S2.6,

S2.7).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests 2.5 Conclusion

Fungal community structure was significantly different between deadwood of Fagus sylvatica and Picea abies occurring in close proximity in temperate forests of Germany.

Wood physico-chemical properties are the main factors corresponding to the fungal communities in these deadwood species. Under the studied natural conditions, microbial-mediated ecosystem functions (i.e. the activities of lignin-modifying enzymes) and processes (wood decomposition rate) were controlled by successional assembly history, interspecific interactions and competition scenarios rather than total OTU/ -species richness (Dickie et al. 2012; Fukami et al. 2010; Hiscox et al. 2015).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests 2.6 References

Abarenkov K, Henrik Nilsson R, Larsson K-H, et al. (2010) The UNITE database for molecular identification of fungi – recent updates and future perspectives. New Phytologist, 186, 281–285. Bader P, Jansson S, Jonsson BG (1995) Wood-inhabiting fungi and substratum decline in selectively logged boreal spruce forests. Biological Conservation, 72, 355– 362. Baldrian P (2004) Increase of laccase activity during interspecific interactions of white- rot fungi. FEMS Microbiology Ecology, 50, 245–253. Baldrian P (2006) Fungal laccases - occurrence and properties. FEMS Microbiology Reviews, 30, 215–242. Blanchette RA (1991) Delignification by wood-decay fungi. Annual Review of Phytopathology, 29, 381–398. Blaser S, Prati D, Senn-Irlet B, Fischer M (2013) Effects of forest management on the diversity of deadwood-inhabiting fungi in Central European forests. Forest Ecology and Management, 304, 42–48. Boddy L (2001) Fungal community ecology and wood decomposition processes in angiosperms: from standing tree to complete decay of coarse woody debris. Ecological Bulletins, 49, 43–56. Chi YJ, Hatakka A, Maijala P (2007) Can co-culturing of two white-rot fungi increase lignin degradation and the production of lignin-degrading enzymes? International Biodeterioration & Biodegradation, 59, 32–39. Coates D, Rayner ADM (1985) Fungal population and community-development in cut beech logs. 3. Spatial Dynamics, Interactions and Strategies. New Phytologist, 101, 183–198. Coleman DC, Whitman WB (2005) Linking species richness, biodiversity and ecosystem function in soil systems. Pedobiologia 49, 479–497. Dickie IA, Fukami T, Wilkie JP, Allen RB, Buchanan PK (2012) Do assembly history effects attenuate from species to ecosystem properties? A field test with wood- inhabiting fungi. Ecology Letters, 15, 133–141. Dowson CG, Rayner ADM, Boddy L (1988a) Inoculation of mycelial cord-forming basidiomycetes into woodland soil and litter. 1. Initial establishment. New Phytologist, 109, 335–341. Dowson CG, Rayner ADM, Boddy L (1988b) Inoculation of mycelial cord-forming basidiomycetes into woodland soil and litter. 2. Resource capture and persistence. New Phytologist, 109, 343–349. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin, 19, 11–15. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R (2011) UCHIME improves sensitivity and speed of chimera detection. Bioinformatics, 27, 2194–2200. Effland MJ (1977) Modified procedure to determine acid-insoluble lignin in wood and pulp. Tappi 60, 143–144. Eisenhauer N, Scheu S, Jousset A (2012) Bacterial diversity stabilizes community productivity. PLoS ONE e34517. doi: 10.1371/journal.pone.0034517 Fengel D, Wegener G (1983) Wood: Chemistry, Ultrastructure, Reactions. Walter de Gruyter, Berlin and New York.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fischer M, Bossdorf O, Gockel S, et al. (2010) Implementing large-scale and long-term functional biodiversity research: The Biodiversity Exploratories. Basic and Applied Ecology, 11, 473–485. Floudas D, Binder M, Riley R, et al. (2012) The paleozoic origin of enzymatic lignin decomposition reconstructed from 31 fungal genomes. Science, 336, 1715–1719. Freitag M, Morrell JJ (1992) Changes in selected enzyme-activities during growth of pure and mixed cultures of the white-rot decay fungus Trametes versicolor and the potential biocontrol fungus Trichoderma harzianum. Canadian Journal of Microbiology, 38, 317–323. Fukami T, Dickie IA, Wilkie JP, et al. (2010) Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. Ecology Letters, 13, 675–684. Fukasawa Y, Osono T, Takeda H (2009) Effects of attack of saprobic fungi on twig litter decomposition by endophytic fungi. Ecological Research, 24, 1067–1073. Gadd GM (2010) Metals, minerals and microbes: geomicrobiology and bioremediation. Microbiology, 156, 609–643. Gardes M, Bruns TD (1993) ITS primers with enhanced specificity for basidiomycetes - application to the identification of mycorrhizae and rusts. Molecular Ecology, 2, 113–118. Hahn F, Ullrich R, Hofrichter M, Liers C (2013) Experimental approach to follow the spatiotemporal wood degradation in fungal microcosms. Biotechnology Journal 8, 127–132. Halme P, Kotiaho JS (2012) The importance of timing and number of surveys in fungal biodiversity research. Biodiversity and Conservation, 21, 205–219. Hammer Ø, Harper DAT, Ryan PD (2001) PAST: Paleontological statistics software package for education and data analysis. Palaeontologia Electronica, 4. Harmon ME, Franklin JF, Swanson FJ, et al. (1986) Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research, 15, 133–302. Hatakka A, Hammel K (2011) Fungal biodegradation of lignocelluloses. In: The Mycota: a Comprehensive Treatise on Fungi as Experimental Systems for Basic and Applied Research. Industrial Applications (ed Hofrichter M), pp. 319–340. Springer, Heidelberg. Heilmann-Clausen J (2001) A gradient analysis of communities of macrofungi and slime moulds on decaying beech logs. Mycological Research, 105, 575–596. Heilmann-Clausen J, Christensen M (2003) Fungal diversity on decaying beech logs - implications for sustainable forestry. Biodiversity and Conservation, 12, 953– 973. Hibbett DS, Donoghue MJ (2001) Analysis of character correlations among wood decay mechanisms, mating systems, and substrate ranges in homobasidiomycetes. Systematic Biology, 50, 215–242. Higuchi T (2006) Look back over the studies of lignin biochemistry. Journal of Wood Science, 52, 2–8. Hiscox J, Savoury M, Müller CT, Lindahl B, Rogers HJ, Boddy L (2015) Priority effects during fungal community establishment in beech wood. ISME Journal doi:10.1038/ismej.2015.38 Hofrichter M, Ullrich R, Pecyna MJ, Liers C, Lundell T (2010) New and classic families of secreted fungal heme peroxidases. Applied Microbiology and Biotechnology, 87, 871v897.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Hoppe B, Kahl T, Karasch P, et al. (2014) Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi. PLoS ONE, e88141. doi:10.1371/journal.pone.0088141 Hoppe B, Krüger D, Kahl T, Arnstadt T, Buscot F, Bauhus F, Wubet T (2015) A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies. Scientific Re- ports, doi:10.1038/srep09456 Jellison J, Connolly J, Goodell B, et al. (1997) The role of cations in the biodegradation of wood by the brown rot fungi. International Biodeterioration & Biodegradation, 39, 165–179. Junninen K, Simila M, Kouki J, Kotiranta H (2006) Assemblages of wood-inhabiting fungi along the gradients of succession and naturalness in boreal pine-dominated forests in Fennoscandia. Ecography, 29, 75–83. Kahl T, Mund M, Bauhus J, Schulze ED (2012) Dissolved organic carbon from European beech logs: Patterns of input to and retention by surface soil. Ecoscience, 19, 1–10. Kögel-Knabner I (2002) The macromolecular organic composition of plant and microbial residues as inputs to soil organic matter. Soil Biology & Biochemistry, 34, 139–162. Kopra K, Fyles J (2005) Woody debris and nutrient cycling: should we care??? SFMN Research Note Series, 8. Krankina ON, Harmon ME, Griazkin AV (1999) Nutrient stores and dynamics of woody detritus in a boreal forest: modeling potential implications at the stand level. Canadian Journal of Forest Research, 29, 20–32. Kubartova A, Ottosson E, Dahlberg A, Stenlid J (2012) Patterns of fungal communities among and within decaying logs, revealed by 454 sequencing. Molecular Ecology, 21, 4514–4532. Lentendu G, Wubet T, Chatzinotas A, Wilhelm C, Buscot F, Schlegel M (2014) Effects of long term differential fertilization on eukaryotic microbial communities in an arable soil: a multiple barcoding approach. Molecular Ecology, 23, 3341-3355. Li W, Godzik A (2006) Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics, 22, 1658–1659. 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 Microbiology Ecology, 78, 91–102. Lindblad I (1998) Wood-inhabiting fungi on fallen logs of Norway spruce: relations to forest management and substrate quality. Nordic Journal of Botany, 18, 243– 255. Luyssaert S, Hessenmoller D, von Lupke N, Kaiser S, Schulze ED (2011) Quantifying land use and disturbance intensity in forestry, based on the self-thinning relationship. Ecological Applications 21, 3272-3284. Martinez AT, Speranza M, Ruiz-Duenas FJ, et al. (2005) Biodegradation of lignocellulosics: microbial chemical, and enzymatic aspects of the fungal attack of lignin. International Microbiology, 8, 195–204. Müller J, Engel H, Blaschke M (2007) Assemblages of wood-inhabiting fungi related to silvicultural management intensity in beech forests in southern Germany. European Journal of Forest Research, 126, 513–527.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Nasholm T, Ekblad A, Nordin A, et al. (1998) Boreal forest plants take up organic nitrogen. Nature, 392, 914–916. Oksanen J, Blanchet F, Kindt R, et al. (2013) vegan: Community Ecology Package. R package version 2.0-8. Olsson J, Jonsson BG, Hjalten J, Ericson L (2011) Addition of coarse woody debris - The early fungal succession on Picea abies logs in managed forests and reserves. Biological Conservation, 144, 1100–1110. Ottosson E, Kubartova A, Edmann M, Jönsson, M, Lindhe A, Stenlid J, Dahlberg A (2015) diverse ecological roles within fungal communities in decomposing logs of Picea abies. FEMS Microbiology Ecology, 10.1093/femsec/fiv012 Ovaskainen O, Nokso-Koivista J, Hottola J, et al. (2010) Identifying wood-inhabiting fungi with 454 sequencing - what is the probability that BLAST gives the correct species? Fungal Ecology 3, 274–283. Ovaskainen O, Schigel D, Ali-Kovero H, et al. (2013) Combining high-throughput sequencing with fruit body surveys reveals contrasting life-history strategies in fungi. ISME Journal 7, 1696–1709. Pollierer MM, Dyckmans J, Scheu S, Haubert D (2012) Carbon flux through fungi and bacteria into the forest soil animal food as indicated by compound- specific 13C fatty acid analysis. Functional Ecology 26, 978–990. Proulx R, Wirth C, Voigt W, et al. (2010) Diversity promotes temporal stability across levels of ecosystem organization in experimental grasslands. PLoS ONE e13382. doi:10.1371/journal.pone.0013382. Purahong W, Hoppe B, Kahl T, et al. (2014a) Changes within a single land-use category alter microbial diversity and community structure: Molecular evidence from wood-inhabiting fungi in forest ecosystems. Journal of Environmental Management, 139, 109–119. Purahong W, Hyde KD (2011) Effects of fungal endophytes on grass and non-grass litter decomposition rates. Fungal Diversity, 47, 1–7. Purahong W, Kahl T, Schloter M, et al. (2014b) Comparing fungal richness and community composition in coarse woody debris in Central European beech forests under three types of management. Mycological Progress, doi:10.1007/s11557-013-0954-y. Raiskila S, Pulkkinen M, Laakso T, et al. (2007) FTIR spectroscopic prediction of mason and acid soluble lignin variation in Norway spruce cutting clones. Silva Fennica, 41, 351–371. 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 Ecology, 4, 437–448. Rajala T, Peltoniemi M, Pennanen T, Makipaa R (2012) Fungal community dynamics in relation to substrate quality of decaying Norway spruce ( Picea abies L. Karst.) logs in boreal forests. FEMS Microbiology Ecology, 81, 494–505. Sarkanen KV, Ludwig C.H. (1971) Lignins: Occurrence, Formation, Structure and Reactions. John Wiley & Sons, New York. Schloss PD, Westcott SL, Ryabin T, et al. (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology, 75, 7537–7541. Schwarze F, Engels J, Mattheck C (2000) Fungal Strategies of Wood Decay in Trees. Springer, Berlin. 51

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Snajdr J, Dobiasova P, Vetrovsky T, et al. (2011) Saprotrophic basidiomycete mycelia and their interspecific interactions affect the spatial distribution of extracellular enzymes in soil. FEMS Microbiology Ecology, 78, 80–90. Stokland JN, Siitonen J, Jonsson BG (2012) Biodiversity in Dead Wood. Cambridge University Press, Cambridge. Torras O, Saura S (2008) Effects of silvicultural treatments on forest biodiversity indicators in the Mediterranean. Forest Ecology and Management 255, 3322-3330. Valmaseda M, Almendros G, Martinez AT (1990) Substrate-Dependent Degradation Patterns in the Decay of Wheat Straw and Beech Wood by Ligninolytic Fungi. Applied Microbiology and Biotechnology, 33, 481-484. Valentin L, Rajala T, Peltoniemi M, et al. (2014) Loss of diversity in wood-inhabiting fungal communities affects decomposition activity in Norway spruce wood. Frontiers in Microbiology, 5, 230. van der Heijden MGA, Klironomos JN, Ursic M, et al. (1998) Mycorrhizal fungal diversity determines plant biodiversity, ecosystem variability and productivity. Nature 396, 72–75. 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, 124-133. White NA, Boddy L (1992) Extracellular Enzyme Localization during Interspecific Fungal Interactions. FEMS Microbiology Letters, 98, 75–79. White T, Bruns T, Lee S, Taylor J (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: PCR Protocols: A Guide to Meth- ods and Applications (eds Innis MA, Gelfand DH, Sninsky JJ, White TJ) pp. 315–322. Academic Press, San Diego and Heidelberg Wubet T, Christ S, Schöning I, Boch S, Gawlich M, Schnabel B, Fischer M, Buscot F (2012) Differences in soil fungal communities between European beech (Fagus sylvatica L.) dominated forests are related to soil and understory vegetation. PLoS ONE 7(10): e47500. doi:10.1371/journal.pone.0047500

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests 2.7 Supplemental information

Assignment of decay classes and dendrochronological dating

Deadwood logs were assigned to four decay classes based on information on remaining mass (%) by k-means cluster analysis as described in Hoppe et al. (2014) and Kahl et al. (2012). This approach was performed to use the decay class as a categorical variable, which is independent from an observers guessing in the field. In addition we also assigned decay classes according to a “conventional” independent approach commonly used in Germany (compare Müller-Using & Bartsch 2009), during the initial sampling campaign. Both approaches significantly correlated (ρ = 0.74, P = 0.0001).

Decay rates were calculated based on a single exponential model (Harmon et al. 1986) using information on mass loss (density and volume loss) and time since death obtained by dendrochronological dating of the deadwood. Tree rings in heavily decayed dead wood were made visible by freezing the fresh piece of dead wood and then placing it in molten paraffin for 2-3 days (melting point 56-59°C, Merck, Darmstadt, Germany)

(Krüger et al. 2014). The paraffin replaces the evaporating water and stabilizes the dead wood after cooling. Tree rings were made visible by planing the surface with a razor blade.

Detailed WIF dynamics and sharedness of cOTUs and families: Fagus vs. Picea logs

(A) Detailed WIF dynamics in Fagus and Picea logs

In Fagus deadwood (Fig. S2.8), the decay class 1 was highly dominated by Xylariaceae (total abundance 26%; mainly comprising Annulohypoxylon cohaerens 17.4%, Hypoxy- lon rubiginosum 5.45%), followed by Polyporaceae (total abundance 14.6%; mainly Trametes versicolor 12.3%), Physalacriaceae (total abundance 10.4%; mainly Armil-

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests laria cepistipes 6.39%, Armillaria gallica 3.78%) and Leotiaceae (total abundance 8.18%; mainly Neobulgaria pura 7.43%). In decay class 2, Polyporaceae (mean abundance 17.8%; mainly Fomes fomentarius 13.3% and Meruliaceae (mean abundance 17%; mainly Phlebia livida 8.37% were dominant fungal families followed by Xylaria- ceae that dropped in their abundances to 10% (dominated by Xylaria hypoxylon 7.65%) when compared to decay class 1. With advancing decay (decay class 3), Xylariaceae further dropped to 5.15% (dominated by Kretzschmaria deusta 4.54%), whereas other families that dominated the deadwood decay class 2 remained dominant. In decay class 4, Mycenaceae (mean abundance 23%; dominated by Mycena alnetorum 16.1%) became the most dominant family followed by Marasmiaceae (mean abundance 15.5%; mainly Megacollybia platyphylla 7.99%) and Polyporaceae (mean abundance 10.1%; dominated by Trametes versicolor 10.1%). Xylariaceae (mean abundance 0.74%) almost disappeared from this most advanced decay class. In Picea deadwood (Fig. S2.9), all decay classes were highly dominated by Agaricomycetes I.S. (mean abundance 26.4%, 42.7%, 29.6%, 17% in decay classes 1-4, respectively; almost fully corresponding to Resinicium bicolor). Stereaceae were frequently detected in decay class 1 (mean abundance 21.7%; dominated by Amylostereum areolatum 19.9%), however, they declined in abundance in decay class 2 and 3 (mean abundances 1.18% and 2.69%, respectively) and increased in their abundance in decay class 4 again (mean abundance 7.31%; dominated by Amylostereum chailletii 7.29%). Physalacriaceae were frequently detected in decay class 1 (mean abundance 14.2%; dominated by Armillaria cepistipes 6.85% and Armillaria gallica 6.11%) but were of low abundance in other decay classes. Bondarzewiaceae (represented by Heterobasidi- on sp.) were frequently detected in all decay classes (mean abundances 10.3 – 18.9%) except in decay class 2 (1.87%). Mycenaceae were frequently found in decay classes 1 and 2 (mean abundances 7.88% and 12.5% both dominated by Mycena rubromarginata), whereas in decay class 3 and 4 they were much less frequent (mean abundances 2.77% and 3.71%). Fomitopsidaceae were frequently detected from decay class 2, 3 and 4 (mean abundances 7.44%, 17.1% and 5.81% dominated by Fomitopsis pinicola). Schizoporaceae were only highly frequently detected in decay class 4 (mean abundance 11.9%; dominated by Hyphodontia alutacea 5.73%, and Hyphodontia alutaria 3.08%).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

(B) Sharedness of cOTUs and families between Fagus and Picea

We examined the sharedness of WIF cOTUs and genera between the tree species (Fig. S2.7). At the level of species (Fig. S2.7A), this was only possible for species that received unequivocally the same epitheth. Among the 160 where this was the case, 74 species (22 Ascomycota, 50 Basidiomycota, and 2 zygomycetes) were found on both Picea and Fagus logs. Some of them are not normally reported on this particular substrate, e.g. Xylariaceae species Annulohypoxylon cohaerens, Hypoxylon fragiforme, Hypoxylon rubiginosum, Kretzschmaria deusta and Nemania serpens are known from literature to exclusively occur on Fagus logs and few other deciduous tree species, but in our case could also be detected with marginal abundances on Picea logs. At the level of genus, to assume sharedness identification to sp. of same genus minimally suffices, and with this, 35 genera are unique to Fagus, 34 to Picea, and 58 genera are shared (Fig. S2.7B).

Table S2.2: Sampling design: Distribution of deadwood logs according to tree species and the respective 9 forest plots which were assigned to three different forest management types. Management type Extensively man- Age-class man- Age-class managed beech forests aged beech forests aged spruce forests (3 plots) (3 plots) (3 plots)

Fagus sylvatica 8 8 8

Picea abies 8 6 10 Deadwood tree species

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table S2.3: List of MID tags for deadwood logs (samples) and additional characterization of sampling design. For further details regarding remaining mass of the respective deadwood log, compare Fig. S1 in Hoppe et al. (2014). MID tag used Gauss-Krüger CWD Tree Forest Plot Coordinates Altitu- in pyro- coordinates Soil type ID species manag. ID de sequencing RW HW Latitude Longitude F. spruce age- AEW 48.478064 9.334397 Cam- 8270 ACGAGTGCGT 3524799 5371201 763 m sylvatica class 1 °N °E bisol spruce age- AEW 48.478064 9.334397 Cam- 8271 ACGCTCGACA P. abies 3524799 5371201 763 m class 1 °N °E bisol spruce age- AEW 48.478064 9.334397 Cam- 8273 AGACGCACTC P. abies 3524799 5371201 763 m class 1 °N °E bisol spruce age- AEW 48.478064 9.334397 Cam- 8276 AGCACTGTAG P. abies 3524799 5371201 763 m class 1 °N °E bisol F. spruce age- AEW 48.478064 9.334397 Cam- 8277 ATCAGACACG 3524799 5371201 763 m sylvatica class 1 °N °E bisol spruce age- AEW 48.37998 9.351452 8820 ATATCGCGAG P. abies 3526110 5360300 749 m Leptosol class 2 °N °E F. spruce age- AEW 48.37998 9.351452 8821 CGTGTCTCTA 3526110 5360300 749 m Leptosol sylvatica class 2 °N °E spruce age- AEW 48.37998 9.351452 8822 CTCGCGTGTC P. abies 3526110 5360300 749 m Leptosol class 2 °N °E spruce age- AEW 48.37998 9.351452 8823 TAGTATCAGC P. abies 3526110 5360300 749 m Leptosol class 2 °N °E spruce age- AEW 48.37998 9.351452 8824 TCTCTATGCG P. abies 3526110 5360300 749 m Leptosol class 2 °N °E spruce age- AEW 48.37998 9.351452 8825 TGATACGTCT P. abies 3526110 5360300 749 m Leptosol class 2 °N °E spruce age- AEW 48.412253 9.355592 Cam- 8279 TACTGAGCTA P. abies 3526400 5363890 715 m class 3 °N °E bisol F. spruce age- AEW 48.412253 9.355592 Cam- 8280 CATAGTAGTG 3526400 5363890 715 m sylvatica class 3 °N °E bisol spruce age- AEW 48.412253 9.355592 Cam- 8281 CGAGAGATAC P. abies 3526400 5363890 715 m class 3 °N °E bisol F. spruce age- AEW 48.412253 9.355592 Cam- 8282 ATACGACGTA 3526400 5363890 715 m sylvatica class 3 °N °E bisol F. spruce age- AEW 48.412253 9.355592 Cam- 8283 TCACGTACTA 3526400 5363890 715 m sylvatica class 3 °N °E bisol F. spruce age- AEW 48.412253 9.355592 Cam- 8807 CGTCTAGTAC 3526400 5363890 715 m sylvatica class 3 °N °E bisol F. spruce age- AEW 48.412253 9.355592 Cam- 8809 TCTACGTAGC 3526400 5363890 715 m sylvatica class 3 °N °E bisol F. beech age- AEW 48.399088 9.244828 Cam- 8842 TGTACTACTC 3518205 5362394 783 m sylvatica class 4 °N °E bisol F. beech age- AEW 48.399088 9.244828 Cam- 8845 ACGACTACAG 3518205 5362394 783 m sylvatica class 4 °N °E bisol F. beech age- AEW 48.399088 9.244828 Cam- 8846 CGTAGACTAG 3518205 5362394 783 m sylvatica class 4 °N °E bisol beech age- AEW 48.419609 9.414683 Cam- 8826 TACGAGTATG P. abies 3530770 5364730 809 m class 5 °N °E bisol F. beech age- AEW 48.419609 9.414683 Cam- 8827 TACTCTCGTG 3530770 5364730 809 m sylvatica class 5 °N °E bisol F. beech age- AEW 48.419609 9.414683 Cam- 8828 TAGAGACGAG 3530770 5364730 809 m sylvatica class 5 °N °E bisol F. beech age- AEW 48.394042 9.445939 Cam- 8810 TCGTCGCTCG 3533100 5361900 767 m sylvatica class 6 °N °E bisol beech age- AEW 48.394042 9.445939 Cam- 8811 ACATACGCGT P. abies 3533100 5361900 767 m class 6 °N °E bisol beech age- AEW 48.394042 9.445939 Cam- 8812 ACGCGAGTAT P. abies 3533100 5361900 767 m class 6 °N °E bisol beech age- AEW 48.394042 9.445939 Cam- 8813 ACTACTATGT P. abies 3533100 5361900 767 m class 6 °N °E bisol F. beech age- AEW 48.394042 9.445939 Cam- 8815 ACTGTACAGT 3533100 5361900 767 m sylvatica class 6 °N °E bisol beech age- AEW 48.394042 9.445939 Cam- 8817 AGACTATACT P. abies 3533100 5361900 767 m class 6 °N °E bisol

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

F. beech age- AEW 48.394042 9.445939 Cam- 8818 AGCGTCGTCT 3533100 5361900 767 m sylvatica class 6 °N °E bisol beech age- AEW 48.394042 9.445939 Cam- 8819 AGTACGCTAT P. abies 3533100 5361900 767 m class 6 °N °E bisol F. beech age- AEW 48.396228 9.261357 8832 ATAGAGTACT 3519430 5362080 773 m Leptosol sylvatica class 7 °N °E beech age- AEW 48.396228 9.261357 8833 CACGCTACGT P. abies 3519430 5362080 773 m Leptosol class 7 °N °E beech age- AEW 48.396228 9.261357 8834 CAGTAGACGT P. abies 3519430 5362080 773 m Leptosol class 7 °N °E F. beech age- AEW 48.396228 9.261357 8835 CGACGTGACT 3519430 5362080 773 m Leptosol sylvatica class 7 °N °E beech F. AEW 48.38258 9.382386 Cam- 8285 TACACACACT natural 3528400 5360600 779 m sylvatica 8 °N °E bisol forest beech F. AEW 48.38258 9.382386 Cam- 8286 TACACGTGAT natural 3528400 5360600 779 m sylvatica 8 °N °E bisol forest beech AEW 48.38258 9.382386 Cam- 8288 TACAGATCGT P. abies natural 3528400 5360600 779 m 8 °N °E bisol forest beech AEW 48.38258 9.382386 Cam- 8289 TACGCTGTCT P. abies natural 3528400 5360600 779 m 8 °N °E bisol forest beech AEW 48.38258 9.382386 Cam- 8290 TAGTGTAGAT P. abies natural 3528400 5360600 779 m 8 °N °E bisol forest beech AEW 48.38258 9.382386 Cam- 8294 TCGATCACGT P. abies natural 3528400 5360600 779 m 8 °N °E bisol forest beech F. AEW 48.38258 9.382386 Cam- 8295 TCGCACTAGT natural 3528400 5360600 779 m sylvatica 8 °N °E bisol forest beech F. AEW 48.369336 9.41522 8296 TCTAGCGACT natural 3530840 5359140 753 m Leptosol sylvatica 9 °N °E forest beech F. AEW 48.369336 9.41522 8297 TCTATACTAT natural 3530840 5359140 753 m Leptosol sylvatica 9 °N °E forest beech F. AEW 48.369336 9.41522 8298 TGACGTATGT natural 3530840 5359140 753 m Leptosol sylvatica 9 °N °E forest beech AEW 48.369336 9.41522 8801 TGTGAGTAGT P. abies natural 3530840 5359140 753 m Leptosol 9 °N °E forest beech AEW 48.369336 9.41522 8804 ACAGTATATA P. abies natural 3530840 5359140 753 m Leptosol 9 °N °E forest

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table S2.4: Decay rate, remaining mass and wood physico-chemical parameters of Fagus (FASY) and Picea (PIAB) deadwood logs in different decay classes (DC 1 through 4; including number of logs per decay class).

FASY_1 (8) FASY_2 (6) FASY_3 (7) FASY_4 (3) PIAB_1 (5) PIAB_2 (5) PIAB_3 (5) PIAB_4 (9) Decay rate k (y-1) 0.05 ± 0.01 0.05 ± 0 0.07 ± 0.01 0.09 ± 0.02 0.04 ± 0.01 0.04 ± 0.01 0.08 ± 0 0.08 ± 0.01 Remai-ning mass (%) 80.28 ± 2.82 58.16 ± 1.46 45.57 ± 1.4 30.75 ± 1.9 75.25 ± 2.01 58.26 ± 2.4 47.64 ± 1.5 31.19 ± 0.97 Density (g*cm-3) 0.51 ± 0.02 0.37 ± 0.01 0.31 ± 0.01 0.2 ± 0.03 0.35 ± 0.01 0.27 ± 0.01 0.22 ± 0.01 0.15 ± 0.01 Rel. wood moisture (% dry 114.21 ± 138.22 ± 155.15 ± 48.72 ± 65.45 ± 163.06 ± mass) 49.83 ± 5.5 14.66 22.94 9.14 11.59 13.38 95.27 ± 8.91 24.62

C% 48.41 ± 0.16 47.92 ± 0.18 47.52 ± 0.37 47.51 ± 0.33 49.29 ± 0.31 49.92 ± 0.22 49.16 ± 0.35 51.39 ± 0.94

N% 0.13 ± 0.01 0.15 ± 0.01 0.18 ± 0.02 0.25 ± 0.02 0.08 ± 0.01 0.08 ± 0 0.08 ± 0 0.14 ± 0.02 324.02 ± 282.87 ± 193.7 ± 629.69 ± 602.56 ± 591.22 ± 422.87 ± C/N 365.22 ± 14.87 22.45 31.53 15.62 48.44 19.24 38.75 52.44 0.175 ± 0.145 ± 0.093 ± 0.172 ± 0.134 ± 0.106 ± 0.079 ± C (g*cm-3) 0.248 ± 0.01 0.007 0.005 0.015 0.005 0.002 0.008 0.003 0.0006 ± 0.0005 ± N (g*cm-3) 0.0007 ± 0 0.0001 0.0005 ± 0 0.0001 0.0003 ± 0 0.0002 ± 0 0.0002 ± 0 0.0002 ± 0 Total Lignin (%) 26.09 ± 0.48 28.52 ± 0.73 31.21 ± 1.14 30.49 ± 2.18 29.5 ± 0.59 31.63 ± 1.06 33.63 ± 2.31 45.23 ± 3.78

pH 5.25 ± 0.1 4.84 ± 0.08 5.12 ± 0.14 4.99 ± 0.13 4.63 ± 0.1 4.33 ± 0.03 4.04 ± 0.16 4.26 ± 0.15

Table S2.5: Mean ligninolytic enzyme activities and macro/ micro nutrients of Fagus and Picea deadwood logs.

Parameters Fagus sylvatica Picea abies

Lac (mU/g) 51.5 ± 11.2 26.6 ± 10.3 MiP (mU/g) 15 ± 4.9 14.4 ± 5.8 MnP (mU/g) 18 ± 9 13.8 ± 6.1 Mn (µg/g) 12.2 ± 3 24.2 ± 3.2 Mg (mg/g) 0.2 ± 0 0.1 ± 0 Ca (mg/g) 0.4 ± 0 0.4 ± 0 K (mg/g) 0.6 ± 0.1 0.4 ± 0.1 Fe (µg/g) 3.8 ± 1.2 5.1 ± 1 Al (µg/g) 0.8 ± 0.2 1.5 ± 0.3 Cu (µg/g) 0.4 ± 0 0.4 ± 0 Zn (µg/g) 30.2 ± 3.3 45.2 ± 2.8 Ni (µg/g) 0.5 ± 0.3 0.1 ± 0

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table S2.6: Total observed OTU richness and estimated OTU richness (Chao1 and ACE) of Fagus and Picea deadwood logs in different decay classes (DC). Deadwood species Parameters DC 1 DC 2 DC 3 DC 4 Total OTU richness 44.8 ± 6.7 62 ± 10.2 68.9 ± 9 62.7 ± 14.8 58.3 ± 5.1 Fagus syl- Chao1 173.3 ± 24.3 163.1 ± 20.6 161.4 ± 22.1 182.6 ± 38.8 168.4 ± 12.6 vatica ACE 162 ± 22.2 171.1 ± 15.6 160.9 ± 20.7 176.2 ± 31.9 165.7 ± 11.1 cOTU richness 36.25 ± 5.38 50.83 ± 8.96 60.57 ± 8 55.67 ± 9.92 49.42 ± 4.57 OTU richness 41.8 ± 5.4 69.6 ± 6.1 68 ± 5.5 74.8 ± 8.6 65.4 ± 4.6 Chao1 125.4 ± 15.3 162.5 ± 19.4 162.1 ± 12.7 170.7 ± 19.6 157.7 ± 10 Picea abies ACE 139.1 ± 19.5 166.8 ± 16.2 173.4 ± 16.8 186 ± 22.5 169.6 ± 11.1 cOTU richness 31.6 ± 8.24 55.2 ± 6.04 53.6 ± 5.17 66.89 ± 8.76 54.33 ± 4.71

Table S2.7: Summary of fungal sequences, OTU statistics and fungal phyla distribution from MOTHUR analysis. Parameter Fagus Picea Total sequences 39,640 42,163 Average sequences per sample 1,651.63 1,756.79 Total OTUs in all samples 506 504 Average OTUs per sample 58.29 65.41 Number of OTUs per phylum: Ascomycota 265 217 Basidiomycota 201 242 zygomycetes 6 7 Entomophthoromycota 0 1 Glomeromycota 0 1 Chytridiomycota 1 1 Unknown fungal OTUs 33 35

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table S2.8: Relationships between fungal richness, abundances of fungal families and cOTUs and activities of ligninolytic enzymes revealed by non-parametric Kendall-tau correlation analysis. Significant correlation (P < 0.05) are displayed in bold. Marginal significant correlations (P < 0.1) are displayed in italics. Parameter Enzyme activities in Fagus Enzyme activities in Picea (mU/g) (mU/g) Laccase MiP MnP Laccase MiP MnP Fungal OTU richness (observed) -0.09 -0.23 -0.04 -0.34 -0.14 -0.24 Fungal cOTU richness -0.12 -0.25 -0.10 -0.16 -0.16 -0.26 Fungal OTU richness (Chao1) 0.01 -0.13 0.07 -0.26 -0.04 -0.15 Fungal OTU richness (ACE) 0.02 -0.22 0.09 -0.24 -0.07 -0.18 Family abundances Agaricomycetes I.S. 0.27 0.00 0.11 0.22 0.01 0.06 Bondarzewiaceae NA NA NA 0.23 0.37 0.45 Chaetosphaeriaceae -0.16 -0.30 -0.25 -0.12 -0.07 -0.08 Coniochaetaceae -0.18 -0.06 -0.20 -0.34 -0.23 -0.27 Dacrymycetaceae 0.02 -0.13 -0.32 -0.43 -0.38 -0.44 Diatrypaceae -0.18 -0.01 0.07 NA NA NA Fomitopsidaceae -0.21 -0.18 0.07 -0.24 -0.21 -0.24 Helotiaceae -0.06 -0.01 -0.11 -0.09 -0.02 -0.14 Helotialaceae NA NA NA -0.27 -0.14 -0.25 Herpotrichiellaceae -0.11 0.07 -0.15 -0.29 -0.17 -0.36 Hyaloscyphaceae -0.12 -0.37 -0.19 -0.01 -0.02 -0.02 Leotiaceae 0.02 0.23 -0.08 NA NA NA Marasmiaceae 0.05 0.04 0.22 -0.05 -0.02 -0.12 Meruliaceae 0.23 0.06 0.23 -0.23 -0.10 -0.17 mitosporic Helotiales -0.04 -0.42 -0.09 0.09 0.04 0.12 Mycenaceae 0.19 0.10 -0.20 -0.30 -0.14 -0.33 Peniophoraceae 0.10 0.22 0.04 NA NA NA Physalacriaceae 0.11 0.01 0.07 0.22 0.06 0.18 Polyporaceae 0.09 -0.08 0.23 -0.07 -0.12 -0.14 Schizoporaceae 0.30 0.01 0.10 -0.22 0 -0.06 Stereaceae NA NA NA 0.08 0 0.03 Strophariaceae -0.21 0.24 0.25 -0.15 -0.11 0.14 Trechisporaceae NA NA NA 0.02 0.22 0.19 Xylariaceae -0.02 0.32 0.17 -0.08 0.01 -0.11 cOTU abundances Amylostereum areolatum NA NA NA -0.11 -0.20 -0.16 Amylostereum chailletii NA NA NA 0.16 0.13 0.09 Annulohypoxylon cohaerens -0.07 0.37 -0.19 NA NA NA Armillaria cepistipes 0.02 -0.04 -0.02 0.14 0.00 0.14 Armillaria gallica -0.20 -0.22 -0.03 0.32 0.17 0.30 Ascocoryne cylichnium 0.14 -0.30 -0.12 -0.11 -0.02 -0.09 Ascocoryne sarcoides -0.12 0.08 -0.18 NA NA NA Bjerkandera adusta 0.10 0.30 0.19 NA NA NA Botryobasidium botryosum NA NA NA -0.03 0.10 0.01 Calocera cornea -0.01 -0.22 -0.22 NA NA NA Ceriporiopsis gilvescens 0.13 -0.01 0.18 NA NA NA Dacrymyces stillatus NA NA NA -0.30 -0.30 -0.35 Eutypa spinosa -0.16 -0.03 0.18 NA NA NA Fomes fomentarius 0.12 0.12 0.26 NA NA NA Fomitopsis pinicola -0.21 -0.18 0.07 -0.19 -0.22 -0.20 Heterobasidion sp. NA NA NA 0.23 0.37 0.45 Hyphodontia alutacea NA NA NA -0.25 -0.11 -0.19 Hyphodontia alutaria NA NA NA 0.05 0.23 0.16 Hypoxylon fragiforme -0.09 0.32 -0.06 NA NA NA

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Parameter Enzyme activities in Fagus Enzyme activities in Picea (mU/g) (mU/g) Laccase MiP MnP Laccase MiP MnP Hypoxylon rubiginosum 0.33 0.42 0.19 NA NA NA Marasmius alliaceus 0.09 0.22 0.31 NA NA NA Megacollybia platyphylla -0.05 -0.23 0.09 NA NA NA Mycena alnetorum 0.43 0.12 0.13 -0.05 0.08 -0.06 Mycena purpureofusca -0.08 -0.24 -0.42 NA NA NA Mycena rubromarginata NA NA NA -0.41 -0.25 -0.41 Neobulgaria pura 0.08 0.35 0.02 NA NA NA Oudemansiella mucida 0.14 0.21 0.20 NA NA NA Peniophora incarnata 0.10 0.22 0.04 NA NA NA Peniophorella praetermissa 0.26 -0.03 0.01 0.03 -0.05 -0.10 Phialocephala dimorphospora -0.12 -0.37 -0.18 NA NA NA Phlebia livida 0.05 0.06 0.20 -0.12 0.05 -0.19 Resinicium bicolor 0.21 0.09 0.02 0.29 0.10 0.12 Scopoloides hydnoides 0.02 -0.29 -0.18 NA NA NA Sistotrema brinkmanii 0.01 0.13 0.09 -0.09 -0.13 -0.11 Trametes versicolor 0.00 -0.15 0.09 NA NA NA Xylaria hypoxylon -0.13 0.16 0.00 NA NA NA

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fagus Picea Parameter τ P τ P Xylariaceae -0.44 0.003 -0.07 0.605 cOTU abundances Amylostereum areolatum NA NA -0.18 0.252 Amylostereum chailletii NA NA -0.03 0.846 Annulohypoxylon cohaerens -0.12 0.428 NA NA Armillaria cepistipes -0.09 0.524 0.02 0.879 Armillaria gallica 0.00 1.000 0.29 0.059 Ascocoryne cylichnium 0.19 0.193 -0.03 0.868 Ascocoryne sarcoides 0.06 0.678 NA NA Bjerkandera adusta -0.07 0.629 NA NA Botryobasidium botryosum NA NA 0.43 0.005 Calocera cornea 0.25 0.091 NA NA Ceriporiopsis gilvescens 0.01 0.919 NA NA Dacrymyces stillatus NA NA -0.14 0.351 Eutypa spinosa 0.05 0.729 NA NA Fomes fomentarius -0.01 0.929 NA NA Fomitopsis pinicola 0.10 0.480 -0.15 0.321 Heterobasidion sp. NA NA 0.08 0.617 Hyphodontia alutacea NA NA 0.32 0.036 Hyphodontia alutaria NA NA 0.32 0.036 Hypoxylon fragiforme -0.25 0.083 NA NA Hypoxylon rubiginosum -0.23 0.122 NA NA Marasmius alliaceus -0.07 0.626 NA NA Megacollybia platyphylla 0.18 0.210 NA NA Mycena alnetorum 0.04 0.790 0.45 0.003 Mycena purpureofusca 0.31 0.034 NA NA Mycena rubromarginata NA NA 0.01 0.973 Neobulgaria pura -0.29 0.049 NA NA Oudemansiella mucida 0.07 0.629 NA NA Peniophora incarnata -0.22 0.131 NA NA Peniophorella praetermissa 0.14 0.353 -0.21 0.177 Phialocephala dimorphospora 0.41 0.005 NA NA Phlebia livida -0.13 0.374 NA NA Resinicium bicolor -0.15 0.319 -0.19 0.219 Scopoloides hydnoides 0.12 0.431 NA NA Sistotrema brinkmanii -0.11 0.449 -0.15 0.315 Trametes versicolor 0.32 0.031 NA NA Xylaria hypoxylon -0.31 0.033 NA NA

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Table S2.10: Fungal cOTUs and their potential roles in deadwood decomposition. FASY= Fagus sylvatica, PIAB = Picea abies. Potential roles in Fungal cOTUs Sample deadwood decomposi- Published ecological role Reference (Abundance in%) tion 8280 Fomes fomentarius laccase producer important producer of ligno- Větrovský et (FASY) (75.22) cellulose-degrading en- al. (2013) zymes endo-1,4-β- glucanase, 1,4-β- glucosidase, cellobiohydro- lase, endo-1,4-β-xylanase, 1,4-β-xylosidase, MNP, and laccase 8282 Annulohypoxylon MiP producer Annulohypoxylon spp. are Robl et al. (FASY) cohaerens (46.18) excellent producers of hy- (2013) drolytic enzymes 8283 Mycena alnetorum laccase producer laccase and peroxidase pro- Eduardo et (FASY) (48.22) ducer al. (1997) 8298 Trametes versicolor important decomposer important decomposer Valmaseda (FASY) (30.43) (causing high mass (causing high mass and et al. (1990) loss) lignin loss) 8810 Armillaria gallica MnP producer Armillaria spp. caused high Valmaseda (FASY) (18.92) mass and lignin loss et al. (1990) 8818 Kuehneromyces mu- laccase producer laccase producer Cho et al. (FASY) tabilis (36.62) (1999) 8820 Resinicium bicolor MiP and MnP producer white-rot fungus able to Shah et al. (PIAB) (66.50) produce ligninolytic en- (2014) zymes, especially laccase 8823 Amylostereum chail- important decomposer saprotrophic decay fungus Vasiliauskas (PIAB) letii (65.61) (causing high mass occurring on fallen conifer et al. (1998) loss) logs 8825 Resinicium bicolor laccase producer white-rot fungus able to Shah et al. (PIAB) (62.97) produce ligninolytic en- (2014) zymes, especially laccase 8828 Trametes versicolor important decomposer important decomposer Valmaseda (FASY) (87.51) (causing high mass (causing high mass and et al. (1990) loss) lignin loss) 8832 Phlebia livida (43.49) laccase producer Phlebia spp. known as lac- Arora & (FASY) case producers Rampal (2002) Arora DS, Rampal P (2002) Laccase production by some Phlebia species. Journal of Basic Microbiology, 42, 295–301. Cho NS, Park JM, Choi JM, et al. (1999) The effects of wood rotting fungi and laccase on destaining of dyes and KP bleaching effluent. Journal of Korean Wood Science and Technology. 27, 72–79. Eduardo V, Carlos R,Sigisfredo, G. (1997) Macro-microscopic and qualitative enzymatic characterization of mycelial strains obtained from basidiocarps of Mycena species (Agaricales) in Chile. Revista Chilena de Historia Natural, 70, 521–530. Hoppe B, Kahl T, Karasch P, et al. (2014) Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi. PLoS ONE, e88141. doi:10.1371/journal.pone.0088141 Krüger I, Muhr J, Hartl-Meier C, Schulz C, Borken W (2014) Age determination of coarse woody debris with radiocarbon analysis and dendrochronological cross-dating. European Journal of Forest Research 133, 931–939. Kahl T, Mund M, Bauhus J, Schulze ED (2012) Dissolved organic carbon from European beech logs: Patterns of input to and retention by surface soil. Ecoscience 19, 1–10. doi: 10.2980/19-4-3501. Müller-Using S & Bartsch N (2009) Decay dynamic of coarse and fine woody debris of a beech (Fagus sylvatica L.) forest in Central Germany. European Journal of Forest Research 128, 287-296. Robl D, Delabona PD, Mergel CM, et al. (2013) The capability of endophytic fungi for production of hemicellulases and related enzymes. BMC Biotechnology, 13, 94. Shah H, Yusof F, Alam Z. (2014) Production of laccase by Resinicium bicolor in submerged cultures: application of the Plackett-Burman experimental design to screen major factors. Proceedings of the International Conference on Advances in Civil, Structural, Environmental & Biotechnology, 142–146. Valmaseda M, Almendros G, Martinez AT (1990) Substrate-dependent degradation patterns in the decay of wheat straw and beech wood by ligninolytic fungi. Applied Microbiology and Biotechnology, 33, 481–484.

Vasiliauskas R, Stenlid J, Thomsen IM (1998) Clonality and genetic variation in Amylostereum areolatum and A. chailletii from northern Europe. New Phytologist, 139, 751–758. Větrovský T, Baldrian P, Gabriel J (2013) Extracellular enzymes of the white-rot fungus Fomes fomentarius and purification of 1,4-β-glucosidase. Applied Microbiology and Biotechnology, 169, 100–109.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.5: Relationships between fungal OTUs richness and decay classes (as determined by density (a) and remaining mass (b)), green = Fagus sylvatica, red = Picea abies.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.6: Heatmap displaying cOTU abundances and presences per sample. This figure is a companion to Fig. 2.2 of the main text and contains only the cOTUs that appear labeled in that figure. The rows are sorted by the number of sequences on the family level, which also only includes the cOTUs appearing in this table.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.7: Comparison of sharedness and species (= cOTU) distribution. Green colors indicate Fagus, red colors Picea. (a) Table of 74 species that are shared between Fagus and Picea made as MS Excel in-cell dot chart. The left and right columns are filled by a heatmap corresponding to overall cOTU size in number of sequences while the bars are relative to cOTU size within the tree species. Up to 150 sequences this dot chart is size-correct. (b) Area-correct Venn diagram for sharedness of genera (Pan-Omics Research, Pacific Northwest National Laboratory).

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.8: Heatmap (green-white) of mean abundances of highly detected fungal families and cOTUs in different decay classes (DC, labeled FASY_1 through FASY_4) in Fagus sylvatica deadwood logs.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.9: Heatmap (red-white) of mean abundances of highly detected fungal families and cOTUs in different decay classes (DC, labeled PIAB_1 through PIAB_4) in Picea abies deadwood logs.

2 Linking molecular wood-inhabiting fungal diversity and community dynamics to ecosystem functions and processes in Central European forests

Fig. S2.10: Some examples: Fagus sylvatica logs showing pseudosclerotial plates (PSPs) due to the colonization by Xylariaceae.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3 Network analysis reveals ecological links between N-

fixing bacteria and wood-decaying fungi

3.1 Introduction

Deadwood is an important structural component of forest ecosystems. It influences numerous ecosystem functions (Cornwell et al. 2009; Harmon et al. 1986), including carbon (C) sequestration (Chambers et al. 2000; Herrmann & Bauhus 2012; Kahl et al.

2012; Litton et al. 2007), nutrient cycling (Brunner & Kimmins 2003), and provision of habitats for wood-dwelling organisms (Lonsdale et al. 2008; Rondeux & Sanchez

2010). Although many investigations have focused on the diversity of fungi, particularly

Agaricomycotina (basidiomycetes) and some Pezizomycotina (ascomycetes), in terms of assemblages (Heilmann-Clausen 2001; Heilmann-Clausen & Christensen 2003;

Müller et al. 2007) or their role in wood decomposition (Boddy & Watkinson 1995), the participation of bacteria in the processes involved is largely unexplored. It is known that substrate qualities, such as nutrient and water contents, strongly influence wood colonization by microbes (Volkenant 2007). The amount of nitrogen (N) available in wood is highly restricted (Cornwell et al. 2009), with carbon to nitrogen ratios generally ranging from ca. 350-800:1 (Spano et al. 1982), but wood-decaying fungi can completely min- eralize and metabolize most wood residues such as cell wall lignocellulose complexes

(Blanchette 1991) and are capable of mobilizing enough N to produce not only their vegetative hyphae but also sporocarps and millions of spores. Cowling and Merrill

(1966) hypothesized that associations with N-fixing bacteria may enable wood- 70

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi inhabiting fungi to meet their substantial N requirements. Nitrogen fixation, the energet- ically expensive reduction of atmospheric dinitrogen to two molecules of ammonia, is enabled by adenosine triphosphate (ATP) generated from the decomposition of cellulose (Weißhaupt 2012). This connects bacterial colonization of deadwood with fungal processes of wood decay. The presence and activity of N-fixing bacteria in both living and deadwood have been assessed by acetylene reduction assays in various studies (Aho et al. 1974; Brunner & Kimmins 2003; Jurgensen et al. 1984; Larsen et al. 1978;

Seidler et al. 1972; Spano et al. 1982). However, with advances of molecular techniques in the 1980s, Zehr and McReynolds (1989) were the first to establish oligonucleotide primers to amplify the nifH gene complex that encodes dinitrogenase reductase. NifH is still the standard target in studies on N-fixing prokaryotes in various natural environments (Gaby & Buckley 2011). Many investigations have been conducted on the molecular ecology of diazotrophic (N-fixing) communities in nitrogen-limited substrates, such as forest soils, salt marshes and oligotrophic marine sediments (Zehr et al. 2003).

Wang et al. (2013) recently reported on the distribution of nifH genes in four terrestrial climatic zones across the USA, where they surprisingly discovered an 80% overlap on the 95% amino acid identity threshold among their and already known genes. We are, however lacking information on nifH gene distribution in deadwood to date. In addition, several authors have investigated the diversity and community structure of bacteria in deadwood as well as functional traits related to white-rot fungi (Valaskova et al. 2009;

Zhang et al. 2008), but without focusing on diazotrophic bacteria. The work presented here emanated from the initial idea to survey bacterial community structure on deadwood to gain information whether potential N-fixers are present or not. This prompted us to immediately investigate the presence and distribution of nifH genes in deadwood.

The objectives of the present study were to: a) explore the diversity of nifH sequences in 71

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi deadwood, b) test the hypothesis that the community composition of N-fixing bacteria correlates with the diversity of fungi fructifying on deadwood, and c) assess the likelihood that these sequences encode functional enzymes by phylogenetic methods.

3.2 Material and methods

3.2.1 Experimental design

The study plots are located in the UNESCO Biosphere Reserve “Schwäbische Alb” in southern Germany, one of three experimental sites included in the German Biodiversity

Exploratories (Fischer et al. 2010), designed to provide a large-scale, long-term open platform for functional biodiversity research along a north-south gradient in Germany.

The main objectives of this endeavor are to elucidate the influence of land use and management type on biodiversity and ecosystem functioning. The mean annual temperature at the “Schwäbische Alb” exploratory is 6-7°C and annual precipitation ranges between ca. 700 and 1000 mm. Our survey was conducted on deadwood logs (hereafter “logs”) in very intensively investigated 1 ha plots (VIP), representing the following three forest management types: (i) extensively managed beech forests, where timber harvesting stopped several decades ago, (ii) managed beech forests dominated by Fagus sylvatica and (iii) managed spruce forests dominated by Picea abies.

3.2.2 Deadwood

In April 2009, logs located on the forest floor in the VIPs were randomly selected and their properties (length, diameter, tree species) were characterized. For the present study on N-fixing bacterial genes, subsets of deadwood logs were randomly selected representing each of the two focal tree species (P. abies and F. sylvatica) in plots represent-

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi ing the three management types. The random subset selection assured that Fagus logs were present in Picea-dominated plots and vice versa, giving 45 logs in total (a sampling scheme is displayed in Fig. S3.7). In June 2009 wood chips from the logs were sampled using a cordless Makita BDF451 drill (Makita, Anja, Japan) equipped with a 2 x 42 cm wood auger. The number of cores drilled depended on the volume and length of the logs. A minimum of three cores were drilled up to a log length of 5 m. Each additional 5 m of log length resulted in another drill core. A maximum of 7 cores was sampled in a 25 m long log. To avoid contamination between samples, the wood auger was flamed and wiped with ethanol between each core. The drill was operated slowly and introduced at an angle of ~45º to a line perpendicular to the log axis. To avoid overheat- ing the sample, the operation was paused periodically. Depending on the log’s diameter at the point of drilling, the auger was either drilled through it or inserted to its maximum length. The wood samples were kept on dry ice and later stored at -80°C upon return to the lab. The total volume of wood chips from each drill core was ground under liquid nitrogen into a fine powder using a Retsch MM400 swing mill (Retsch, Haan, Germa- ny). Wood C and N concentrations were determined through total combustion. For this purpose 10 mg of each wood sample was weighed into a tin capsule and analyzed using a Truspec elemental analyzer (Leco, St. Joseph, MI, USA). The remaining mass after decay (%) of each log was calculated using information on its dimensions (length, diameter and volume loss), density and the density of fresh undecayed wood, according to

Kahl et al. (2012). Deadwood logs were assigned to 4 decay classes based on remaining mass (%) data by k-means cluster analysis (Hartigan & Wong 1979) in R v. 2.11.1. This classification is more reliable than the conventional method of educated guessing by an observer.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.2.3 Fungal sporocarp inventories

Sporocarps are part of the currently actively growing fungi in the wood substrate and were chosen as study objects rather than conducting a more comprehensive nucleic acid based assessment of fungal diversity that may include inactive fungi. Sampling of sporocarps took place on the preselected subset of deadwood logs at three different occasions in June (contemporary with wood sampling for molecular analyses of nifH genes),

September and October/November 2009 to cover the full aspect of fructification of particular species according to their phenology across the course of the year. All sporocarps larger then 1cm were sampled, excluding fully resupinate corticoid fungi (Basidiomyco- ta) and non-stromatic pyrenomycetes and discomycetes of the phylum Ascomycota and morphologically identified to the species level if possible. Dried specimens were depo- sited at the herbarium LZ (University of Leipzig).

3.2.4 DNA isolation

Total community DNA from 1 g of each previously homogenized wood sample, which was divided into four 1.5 ml microcentrifuge tubes, was isolated using a modified

CTAB-protocol (Doyle & Doyle 1987). Briefly, 900 µl of 2x CTAB buffer (2% [wt/vol] hexadecyltrimethylammonium bromide; 100 mM TrisHCl, pH 8.0; 1.4 M NaCl; 20 mM

EDTA; 1.5% polyvinyl-pyrrolidone (PVP), 0.2% [vol/vol] beta-mercaptoethanol), was added to the sample. Tubes were incubated at 55°C for one hour. Nucleic acids were separated from proteins and cell debris by adding 500 µl of 24:1 chloroform:isoamyl alcohol and subsequent centrifugation at maximum speed for 10 min followed by another 500 µl chloroform addition and centrifugation at maximum speed for 5 min. DNA was precipitated with 0.08 volumes of 7.5M ammonium acetate and 0.54 volumes of isopropanol and washed twice with 99% ethanol. Dried DNA pellets were dissolved in

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

100 µl molecular grade water (all chemicals supplied by Merck, Darmstadt, Germany and Carl Roth, Karlsruhe, Germany).

3.2.5 PCR, cloning and initial sequence analysis

All DNA extracts from the wood samples of each log were pooled into a composite extract prior to PCR. The primer pair PolF (5'- TGC GAY CCS AAR GCB GAC TC -3') and PolR (5’- ATS GCC ATC ATY TCR CCG GA -3') (Poly et al. 2001) was used to amplify a 360 bp fragment of the nifH gene. Each composite DNA extract was amplified separately by PCR in triplicate 20 µl reaction mixtures containing 4 µl FIREPol 5x

Master Mix (Solis BioDyne, Tartu, Estonia), 10 µM of each primer and approximately

20 ng template DNA. PCR was performed with an initial denaturation at 94°C for 5 min followed by 34 cycles of 94°C for 1 min, 55°C for 1 min and 72°C for 1 min 30 s and a final elongation step of 72°C for 5 min. After checking the quality of the PCR products by separation on a 1.5% agarose gel the replicates were pooled and purified using an

E.Z.N.A. Cycle-Pure Kit (Omega Bio-Tek, Norcross, GA, USA). Cloning was done with the pGEM-T Vector System (Promega, Mannheim, Germany) and Escherichia coli

JM109 according to the manufacturer’s instructions. Approximately 32 clones per library were screened by PCR re-amplification of the insert using M13F and M13R primers and the following PCR-conditions: 95°C for 5 min, 32 cycles of 95°C for 40 s,

54°C for 30 s and 72°C for 60 s then a final elongation step of 72°C for 10 min. Insert re-amplicons were purified with ExoSAP-IT (USB Corporation, Cleveland, OH, USA) then used in cycle sequencing with M13F as a sequencing primer and a Big Dye Termi- nator Cycle Sequencing Reaction Kit v.3.1 (Applied Biosystems, Foster City, CA,

USA). After an ethanol precipitation sequencing was completed using an ABI 3730xl

DNA Analyzer (Applied Biosystems). PolF/R primer residues were trimmed using Se-

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi quencher 4.10 (Genecodes, Ann Arbor, MI, USA). Amplified sequences were identified by BLASTn queries against NCBI GenBank using standard settings but targeting only the bacteria division. New nucleotide sequences and their OTU (operational taxonomic unit) assignments are available under accession numbers HF559482-HF560561.

3.2.6 Compilation of a nifH Database

We have obtained, through importing into BOSQUE v. 1.81. (Ramirez-Flandes & Ulloa

2008), the GenBank flatfiles of all nucleic acid accession numbers contained in the

ARB database provided online by the Zehr Marine Microbiology lab, University of Cal- ifornia at Santa Cruz (updated February 17, 2012) and sent by the Buckley lab (corresponding to Gaby and Buckley (2011)). Sequences that were shorter than 100 bases or of apparently non-bacterial origin were immediately excluded. Sequences that were either longer than 10,000 bases or originated from whole genome sequencing were also excluded due to difficulties in aligning multiple sequences of widely varying lengths. A local BLASTn search against our own sequences and preliminary alignment attempts on the MAFFT v. 6 server, accessed through logging in at www.bioportal.uio.no (Katoh &

Toh 2008; Kumar et al. 2009; Noy et al. 2009) were used to identify sequences for removal as well as those requiring reverse complement conversion. Another BLASTn search of our own sequences against the entire DDBJ database identified another set of nifH sequences to add to the growing nifH compilation. Then, duplicates arising from the different sources and all PolF and PolR primer annealing sites were removed while maintaining all sequence data in BioEdit v. 7.0.9.0 (Hall 1999). After using UCHIME

(Edgar et al. 2011) in USEARCH v. 6.0 (Edgar 2010) in de-novo and cross-wise modes

(our sequences vs. external sequences) we flagged potential chimeras in both our wood- derived sequences and sequences from the public databases. Information from the Gen-

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Bank files pertaining to origin and ecology was appended to the FASTA file headers of all hitherto retained sequences, using the Ecobyte Replace Text v. 2.2 program. The dataset of 25,303 previously published sequences together with the novel deadwood nifH sequences were submitted to CD-HIT (Li & Godzik 2006) via www.bioportal.uio.no using a 97% cutoff threshold for building OTUs. OTU membership was appended to the FASTA headers of all sequences.

3.2.7 Protein alignment and phylogenetic analyses

We found it impossible to align even a subset of the data at the nucleotide level with confidence, thus following several other authors (Chien & Zinder 1996; Raymond et al.

2004; Young 2005) we performed phylogenetic analyses and structure comparisons only on predicted proteins. We opted to minimize the large dataset for tractability, even for computationally intensive phylogenetic analyses, by: a) using only amino acid sequences, thus also allowing comparisons at the protein level and assortment into groups described in the literature; b) using a consensus protein sequence per OTU rather than arbitrarily choosing reference protein sequences for non-singletons; c) representing only the OTUs with contributions from wood-derived nifH and the largest overall OTU. All new sequences and those of OTU1 were imported again into Sequencher. The sequences known to be part of a particular OTU were then re-assembled into a contig while singleton sequences remained separate, using amended FASTA headers to allow for automated contig building. In each contig or separate sequence we then displayed all three potential open reading frames and for ones without premature stop codon used

BLASTp to find the one matching nifH in the public databases. Next we exported from the correct reading frame the consensus amino acid sequence for non-singleton OTUs

(=contigs) or individual amino acid sequence for singletons. For OTUs of mixed origin

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

(containing our sequences and sequences from the data compiled from other sources), one consensus protein sequence or individual sequence (if there was only one) from our data was exported, and a consensus sequence or individual sequence (if one) from the remainder. For the largest group, OTU1, which did not contain any of our sequences, this was done analogously, but separately exporting an individual sequence deviating from all others by a single amino acid. All protein sequences exported in this manner were then aligned using KALIGN2 (Lassmann et al. 2009)

(www.ebi.ac.uk/Tools/msa/kalign/) and slightly modified manually at such indel positions where visual inspection suggested that alignment columns have been erroneously offset. Ultimately the alignment contained 111 columns. We used the 7-residue-sliding window hydrophobicity scheme in KALIGNVU (Lassmann & Sonnhammer 2006), applying the Kyte-Doolittle method (Kyte & Doolittle 1982) to explore variability in the proteins and the WebLogo Server (Crooks et al. 2004), which calculates amino acid frequency and entropy across each alignment column. Comparability to findings of other studies was maintained by using the information in the ARB file from the Zehr lab on the group memberships according to (Raymond et al. 2004) for the previously published sequences represented in the phylogenetic alignment and the near-BLASTn hits. Data were converted to NEXUS format using ALTER (Glez-Pena et al. 2010). ProtTest 2 at webserver darwin.uvigo.es (Abascal et al. 2005) (with PHYML and a BIONJ starting tree (Gascuel 1997; Guindon & Gascuel 2003)) was used to select the best model of protein sequence evolution by the Akaike Information Criterion (AIC). Hence, we used

PhyloBayes v. 3.3 (Lartillot et al. 2009) for Bayesian inference, with command “- nchain 2 100 0.3 50 -wag -s -dgam 6”, i.e. the WAG+Γ model chosen after ProtTest analysis (best model according to AIC: LG+I+Γ, not available in chosen programs).

After examining the likelihood trace file, MESQUITE v. 2.74 (Maddison & Maddison 78

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

2011) was used to generate one 50% majority rule consensus tree from 13,500 trees,

6,750 each after 250 burn-in trees from the two Markov chains, both leveling around ln

L= -4,000. A parameter file for GARLI (Zwickl 2006) was generated in CIPRES v. 3.2

(Miller et al. 2010) by the GARLI.conf Creator, then we performed GARLI v. 2.0 100 replicate ML bootstrapping with WAG+Γ+I (parameters estimated) in CIPRES on the

XSEDE beta server. Again, a 50% majority rule consensus tree was generated in MES-

QUITE from the 100 best trees of the bootstrap replicates. 100-replicate MP bootstrapping was performed in MEGA5 (Tamura et al. 2011) with all gap-containing alignment columns considered, using close-neighbor-interchange (CNI) heuristics and saving a condensed consensus tree with MP bootstrap threshold of 50. The unrooted phylogenetic trees were displayed, converted, compared and edited by a number of tree editors as suitable for the output type of tree file, most notably DensiTree v. 2.01 (Bouckaert

2010) to look for conflicts between phylogenetic methods and TreeGraph2 v. 2.0.47-

206 (Stöver & Müller 2010) to highlight tips based on the information contained in extended labels. Final combination of the Bayesian phylogenetic consensus tree with support values, clade labeling and alignment variability was done using Corel PhotoPaint

X3 v. 13 (Corel, Ottawa, ON, Canada). Conflicting phylogenetic signals in the protein alignment were also assessed using SplitsTree v. 4.12.3 (Huson & Bryant 2006) with the model WAG+Γ (Γ shape 0.521) +I (proportion of invariable sites 0.094) in the Pro- teinMLdist method, producing a neighbor-network reticulogram.

3.2.8 Statistical analysis

The effects of tree species, decay stage and forest management type on the nifH OTU community structure in logs were analyzed by perMANOVA using the vegan package in R. Multivariate regression trees (MRT) (De'Ath 2002) were subsequently applied to

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi describe and display the relationships of the OTUs with the independent variables by repeatedly splitting the data based on Euclidean distances. Data were then visualized in

Principal Component Analysis (PCA) biplots of the group means from the prior MRT.

Intersect correlations were calculated for each axis. Axes with strong correlations

(> 0.8) potentially account for significant between-group variation. Both PCA and MRT were performed using the mvpart package v. 1.4.0 in R. The effects of C and N content per density unit, diversity of nifH OTUs and decay classes were assessed using multiple regressions with stepwise backward selection. C and N content data were log- transformed to meet the assumptions of normality. The significance and importance of the independent variables for the model were tested using ANOVA. Polynomial and linear regression were applied to display the important factors on the fructification of fungi. ANOVAs were conducted to assess the significance of the effects.

We used EcoSim (Gotelli & Entsminger 2009) to test for non-random co-occurrence patterns based on presence-absence species distribution, by calculating C-scores (Stone

& Roberts 1990) and checkerboard indices, which respectively evaluate the tendency of species not to co-occur and indicate the number of species pairs that never co-occur; using the default settings. Furthermore, we tested for non-random associations between pairs of nifH OTUs and sporocarps using the PAIRS program (Ulrich 2008). Only fungal species and nifH OTUs present in at least three samples were evaluated. A total of

100 random matrices were obtained to generate C-scores using the fixed row and fixed column constraints algorithm. Significant species under-dispersion or over-dispersion

(at the 5% probability level) is indicated by Z-transformed scores (observed C-score - expected C-score) above 1.96 or below -1.96 (Ulrich & Zalewski 2006). Cytoscape

(Shannon et al. 2003) was used to visualize the correlations generated through PAIRS, keeping only network edges involving at least one nifH OTU. 80

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.3 Results

3.3.1 PCR clone library analysis and clustering

In total, 1,080 sequences were obtained by sequencing the PCR clone library derived from the deadwood samples. They were subsequently incorporated into a nifH compilation dataset of 25,303 sequences from GenBank, thus enlarging it to 26,383 sequences.

Clustering of the sequences using CD-HIT resulted in 7,730 OTUs of which 5,130 appeared as singletons. Among the total OTU set, the 1,080 sequences obtained from deadwood clustered into 176 OTUs of which 70 were singletons. Only eight of these

176 OTUs included both sequences detected in the logs and sequences present in the

25,303 compiled GenBank entries. The rank abundance of the 200 largest OTUs encompassing the 12 most abundant OTUs from the deadwood samples coupled with the nearest BLAST hits from GenBank is presented in Fig. 3.1 and its inserted table. OTU5, the fifth largest OTU (Fig. 3.1), consisted of 260 sequences, 206 of which were obtained from P. abies, 52 from F. sylvatica logs and two from GenBank entries identified as Methyloferula stellata strains isolated from peat. OTU42, the second largest OTU including sequences from this study (Fig. 3.1), consisted of one sequence type revealed from deadwood together with 69 sequences from GenBank identified as Azospirillum brasilense, isolated from various sources (wheat rhizosphere, paddy soils, sea grasses).

The most abundant OTU1 comprised 679 GenBank sequences exclusively obtained from assays of marine samples and none were found in the analyzed logs. Roughly 11% of the 176 OTUs could be assigned to Rhizobiales at ≥ 97% similiarity percentage through BLASTn against GenBank (Table S3.3). At a threshold of ≥ 90% 65 OTUs were identified as deriving from Rhizobiales while another 15 OTUs were associatied to dinitrogenase reductase genes from orders Rhodocyclales, Pseudomonadales, Rhodospi- 81

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi rillales, Sphingomonadales and Burkholderiales. The majority of the OTUs could only be identified as nifH associated to uncultured bacteria (Fig. 3.1).

As we assigned the 25,303 sequence types from GenBank to their source environments

(according to 17 classifications, including isolates), we could analyze the distribution of all 7,730 nifH OTUs in relation to habitat types (Fig. 3.2). The database compiled from

GenBank revealed that prior to our study, no nifH sequences associated with deadwood had been identified, apart those associated with guts of wood-inhabiting insects (168

OTUS of the 349 “Terrestrial Animal Symbionts”). Most sequences clustered into

OTUs assigned to the “Soil and Belowground” or “Marine” environments, with 3,181 and 2,809 counts, respectively. “Hotsprings” and “Wastewater” were the environments with the fewest OTU counts.

Fig. 3.1: Rank abundance chart displaying the distribution of the 12 most abundant nifH OTUs derived from the deadwood dataset within the compiled nifH dataset comprising 26,383 sequences. Only the 200 largest OTUs are shown due to space limitations. Colored bars indicate deadwood tree species (green, Fagus sylvatica; red, Picea abies). The inserted table lists the best BLASTn hit reference sequences in NCBI Genbank for the same 12 most abundant wood-derived OTUs from our study.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. 3.2: Distribution of the 7,730 nifH OTUs according to the environments where they have been detected, and whether described as originating from an isolate in GenBank. 168 of the 176 OTUs derived from this deadwood study have been exclusively identified in wood samples. The integrated heatmap displays proportions of rare sequence types (singletons).

3.3.2 Protein phylogenetic analysis of the functionality of nifH genes from

deadwood

We performed a protein phylogenetic study on the 176 OTUs corresponding to all sequences we detected in deadwood plus the most abundant OTU (OTU1), which did not include nifH sequences from our deadwood. In this part of the study a consensus protein sequence was exported from Sequencher for each non-singleton OTU. Singleton sequences and a deviant within OTU1 were translated into protein sequences. From the eight deadwood OTUs that also contained sequences from other studies found in distinct habitats (OTUs 5, 42, 52, 119, 332, 401, 515 and 1544), one translated protein sequence not corresponding to the wood habitat was exported as an individual sequence

(if one) or consensus sequence (if more than one) and added to the analyses. In total,

186 protein sequence types (also annotated with source tree species if derived from deadwood) were used for phylogenetic calculations. Phylogenetic analyses revealed that many sequences assemble in a long comb, with short branches, henceforth called “Su- pergrade” (Fig. 3.3; three parts). All protein motifs within that clade appeared to be similar and exclusively correspond to the group called Group I by Raymond et al. (2004).

The part of the tree designated “Superclade” was better resolved and was dissected into seven nodes (1-7) (Fig. 3.3). Nodes 3/4 and 5 to 7 (Fig. 3.3) were exclusively populated

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi by sequences of Groups II and III (Raymond et al. 2004), respectively, mainly derived from F. sylvatica logs. A notable exception was node 2, the base of a long branch, containing sequences of Group IV, previously characterized as less certain to be really involved in nitrogen fixation. The transitional part of the tree between the poorly and relatively well resolved branches, designated as “Intermediates”, hosts members of nifH

Groups I and III. The protein phylogeny did not show a full separation of nifH sequences by tree species, indeed the highly conserved protein motifs in the Supergrade appeared to contain almost equal numbers of sequences from each tree species. Only the

OTUs that assembled in nodes 3-7 of the Superclade were predominantly from Fagus sylvatica. Protein sequences at labeled nodes showed distinct differences in amino acid composition as displayed in the alignments visualized by using KALIGNVU (Lassmann et al. 2009) and WebLogo (Crooks et al. 2004) (Fig. 3.3). All protein sequence types contained the iron sulfur coordinating cysteines that are marked with black squares in

Fig. 3.3. Splitstree analysis (Fig. S3.8, see Supplemental Information) also divided the phylogenetic tree into relatively well and poorly resolved parts. The broader tree structure of the reticulogram (Fig. S3.8) is the same as in Fig. 3.3, but better shows phylogenetic distances because it is given as unrooted network. 53 potential chimeras that appeared in 9 OTUs were only detected within the Supergrade, which has weak phylogenetic resolution.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi r- e- o- Doolittle hyd - k values at inte , followed by number of of by number , followed sts, Extensiv = extensively Picea een values = GARLI Maximum values een rom in trees,in unrooted. Blac - and with the term PotChim. The width of

, PIAB = f v.0.972 ) shows the position within the complete complete the within position the shows ) v.0.972 100 MP bootstrap support. Green color indicates indicates color Green support. 100 MP bootstrap italics

Fagus s with MOTUs solely of one tree species origin, collapsed but keeping the collapsed but keeping tree origin, one MOTUs of species solely s with labels with sequences from other sources: ecological data other of summary nearsources: labels BLASTsequences from with hit,

79 MP bootstrap support, dark pink = 80 - = pink dark support, 79 MP bootstrap - origin andorigin dark blue color origin mixed (with bars showing ratio of [green] vs. [red]).Terminal labels s with

protein sequences. 50% majority rule consensus tree of 13,500 PhyloBayes post burn post PhyloBayes 13,500 of tree consensus rule 50% majority sequences. protein

nifH origin, red color Picea color red origin,

Fagus m

l branches represents the number of sequences (size correct up to 10 sequences). To the right, amino acid sequence logos and Kyte logos acid amino sequence To right, correct to 10 sequences up (size the of sequences). the represents number branches l

: Phylogenetic tree of of tree Phylogenetic : 3 . 3 Fig. Gr 50). > (if support bootstrap (MP) Parsimony Maximum MEGA5 = values Pink 0.5). > (if Probability Posterior Bayesian = nodes clade monophyletic represent triangles Terminal 50). > (if support bootstrap Likelihood internal distance (substitutions per site, see scale bar), light in = pink 50 MOTUs solely fro this quences MOTUfrom study: ID (SMOTU = singleton MOTU), of sequences, number total FASY = from in sequences order,the same type(s) = then (AC.Conif forest management managed spruce forests, = beech AC.Decid fore managed and of beech order.sequences number in Terminal same forests) managed sequences MOTU. in that of contain MOTUs that as nucleotide potential chimeras sequences flagged appear in visible termina (based screenshot from on labeled alignments shape for tree the tree. TheArchaeopteryx on small nodes phobicity tree. phylogenetic

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.3.3 nifH sequence diversity

Analysis of the richness and community distribution of the 176 nifH OTUs revealed 3 to

14 different OTUs per deadwood log, and significantly higher richness in F. sylvatica than in P. abies logs (Fig. S3.9A; P = 0.028). PerMANOVA analysis revealed that both tree species and decay class significantly explained the variation of the nifH community on deadwood (P = 0.001), but not forest management type (Table 3.1). Multivariate regression analysis clearly separated the nifH communities according to tree species

(Fig. 3.4A). Within deadwood of the different species, the decay class was the dominant factor for the separation. Notably the communities in Fagus sylvatica logs and Picea abies logs of the least decomposed stage 1 were clearly distinct from the communities of more decayed deadwood logs.

Fig. 3.4: nifH community structure A: Multivariate regression tree of nifH OTU community composition estimated from sequences of the clone library obtained from deadwood of Picea abies and Fagus sylvatica. Analyses were conducted for different decay classes, based on the remaining mass per deadwood log after decay using k-means cluster analysis. B: Principal Component Analysis biplots of the group means of the multivariate regression tree. The larger circles (per color) represent the multivariate group means, the individual logs are denoted by smaller circles, with matching colors and designation to Fig. 3.4A. The identity of selected OTUs with characteristic discriminatory loading is specified. Each OTU label is located at its weighted mean from the group means. Intersect correlation is given in brackets.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Table 3.1: Results of perMANOVA analysis of Bray-Curtis dissimilarities in nifH OTU community structure in relation to tree species, decay class (based on remaining mass after decay) and management type and their interactions, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseu- do-F = F value by permutation, boldface indicates statistical significance at P < 0.05, P-values based on 999 permutations (lowest P-value possible is 0.001). Df SS MS F R² P Tree species 1 1.376 1.376 3.929 0.079 0.001 Decay class 1 1.073 1.073 3.064 0.061 0.001 Management type 2 0.757 0.378 1.08 0.043 0.322 Tree species x Decay class 1 0.589 0.589 1.681 0.034 0.024 Tree species x Management type 2 0.575 0.287 0.821 0.033 0.847 Decay class x Management type 2 0.712 0.356 1.016 0.041 0.447 Tree species x Decay class x Management type 2 0.84 0.42 1.198 0.048 0.157 Residuals 33 11.56 0.35 0.661 Total 44 17.48 1

On F. sylvatica logs a broader branching of decay stage 2 from stages 3 and 4 explained the separation of nifH community structure. The first principle component (PC) of the

PCA (Fig. 3.4B) explained roughly 41% of the total variation, and mainly separated

OTUs associated with F. sylvatica logs of intermediate decay stage 2 (driven by

OTU52, the fourth most abundant in the dataset) and decay stages 3 and 4 from sequences associated with P. abies logs of decomposition stage 2. The second PC explained nearly 30% of the community variation and separated the nifH sequence types detected on F. sylvatica logs of decay classes 1 and P. abies logs as well as OTU5, the most abundant OTU in this dataset (with 258 sequences, 206 of them from P. abies deadwood).

3.3.4 Fungal diversity and influencing factors

In total 158 fungal species were detected on the deadwood logs and 131 (83%) of them could be identified to the species level (Table S3.4). Fungal species richness ranged 89

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi from 2 to 20 observed species on Fagus logs and from 2 to 14 on logs of Picea abies

(Fig. S3.10). Mean species richness was significantly higher (ANOVA, P = 0.027) on

Fagus sylvatica (9.14 ± 1.01) versus Picea abies (6.04 ± 0.71) (Fig. S4). We also observed a wider Basidiomycota to Ascomycota ratio on Picea abies. Barely 2% of the detected species belonged to the phylum Ascomycota, while they accounted for 40.3% of all taxa being observed on Fagus sylvatica (Fig. S3.11). ANOVA of factors affecting sporocarp richness based on multiple regression analyses also revealed that tree species significantly (P = 0.005) affected the fungal diversity (Table 3.2), as well as N content

(P = 0.003), decay class (based on the remaining mass after decay; P = 0.011), and nifH

OTU richness (P < 0.001). In contrast, C content did not influence the diversity of sporocarps on deadwood (P = 0.896). Polynomial and linear regressions were performed separately for the F. sylvatica and P. abies logs to avoid redundancy in the analyses, since tree species had the strongest effects on chemical constitution and nifH OTU richness, as displayed in Figs. S3.9 and S3.14. Fungal diversity was related to the decay class. The highest number of sporocarps was observed on logs of intermediate stages of mass loss (Fig. 3.5A F. sylvatica R² = 0.1884, P < 0.05, P. abies R² = 0.1206, P = 0.14), which was significant on logs of Fagus. We also observed a negative correlation between N content and sporocarp richness (Fig. 3.5B; F. sylvatica R² = 0.1856, P = 0.04) that was significant for F. sylvatica logs. There was no correlation between remaining mass after decay and nifH OTU richness (R² = 0.0701, P = 0.28, Fig. S3.12A, F. sylvatica only), but polynomial regression revealed that the number of OTUs peaked at

54.9% water content (R² = 0.268, P < 0.02, Fig. S3.12B), which in turn significantly correlates with the intermediate decay stages (R² = 0.7396, P < 0.001, Fig. S3.12C).

Furthermore, we also observed a significantly positive correlation between diversity of

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi nifH OTUs and sporocarp species richness on logs of both tree species (Fig. 3.5C: F. sylvatica R² = 0.2803, P = 0.011; P. abies R² = 0.3809, P = 0.0017).

Table 3.2: ANOVA table of effects of the indicated factors on fungal fructification ability. Complete model summary representing R2, F, P statistics. Abbreviations of the depicted ANOVA table Df = degrees of freedom, SS = sum of squares, MS = mean sum of squares. The summary model is as follows: R², F, and p were 0.5208, 8.476 and <0.001 (significant), respectively. Boldface indicates statistical significance. Sporocarp richness Df SS MS F P Tree species 1 1.784 1.784 8.812 0.005 log (Ng/cm3) 1 2.015 2.015 9.952 0.003 log (Cg/cm3) 1 0.004 0.004 0.017 0.896 Decay class (remaining mass after decay) 1 1.457 1.457 7.198 0.011 nifH OTU richness 1 3.32 3.32 16.399 <0.001 Residuals 39 7.896 0.203

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. 3.5: Interrelations between sporocarp richness and remaining mass after decay in%, nifH OTU richness and log-transformed nitrogen content per density unit (N (g/cm³)) (A, B, C). The figure displays interrelations separately per deadwood species.

3.3.5 Co-occurrence patterns

C-score and checkerboard analyses revealed structured, non-random associations between bacterial nifH and fungal sporocarps. Observed C-score and checkerboard index values (20.1712 and 940.0, respectively; Fig. S3.13) were both significantly higher than the expected values from randomized datasets (C-score=19.80343, P < 0.0001 and checkerboard = 841.0292, P < 0.0001). The pairwise relationships in this assembled community were thus further detailed using PAIRS. A null model analysis using fixed rows and columns resulted in 54 correlations among nifH OTUs and sporocarps with significant Z-transformed scores above 1.96 and below -1.96. By incorporating information on preferential occurrences of fungi and nifH OTUs according to deadwood species (using a threshold of 75% to classify them as either Fagus or Picea affiliated; oth- erwise substrate “generalists”) we identified both positive associations and avoidance patterns in the context of the two different wood-species substrates (Fig. 3.6). The

Fagus and Picea subnetworks entail four and two positive co-associations within nifH

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi sequences, respectively. The ambiguous subnetwork of generalists contains more nifH than fungi, but only five positive co-associations between different nifH. The Fagus subnetwork is the largest, dominated by fungi with well-described functions as wood- decayers and mainly abundant nifH OTUs. There are no positive associations bridging the Fagus and Picea subnetworks. Most associations that link either Fagus or Picea with the ambiguous subnetwork are positive.

Network Analyzer (Assenov et al. 2008) was used to identify hub nodes from their degree distribution. The 50 nodes were connected to 3.16 neighbors on average. nifH

OTUs 92, 74, 217 and 52 (which are highly connected to other nodes: > 10 degrees) could be identified as hubs, presumed to be stable towards random node removal from the network. Ten nifH OTUs in the ambiguous subnetwork (meaning they occurred more evenly on both Fagus and Picea) also show positive associations with fungi. For example, nifH OTU72 co-occurs with Xylariales Hypoxylon fragiforme and Xylaria hyopoxylon. NifH OTU5 (detected on 11 Fagus and 19 Picea logs) only avoids

Xylariales Annulohypoxylon cohaerens and Xylaria hyopoxylon as well as the Helotia- les Bisporella citrina (all known nearly exclusively from Fagus logs). Six nifH OTUs

(72, 274, 451, 592, 873 and 1075) significantly co-occur with fungal species that were solely present on either deadwood species.

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. 3.6: Network organized around 23 nifH OTUs and 27 fungal species (abbreviations according to the legend). Fungi (sporocarps) and nifH OTUs serve as connected nodes, solid lines display co-occurrence patterns (Z-Score < -1.96) and dotted lines avoidance patterns. Edge widths display significance levels from thinnest = 0.049 to thickest = 0.0017. Differently shaped and colored nodes/ hubs display taxonomic differences on phylum level and their ecological role in wood decay. Subnetworks are grouped by tree species, and colored background circles indicate affiliations of included taxa to substrate deadwood species (green = Fagus affiliated, red = Picea affiliated, blue = unaffiliated “Generalists”).

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.4 Discussion

3.4.1 Characterization and diversity of nifH sequences in deadwood

Gaby and Buckley (2011) stated that the diversity of nifH is not evenly distributed across various environments and that further investigations are required to understand the links and specificity of diazotrophic communities to their substrates. Our approach led to the discovery of a rich diversity of novel nifH sequences in hitherto unexplored deadwood substrate that corresponds to presumably active nifH. Assigning the 1,080 new sequences to OTUs with a large backing compilation of previously published nifH sequences from public databases allowed us to examine specificities of the nifH pools in source environments. The finding that only eight among the 176 (containing sequences from this study) included sequences derived from other environments indicates that deadwood is a specific substrate for N-fixing bacteria. The proportion of singletons

(39%) within this deadwood dataset of 176 OTUs was below the average value (66%) for all OTUs derived from 17 source classifications (16 environments plus isolates from cultures), as displayed in Fig. 3.2. This can be explained by the proximity of the investigated logs, all of which originated from nine plots at a single experimental site in Ger- many, while the OTUs retrieved from GenBank were from geographically widely distant sources. OTU composition mainly differed between the tree species. While 63 and

87 OTUs were exclusively detected in P. abies and F. sylvatica deadwood, respectively, only 26 OTUs were common to both tree species. BLASTp and additional comparison with the Zehr database indicated that almost all detected nifH sequences were members of nifH Groups I, II and III designated by Raymond et al. (2004), with the notable exception of node 2 at the base of a long branch. These groups are thought to be functional nifH genes. Accordingly, the amino acid sequence around the cysteines required for 95

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

formation of Fe4S4 clusters at alignment positions 54 and 93 (Fig. 3.3) also correspond to these groups, as depicted in Fig. 6 of the cited article (Raymond et al. 2004). Collec- tively, the evidence suggests that almost all nifH sequences detected in this study encode real dinitrogen reductase. OTUs that assembled in the Supergrade of the phylogenetic tree exclusively correspond to nifH Group I, which primarily originate from Cya- nobacteria and Proteobacteria (Raymond et al. 2004). The latter produce dinitrogenase reductases with Fe4S4 clusters that are inactivated and inhibited by O2 (Berman-Frank et al. 2001). We found that roughly 78% of our 1,080 sequences belong to Group I, which also was the most numerous group according to Wang et al. (2013). As wood decay advances, water content is likely to rise and oxygen levels to fall, at least seasonally

(Volkenant 2007), and higher N-fixation rates have been detected under anaerobic than aerobic conditions in fallen tree boles of Pseudotsuga menziesii (Silvester et al. 1982).

Oxygen depletion in more water-logged or decayed logs could allow the activity of oxygen-sensitive nifH genes from Group I. Accordingly, the transitional part of the tree between the poorly and relatively well resolved branches, designated “Intermediates”, hosts members of both nifH Groups I and III. Only the six OTUs in node 2 contained sequences that point to membership of nifH Group IV. The 36 deadwood nifH sequences (from only eight logs) included in these six OTUs have GenBank matches with at most 69-86% identity according to BLASTn searches. The further resolved nodes 3, 4 and 5 to 7 were exclusively comprised of sequences affiliated with Groups II and III, respectively. OTUs that clustered in these branches were mainly derived from F. sylvatica, possibly due to differences in the chemical constitution of the two species’ wood. Wazny and Wazny (1964) measured micronutrient concentrations in 34 tree species and found vanadium concentrations were highest in Pinus nigra and P. abies, which seems inconsistent with our finding of Group III sequences (alternative nitrogen- 96

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi ases, in which for instance vanadium replaces molybdenium) mainly in dead F. sylvatica wood, but should be subject to further investigation.

3.4.2 Correlations of nifH community structure with environmental settings

Phylogenetic reconstruction revealed that tree species strongly influence nifH diversity, a conclusion supported by the multivariate statistics and network analyses. Only 26

OTUs were detected in both F. sylvatica and P. abies logs, and OTU richness was significantly higher in F. sylvatica logs (Fig. S3.9A). Differences in chemical and structural parameters of the species’ wood could explain this pattern. For example, wood density and C content (g*cm-³) were both significantly higher in Fagus sylvatica logs

(P < 0.001, Fig. S3.14). The four stages of decay, distinguished on the basis of remaining mass, also significantly separated the nifH community. Notably, there were significant differences between the structures of the nifH communities associated with F. sylvatica logs of initial decay stage 1 and later stages (Fig. 3.4A). The results for P. abies logs revealed similar segregation of the pools of nifH sequences associated with initial and advanced decay stages. Hicks et al. (2003) and Spano et al. (1982) reported contrasting fixation rates in coniferous wood, finding them to be highest in moderate phases of decay and later phases of decay, respectively. How well nifH diversity relates to actual N-fixation rates in deadwood is hence still an open question. The intensity of forest management in the different forest plots did not influence the distribution of nifH sequence types: the highest variations in OTU richness were found in logs of extensively managed forests, but mean values did not differ significantly among the different management types. Brunner and Kimmins (2003) proposed that the amount of available deadwood significantly affects N-fixation, finding ranges in rates from 1 to 2.1 kg*ha-

1*year-1, and hence long-term N accumulation in unpolluted ecosystems. As we have no

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi basis to estimate actual N-fixation rates in our sampled logs and correlate them with as yet unavailable data on quantities of deadwood at the study sites, we can hardly address this issue. However, Shaffer et al. (2000) observed a decline in the nifH gene pool in litter of Douglas fir forests when they were clearcut, and hence proposed that the lost diversity could potentially contribute more to N-fixation in forest litter than in litter from plants that regrow in clearcuts.

3.4.3 Interrelation of fungal fructification and N-fixing bacteria through N-

availability

The impression that deadwood is a specific substrate whose decomposer communities harbor N-fixers is reinforced by variations in wood decay progression, fruiting body production and variations in N concentration during wood decomposition. Deadwood is a complex, heterogeneous and dynamic environment. Thus, several factors probably contribute to the presence of unique pools of presumably functional nifH sequences in our sampled deadwood community. Our finding of fewer singletons than in some other environments points to greater sequencing depth and spatial closeness of our samples, and the weak overlap with sequences from other environments to deadwood as a unique environment. Our hypothesis that nifH community structure is interrelated with fungal occurrences on deadwood was supported by positive correlations between sporocarp- and nifH OTU richness (Fig. 3.5C). Heilmann-Clausen (2001) reported that sporocarp richness was maximal in the intermediate decay stages, when N content in logs was lowest, while N-concentration continuously increased as decomposition proceeds (Fig.

S3.15), as also observed by Volkenant (2007) and Boddy and Watkinson (1995). This may reflect the high need for N in the phase of highest fungal vegetative and generative growth. A similar temporal pattern of N characteristics has been observed for Japanese

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fagus crenata wood (Fukasawa et al. 2009). We propose that further N-accumulation may be due to actively N-fixing bacteria and the decline of sporocarp occurrence and sporocarp biomass during loss of wood mass. This is consistent with pioneering studies by Merrill and Cowling (1966) and Larsen et al. (1978), who first suggested that fungi overcome N deficiencies by interaction with N-fixing bacteria and subsequently confirmed that fixation occurs in living sporocarps, respectively. Other opportunities for N accumulation that should be considered include its release and recycling from eaten and degenerating fruiting bodies as well as spores falling back onto and into the deadwood

(Merrill & Cowling 1966; Weißhaupt 2012). In addition, fungi that grow hyphae not only in deadwood but also in soils will likely access N pools outside deadwood and move it to places where it is needed most. Another potential link between the co- presence of both many nifH sequences and many fungi in deadwood is that specific bacteria may associate with specific fungi, perhaps in symbiotic or at least specialized commensal relationships. Intimate cooperation between fungi and bacteria in the process of decaying lignocellulosic material may thus be another widespread and ecologi- cally important aspect of fungal-bacterial interaction (Frey-Klett et al. 2011). Network analysis based on non-random co-occurrence patterns revealed relationships between sporocaps and nifH OTUs. Both C-score and checkerboard analyses indicated that bacteria and fungi co-occur less often than expected by chance. Their assemblages on deadwood were structured and clearly depend on the different tree species.

The network analysis revealed that co-associations between nifH OTUs and fungi are strongly dependent on substrate qualities. Interpretation of these relationships is not straightforward, which is due to the lack of validation of real N-fixation in our samples.

However, we can assume that these positive correlations among taxa are due to cross- feeding, co-colonization, niche overlaps (Faust & Raes 2012) or a combination of these 99

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi possibilities. These patterns could also follow community assembly rules, as originally proposed by (Diamond 1975) and still debated by ecologists (Horner-Devine et al.

2007). Distributions of fungi and bacteria across various environments are determined by their dispersion and adaption mechanisms (Green & Bohannan 2006; Whitaker

2009). The results of our study provide information that potentially N-fixing bacteria, detected by the presence of diverse nifH genes, are distributed across complex environments analogously to fungi. As previously mentioned, numerous studies have confirmed that biotic N-fixation occurs in deadwood and forest soils, as well as providing valuable contributions to our understanding of N-related traits and resulting ecosystem services.

The dispersion and ecology of the microbial communities involved were described in

2000, but the cited study merely focused on community shifts under different forest management regimes (Shaffer et al. 2000). Our investigation lacks information on real activity, but our results allowed us to detect and discern differences in the communities of N-fixing bacteria in terms of nifH genes in two kinds of substrates. As we included environmental information in the network analysis we were also able to identify conditions that the co-occurring species assemblages preferred or avoided. Explanations for observed patterns of nifH gene diversity include niche differentiations according to water content and oxygen depletion in different decomposition stages, or variations in the chemical constitution of particular tree species based on different wood decay types, resulting in variations in cellulose levels (Rayner & Boddy 1988) and consequently

ATP availability (Weißhaupt 2012). Whether the observed co-occurrence patterns result directly from bacterial-fungal interactions or from more complex indirect interactions remains to be elucidated.

To assess whether N newly fixed by bacteria reaches fungal fruiting bodies, labeling

15 experiments with N are needed. The labeled substances should include N2 and other N 100

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi containing compounds in surrounding soil as alternative N sources, as some fungi may bridge soil, litter and decaying wood, e.g., (Thompson et al. 2012). In addition, data on transcription and activity of the transcripts must be correlated with fungal biomass, diversity and enzymatic activities of lignocellulose degrading enzymes.

101

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.5 References

Abascal F, Zardoya R, Posada D (2005) ProtTest: selection of best-fit models of protein evolution. Bioinformatics 21, 2104-2105. Aho PE, Seidler RJ, Evans HJ, Raju PN (1974) Distribution, enumeration, and identification of Nitrogen-fixing bacteria associated with decay in living White fir trees. Phytopathology 64, 1413-1420. Assenov Y, Ramirez F, Schelhorn SE, Lengauer T, Albrecht M (2008) Computing topological parameters of biological networks. Bioinformatics 24, 282-284. Berman-Frank I, Lundgren P, Chen YB, et al. (2001) Segregation of nitrogen fixation and oxygenic photosynthesis in the marine Cyanobacterium Trichodesmium. Science 294, 1534-1537. Blanchette RA (1991) Delignification by wood-decay fungi. Annual Review of Phytopathology 29, 381-398. Boddy L, Watkinson SC (1995) Wood decomposition, higher fungi, and their role in nutrient redistribution. Canadian Journal of Botany 73, 1377-1383. Bouckaert RR (2010) DensiTree: making sense of sets of phylogenetic trees. Bioinformatics 26, 1372-1373. Brunner A, Kimmins JP (2003) Nitrogen fixation in coarse woody debris of Thuja plicata and Tsuga heterophylla forests on northern Vancouver Island. Canadian Journal of Forest Research 33, 1670-1682. Chambers JQ, Higuchi N, Schimel JP, Ferreira LV, Melack JM (2000) Decomposition and carbon cycling of dead trees in tropical forests of the central Amazon. Oecologia 122, 380-388. Chien YT, Zinder SH (1996) Cloning, functional organization, transcript studies, and phylogenetic analysis of the complete nitrogenase structural genes (nifHDK2) and associated genes in the archaeon Methanosarcina barkeri 227. Journal of Bacteriology 178, 143-148. Cornwell WK, Cornelissen JHC, Allison SD, et al. (2009) Plant traits and wood fates across the globe: rotted, burned, or consumed? Global Change Biology 15, 2431-2449. Cowling EB, Merrill W (1966) Nitrogen in wood and its role in wood deterioration. Canadian Journal of Botany 44, 1539-1554. Crooks GE, Hon G, Chandonia JM, Brenner SE (2004) WebLogo: A sequence logo generator. Genome Research 14, 1188-1190. De'Ath G (2002) Multivariate regression trees: a new technique for modeling species- environment relationships. Ecology 83, 1105-1117. Diamond JM (1975) Assembly of species communities. In: Ecology and Evolution of Communities. (eds. Cody ML, Diamond JM), pp. 342-444. Harvard University Press, Cambridge. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19. Edgar RC (2010) Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26, 2460-2461.

102

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R (2011) UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27, 2194-2200. Faust K, Raes J (2012) Microbial interactions: from networks to models. Nature Reviews Microbiology 10, 538-550. Fischer M, Bossdorf O, Gockel S, et al. (2010) Implementing large-scale and long-term functional biodiversity research: The Biodiversity Exploratories. Basic and Applied Ecology 11, 473-485. Frey-Klett P, Burlinson P, Deveau A, et al. (2011) Bacterial-fungal interactions: hyphens between agricultural, clinical, environmental, and food microbiologists. Microbiology and Molecular Biology Reviews 75, 583-609. Fukasawa Y, Osono T, Takeda H (2009) Dynamics of physicochemical properties and occurrence of fungal fruit bodies during decomposition of coarse woody debris of Fagus crenata. Journal of Forest Research 14, 20-29. Gaby JC, Buckley DH (2011) A global census of nitrogenase diversity. Environmental Microbiology 13, 1790-1799. Gascuel O (1997) BIONJ: An improved version of the NJ algorithm based on a simple model of sequence data. Molecular Biology and Evolution 14, 685-695. Glez-Pena D, Gomez-Blanco D, Reboiro-Jato M, Fdez-Riverola F, Posada D (2010) ALTER: program-oriented conversion of DNA and protein alignments. Nucleic Acids Research 38, 14-18. Gotelli NJ, Entsminger GL (2009) EcoSim: Null models software for ecology. Version 7. Acquired Intelligence Inc. & Kesey-Bear. Jericho. Green J, Bohannan BJM (2006) Spatial scaling of microbial biodiversity. Trends in Ecology and Evolution 21, 501-507. Guindon S, Gascuel O (2003) A simple, fast, and accurate algorithm to estimate large phylogenies by maximum likelihood. Systematic Biology 52, 696-704. Hall TA (1999) BioEdit: a user-friendly biological sequence alignment editor and analysis program for Windows 95/98/NT. Nucleic Acids Symposium Series 41, 95-98. Harmon ME, Franklin JF, Swanson FJ, et al. (1986) Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research 15, 133-302. Hartigan JA, Wong MA (1979) Algorithm AS 136: A K-Means Clustering Algorithm. Journal of the Royal Statistical Society, Series C (Applied Statistics) 28 100– 108. Heilmann-Clausen J (2001) A gradient analysis of communities of macrofungi and slime moulds on decaying beech logs. Mycological Research 105, 575-596. Heilmann-Clausen J, Christensen M (2003) Fungal diversity on decaying beech logs - implications for sustainable forestry. Biodiversity and Conservation 12, 953- 973. Herrmann S, Bauhus J (2012) Effects of moisture, temperature and decompostion stage of respirational carbon loss from coarse woody debris (CWD) of important European tree species Scandinavian Journal of Forest Research. Hicks WT, Harmon ME, Myrold DD (2003) Substrate controls on nitrogen fixation and respiration in woody debris from the Pacific Northwest, USA. Forest Ecology and Management 176, 25-35. Horner-Devine MC, Silver JM, Leibold MA, et al. (2007) A comparison of taxon co- occurrence patterns for macro- and microorganisms. Ecology 88, 1345-1353. Huson DH, Bryant D (2006) Application of phylogenetic networks in evolutionary studies. Molecular Biology and Evolution 23, 254-267. 103

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Jurgensen MF, Larsen MJ, Spano SD, Harvey AE, Gale MR (1984) Nitrogen-fixation associated with increased wood decay in Douglas-fir residue. Forest Science 30, 1038-1044. Kahl T, Mund M, Bauhus J, Schulze ED (2012) Dissolved organic carbon from European beech logs: Patterns of input to and retention by surface soil. Ecoscience 19, 1-10. Katoh K, Toh H (2008) Recent developments in the MAFFT multiple sequence alignment program. Briefings in Bioinformatics 9, 286-298. Kumar S, Skjaeveland A, Orr RJS, et al. (2009) AIR: A batch-oriented web program package for construction of supermatrices ready for phylogenomic analyses. BMC Bioinformatics 10. Kyte J, Doolittle RF (1982) A simple method for displaying the hydropathic character of a protein. Journal of Molecular Biology 157, 105-132. Larsen MJ, Jurgensen MF, Harvey AE, Ward JC (1978) Dinitrogen fixation associated with sporophores of Fomitopsis pinicola, Fomes fomentarius, and Echinodontium tinctorium. Mycologia 70, 1217-1222. Lartillot N, Lepage T, Blanquart S (2009) PhyloBayes 3: a Bayesian software package for phylogenetic reconstruction and molecular dating. Bioinformatics 25, 2286- 2288. Lassmann T, Frings O, Sonnhammer ELL (2009) Kalign2: high-performance multiple alignment of protein and nucleotide sequences allowing external features. Nucleic Acids Research 37, 858-865. Lassmann T, Sonnhammer ELL (2006) Kalign, Kalignvu and Mumsa: web servers for multiple sequence alignment. Nucleic Acids Research 34, 596-599. Li W, Godzik A (2006) Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658-1659. Litton CM, Raich JW, Ryan MG (2007) Carbon allocation in forest ecosystems. Global Change Biology 13, 2089-2109. Lonsdale D, Pautasso M, Holdenrieder O (2008) Wood-decaying fungi in the forest: conservation needs and management options. European Journal of Forest Research 127, 1-22. Maddison WP, Maddison DR (2011) Mesquite: a modular system for evolutionary analysis. Version 2.75. Merrill W, Cowling EB (1966) Role of nitrogen in wood deterioration: amounts and distribution of nitrogen in tree stems. Canadian Journal of Botany 44, 1555- 1580. Miller MA, Pfeiffer W, Schwartz T (2010) Creating the CIPRES Science Gateway for inference of large phylogenetic trees. In: Proceedings of the Gateway Computing Environments Workshop (GCE), pp. 1-8, New Orleans. Müller J, Engel H, Blaschke M (2007) Assemblages of wood-inhabiting fungi related to silvicultural management intensity in beech forests in southern Germany. European Journal of Forest Research 126, 513-527. Noy NF, Shah NH, Whetzel PL, et al. (2009) BioPortal: ontologies and integrated data resources at the click of a mouse. Nucleic Acids Research 37, W170-W173. Poly F, Monrozier LJ, Bally R (2001) Improvement in the RFLP procedure for studying the diversity of nifH genes in communities of nitrogen fixers in soil. Research in Microbiology 152, 95-103. Ramirez-Flandes S, Ulloa O (2008) Bosque: integrated phylogenetic analysis software. Bioinformatics 24, 2539-2541. 104

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Raymond J, Siefert JL, Staples CR, Blankenship RE (2004) The natural history of nitrogen fixation. Molecular Biology and Evolution 21, 541-554. Rayner ADM, Boddy L (1988) Fungal communities in the decay of wood. Advances in Microbial Ecology 10, 115-166. Rondeux J, Sanchez C (2010) Review of indicators and field methods for monitoring biodiversity within national forest inventories. Core variable: Deadwood. Environmental Monitoring and Assessment 164, 617-630. Seidler RJ, Aho PE, Evans HJ, Raju PN (1972) Nitrogen fixation by bacterial isolates from decay in living White fir trees [Abies Concolor (Gord. and Glend.) Lindl.]. Journal of General Microbiology 73, 413-416. Shaffer BT, Widmer F, Porteous LA, Seidler RJ (2000) Temporal and spatial distribution of the nifH gene of N2 fixing bacteria in forests and clearcuts in Western Oregon. Microbial Ecology 39, 12-21. Shannon P, Markiel A, Ozier O, et al. (2003) Cytoscape: A software environment for integrated models of biomolecular interaction networks. Genome Research 13, 2498-2504. Silvester WB, Sollins P, Verhoeven T, Cline SP (1982) Nitrogen-fixation and acetylene-reduction in decaying conifer boles - Effects of incubation-time, aeration, and moisture-content. Canadian Journal of Forest Research 12, 646- 652. Spano SD, Jurgensen MF, Larsen MJ, Harvey AE (1982) Nitrogen-fixing bacteria in Douglas-fir residue decayed by Fomitopsis pinicola. Plant and Soil 68, 117-123. Stone L, Roberts A (1990) The Checkerboard Score and Species Distributions. Oecologia 85, 74-79. Stöver BC, Müller KF (2010) TreeGraph 2: combining and visualizing evidence from different phylogenetic analyses. BMC Bioinformatics 11. Tamura K, Peterson D, Peterson N, et al. (2011) MEGA5: Molecular Evolutionary Genetics Analysis Using Maximum Likelihood, Evolutionary Distance, and Maximum Parsimony Methods. Molecular Biology and Evolution 28, 2731- 2739. Thompson TA, Thorn RG, Smith KT (2012) Hypholoma lateritium isolated from coarse woody debris, the forest floor, and mineral soil in a deciduous forest in New Hampshire. Botany 90, 457-464. Ulrich W (2008) Pairs- a FORTRAN programm for studying pair wise associations in ecological matrices. www.uni.torun.pl/~ulrichw. Ulrich W, Zalewski M (2006) Abundance and co-occurrence patterns of core and satellite species of ground beetles on small lake islands. Oikos 114, 338-348. Valaskova V, de Boer W, Klein Gunnewiek PJA, Pospisek M, Baldrian P (2009) Phylogenetic composition and properties of bacteria coexisting with the fungus Hypholoma fasciculare in decaying wood. ISME Journal 3, 1218-1221. Volkenant K (2007) Totholz als Lebensraum von Mycozönosen im fortschreitenden Zersetzungsprozess - Eine Chronosequenzstudie an Fagus sylvatica-Totholz im Nationalpark Kellerwald-Edersee, Dissertation, Universität Kassel. Wang QJ, Quensen JF, Fish JA, et al. (2013) Ecological patterns of nifH genes in four terrestrial climatic zones Explored with targeted metagenomics using FrameBot, a new informatics tool. Mbio 4. Wazny H, Wazny J (1964) Über das Auftreten von Spurenelementen im Holz. Holz als Roh- und Werkstoff 22, 299-304.

105

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Weißhaupt P (2012) Nitrogen uptake of saprotrophic basidiomycetes and bacteria. Dissertation, TU Berlin Whitaker RJ (2009) Evolution: Spatial Scaling of Microbial Interactions. Current Biology 19, 954-956. Young JPW (2005) The phylogeny and evolution of nitrogenases. In: Genomes and genomics of nitrogen-fixing organisms (eds. Palacios R, Newton WE), pp. 221- 241. Springer, Dordrecht. Zehr JP, Jenkins BD, Short SM, Steward GF (2003) Nitrogenase gene diversity and microbial community structure: a cross-system comparison. Environmental Microbiology 5, 539-554. Zehr JP, McReynolds LA (1989) Use of degenerate oligonucleotides for amplification of the nifH gene from the marine Cyanobacterium Trichodesmium thiebautii. Applied and Environmental Microbiology 55, 2522-2526. Zhang HB, Yang MX, Tu R (2008) Unexpectedly high bacterial diversity in decaying wood of a conifer as revealed by a molecular method. International Biodeterioration & Biodegradation 62, 471-474. Zwickl DJ (2006) Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Dissertati- on, University of Texas at Austin.

106

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

3.6 Supplemental information

Table S3.3: Sequence percentage identity of OTUs taxonomically assigned through BLASTn against GenBank (uncultured/ environmental sample sequences excluded). 19 OTUs were assigned to Rhizobial- es at a 95% similarity threshold (65 OTUs at ≥ 90%). A total of 80 OTUs were identified to genus level. Order Genus 100% 99% 98% 97% 96% 95% 90-94% Bradyrhizobium 1 1 2 4 28 Methylocella 1 Methyloferula 3 1 1 Beijerinckiaceae* 1 1 Methylocapsa 2 1 1 Rhizobiales Rhodomicrobium 1 Methylococcus 1 Methylobacterium 2 Methylocystis 9 Xanthobacter 4 Rhodocyclales Azospira 1 Pseudomonadales Pseudomonas 1 Telmatospirillum 1 2 2 Rhodospirillales Azospirillum 1 1 Sphingomonadales Sphingomonas 2 Burkholderia 1 1 Burkholderiales Ideonella 1 Pelomonas 1

107

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Table S3.4: List of sporocarps identified on the respective deadwood trees A) Fagus sylvatica B) Picea abies.

A) Fagus sylvatica

CWD_ID Fagus 08270 08277 08280 08283 08285 08286 08295 08296 08297 08298 08807 08810 08815 08818 08821 08827 08828 08832 08835 08842 08845 08846 sylvatica Antrodiella hoehnelii X Armillaria spec. X X X X X X X X X X X X Ascocoryne cylichni‐ X um Ascocoryne spec. X X X X X X X X X X Ascodichaena rugosa X Athelia epiphylla X X Bispora monilioides X X X X Bisporella citrina X X X Bjerkandera adusta X X Bjerkandera spec. X Botryobasidium aure‐ X um Botryobasidium X vagum Calocera cornea X Ceriporia excelsa X Ceriporia purpurea X Ceriporia spec. X X Ceriporiopsis spec. X Coprinellus micaceus X Coprinus spec. X Dacrymyces spec. X Datronia mollis X Eutypella quaternata X Exidia glandulosa X Fomes fomentarius X X X X Fomitopsis pinicola X X X Galerina spec. X X Ganoderma applana‐ X X X tum Grandinia spec. X Hyalorbilia inflatula X Hymenoscyphus spec. X Hyphodontia sambuci X

108

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Hypholoma fascicula‐ X re Hypholoma lateritium X Hypochnicium polo‐ X nense Hypocrea minutispora X X Hypocrea spec. X Hypoxylon cohaerens X X X X X Hypoxylon fragiforme X X X X Hypoxylon rubigino‐ X X X X sum Hypoxylon spec. X X Kretzschmaria deusta X X X X X X Kuehneromyces X X mutabilis Lenzites betulina X Lycoperdon pyriforme X Marasmiellus spec. X Marasmius alliaceus X X X X X X Megacollybia platy‐ X X X phylla Melogramma spini‐ X ferum Mollisia ligni X X Mollisia spec. X Mutatoderma muta‐ X tum Mycena galericulata X X Mycena speirea X X Mycena crocata X X Mycena renati X X Mycena rubromargi‐ X nata Mycena spec. X Nectria spec. X Nemania confluens X Nemania serpens X X X X X X Neobulgaria pseu‐ X doombrophila Neobulgaria pura X X Neodasyscypha cerina X X Orbilia spec. X X X X Oudemansiella muci‐ X da Panellus serotinus X X X Peniophorella guttu‐ X lifera Peziza obtusapiculata X Phanerochaete veluti‐ X na 109

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Phellinus ferruginosus X Phlebia livida X Physisporinus vitreus X Pluteus pouzerianus X Pluteus phlebophorus X X Pluteus salicinus X Pluteus spec. X Postia subcaesia X Psathyrella rostellata X X Pycnoporus cinnaba‐ X rinus Ruzenia spermoides X Schizopora paradoxa X Scopuloides hydnoi‐ X X X X des Scutellinia scutellata X Sebacina incrustans X Sistotrema brinkman‐ X X nii Stereum rugosum X Stereum spec. X Stereum subtomento‐ X sum Stypella grilletii X Subulicystidium longi‐ X sporum Tomentella bryophila X Trametes gibbosa X X X Trametes hirsuta X X X Trametes versicolor X X X X X Trechispora nivea X Tubaria furfuracea X Xenasmatella vaga X X Xylaria hypoxylon X X X X X X X X X X X

110

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

B) Picea abies

CWD_ID Picea 08271 08273 08276 08281 08288 08289 08290 08294 08801 08804 08811 08812 08813 08817 08819 08820 08822 08823 08824 08825 08826 08833 08834 abies Amylostereum areolatum X X X Amylostereum spec. X Armillaria spec. X X X X X X Athelia decipiens X Athelia epiphylla X Boidinia furfuracea X Botryobasidium subcoronatum X X X X Botryobasidium vagum X X X X X Botryobasidium laeve X Cabalodontia sub- cretacea X Camarops tubulina X Ceriporiopsis mucida X Chrysomphalina grossula X Clavulicium de- lectabile X Dacrymyces stillatus X X X X Dacryobolus sudans X Exidiopsis spec. X Fomitopsis pinicola X X X X X X Galerina marginata X Galerina spec. X Ganoderma applanatum X Gloeophyllum sepi- arium X Gymnopilus penet- rans X X Henningsomyces candidus X Heterobasidion annosum X X X X X Hyphoderma ar- gillaceum X X X Hyphodontia alutaria X Hyphodontia brevi- seta X Hyphodontia nespo- ri X X Hyphodontia palli- dula X Hyphodontia spathulata X 111

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Hyphodontia spec. X Hypholoma capnoi- des X X X Hypocrea pulvinata X Ischnoderma ben- zoinum X Marasmius andro- saceus X Mucronella spec. X Mycena stipata X Mycena cyanorhiza X Mycena metata X X X X X X Mycena rubromarginata X X Mycena spec. X X Neobulgaria spec. X Peniophorella palli- da X X Peniophorella praetermissa X X Pholiota flammans X Physisporinus spec. X Piloderma bicolor X Postia caesia X X X X X Postia stiptica X X Postia tephroleuca X Pycnoporellus fulgens X X Resinicium bicolor X X X X X X X X X X X Skeletocutis carneo- grisea X Skeletocutis kuehne- ri X Skeletocutis nivea X Skeletocutis spec. X X Steccherinum ochraceum X Thanatephorus fusisporus X Tomentella lilacino- grisea X Tomentella spec. X Trechispora hy- menocystis X X Trechispora mollu- sca X Trechispora spec. X Trichaptum abieti- num X X X Trichaptum fuscovi- olaceum X Vesiculomyces citrinus X X Xenasma tulasnel- loidea X X 112

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Xenasmatella vaga X X X X X X X

113

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. S3.7: Sampling scheme visualized using Treemap v. 3.1.0. (Macrofocus, Zurich, Switzerland) in squarified layout. Items are grouped by management type. Treemap cell size is proportional to mass loss in% (smaller cells = less decayed logs) Colors represent tree species. (red = Picea abies, green = Fagus sylvatica). Numbers indicate the ID of the deadwood item.

114

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. S3.8: Splitstree reticulogram. The three major parts of the phylogeny (compare phylogenetic tree in Fig. 3.3) are labeled here.

Fig. S3.9: Bargraphs including standard errors displaying nifH OTU richness (A) and nitrogen content per density unit (B) within deadwood tree species.

115

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. S3.10: Scatterplot displaying number (= richness) of fruiting fungal species per deadwood log. Solid vertical lines display mean values of richness, dashed vertical lines median richness per tree species (green = Fagus sylvatica, red = Picea abies). Heatmapped bars to the left and right display density probability as calculated by kernel density estimation using the denstrip package in R (Jackson CH (2008) Displaying uncertainty with shading. Am Stat 62: 340-347.)

Fig. S3.11: Relative abundances (left) of Basidiomycota and Ascomycota on deadwood logs of Fagus sylvatica and Picea abies and mean number of sporocarps per tree species (right).

Fig. S3.12: Interrelation of nifH OTU richness and remaining mass after decay in% (A) and water content in% (B) and water content in% and remaining mass after decay in% (C) on logs of Fagus sylvatica.

116

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. S3.13: Non-random sporocarp – nifH OTU community assembly assessed by C-score distribution and Checkerboard index for observed and expected/ randomized species occurrences.

Fig. S3.14: Boxplots including median, upper and under quartiles and whiskers displaying the interrelation of deadwood species and log transformed carbon content per density unit (A) and log transformed wood density (B).

117

3 Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi

Fig. S3.15: Effects of remaining mass after decay in% on log-transformed nitrogen content per density unit (g*cm³) and N concentration in g*g-1. Interrelations are displayed separately per wood species.

118

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

4 A pyrosequencing insight into sprawling bacterial

diversity and community dynamics in decaying

deadwood logs of Fagus sylvatica and Picea abies

4.1 Introduction

Deadwood is an important structural component in forest ecosystems. It provides shelter and nutrition to various organisms, primarily fungi and saproxylic insects (Harmon et al. 1986; Lassauce et al. 2011; Stokland et al. 2012). It also partakes in numerous ecosystem functions (Cornwell et al. 2009), including carbon sequestration and nutrient cycling (Chambers et al. 2000; Herrmann & Bauhus 2012; Kahl et al. 2012; Litton et al.

2007). Many investigations have focused on the diversity and composition of fungal communities, their roles in wood decomposition (Kubartova et al. 2012; Rajala et al.

2012), and their interactions with different forest management regimes (Müller et al.

2007; Purahong et al. 2014). Conversely, the role of prokaryotes in deadwood and related ecosystem processes has only been examined in a few case studies such as the investigation into bacterial communities in deadwood of an East Asian pine species (Zhang et al. 2008) and the presence of coexisting bacteria with Hypholoma fasciculare sampled in seven tree stumps by Valášková et al. 2009. Substrate properties such as nutrient and water content have been shown to strongly influence wood colonization by microbes

(Rajala et al. 2012; Volkenant 2007). Greaves 1971 first developed a concept concerning a functional classification of wood-inhabiting bacteria: with (1) bacteria that affect

119

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies permeability but do not cause losses in material strength, (2) bacteria that attack wood structures, (3) bacteria that act as integral synergistically members of the total microflora and (4) the “passive” bacteria, which may act as antagonists to other bacteria. Water- logged and oxygen-depleted wood is mainly degraded by bacteria (Kirker et al. 2012) through a process that may be so slow that trees remain intact long enough to think of salvaging submerged logs (Tenenbaum 2004). However, they may be inefficient as wood decayers on their own, though bacteria, especially members of the Actinobacteria are expected to be among the early initial colonizers of deadwood and to permeate or even degrade its lignified cell walls via secretion of cellulases (Daniel & Nilsson 1998;

Lynd et al. 2002; Clausen 1996). The composition of the primary wood-inhabiting bacterial communities may also be a consequence of the composition of the surrounding soil bacterial communities (Sun et al. 2013). Associations among saprotrophic or wood- decaying Basidiomycota and bacteria in deadwood have been examined in several studies and reviews published over the last three decades (Blanchette & Shaw 1978;

Clausen 1996; de Boer & van der Wal 2008; Folman et al. 2008). Both antagonistic (de

Boer et al. 2010; Folman et al. 2008) and mutualistic interactions (Lindahl & Finlay

2006; Merrill & Cowling 1966) have been observed during wood decomposition processes. A recent network analysis revealed non-random co-occurrence patterns of bacterial nitrogenase-encoding nifH genes with fungal sporocarps on deadwood logs of

Fagus sylvatica and Picea abies (Hoppe et al. 2014). This finding provided further evidence of potential mutualistic interactions between fungi and methylotrophic N-fixing bacteria that consume methanol, which is a by-product of enzymatic lignin degradation

(Dedysh et al. 2001; Merrill & Cowling 1966; Vorobev et al. 2009). Hervé et al. (2014) investigated changes in bacterial community structure over time in deadwood inoculated with the lignin-degrading fungus Phanerochaete chrysosporium. They discovered varia- 120

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies tions in bacterial community composition across time to be incidental, but also identified members of the Burkholderiaceae to be always present in the mycosphere.

Apart from these studies, little is known about the community dynamics of wood- inhabiting bacteria during wood decay in temperate ecosystems. The study we here present is among the first to investigate bacterial diversity and community structure in deadwood under field conditions and applying deep 16S rDNA metabarcoding. Specifi- cally, it compares the bacteria in deadwood logs of two common temperate timber tree species grown in geographic proximity, the deciduous Fagus sylvatica and the conifer

Picea abies, at different stages of decay and under different forest management. Deci- duous and coniferous woods have very different physico-chemical properties (Carlquist

2001; Fengel & Wegener 1983). We anticipated that the structures of the bacterial communities would depend strongly on the properties of the deadwood and the identity of the tree species from which it derived. The primary objective of this study was therefore to determine which of these wood properties correspond or determine the composition and diversity of deadwood-inhabiting bacterial communities, and to identify the key players in the bacterial communities in the two deadwood types on at least the family level. In relation to a previously conducted study on the distribution of nifH genes in deadwood (Hoppe et al. 2014), we additionally assumed that N-fixing bacteria from the order Rhizobiales were more abundant during the intermediate stages of wood decay, when fungal sporocarp richness is known to be highest and the provision of nitrogen is crucial (Hoppe et al. 2014).

121

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies 4.2 Material and methods

4.2.1 Experimental design, deadwood selection and physico-chemical properties

The study was conducted on plots of the German Biodiversity Exploratories (Fischer et al. 2010) located in the “Schwäbische Alb” UNESCO Biosphere Reserve in southwestern Germany according to the sampling scheme displayed in Hoppe et al. 2014. We surveyed deadwood logs in nine very intensively investigated 1 ha plots (VIPs), with three plots representing three different forest types and management regimes, respectively: (i) unmanaged beech forests, where timber harvesting stopped several decades

(20 - 70 years) ago, (ii) managed beech forests dominated by Fagus sylvatica and (iii) managed spruce forests dominated by Picea abies, which in both cases are characterised by uniform tree species composition, forest structure and site conditions. In April 2009, a set of 48 logs, equally representing the two tree species 24 logs per tree species (P. abies and F. sylvatica) located on the forest floor were randomly selected and their properties (length, diameter, tree species, e.g.) were characterized (see Table S4.3). The logs were selected to ensure that some of the Fagus logs were located in Picea- dominated plots and vice versa. In June 2009, 3-7 wood samples were collected from each log (according to log length; compare Purahong et al. 2014 and supplementary information) using a cordless Makita BDF451 drill (Makita, Anja, Japan) equipped with a 2 x 42 cm wood auger as described by Hoppe et al. 2014 (further details provided in supplementary information; Fig S4.4). The upper surface layer and bark of the deadwood was removed to avoid external bacterial contamination. Prior to analysis, the wood samples were weighed, dried at 60°C to constant mass, and reweighed. The concentrations of C and N in wood samples were determined by total combustion using a

Truspec elemental analyzer (Leco, St. Joseph, MI, USA). The samples’ densities and 122

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies relative wood moisture contents were calculated based on their dry masses. Sample pH values were determined by shaking 1 g of dried wood in 10 mL of distilled water for

120 minutes and measuring the pH of the resulting aqueous extract. Each deadwood log was assigned to one of 4 decay classes based on its remaining mass (%) by k-means cluster analysis as described by Kahl et al. 2012 and Hoppe et al. 2014. Higher decay classes corresponded to more extensive decay (Table S4.4).

4.2.2 DNA Isolation, PCR and Pyrosequencing

Total community DNA was isolated from 1 g of each previously homogenized wood sample using a modified CTAB-protocol (Doyle & Doyle 1987) as described by Hoppe et al. 2014. All DNA extracts from the wood subsamples of the 48 logs were pooled into a composite extract prior to PCR amplification. In total, 47 16S rDNA gene amplicon libraries were obtained while no amplified product was found from one sample of

Fagus sylvatica. For the amplicon library production we used fusion primers designed with pyrosequencing primer B, a barcode and the primer 341f (Muyzer et al. 1993) as a forward primer and pyrosequencing primer A and the 907r (Lane 1991) to amplify the

V3-V5 region of the eubacterial 16S rDNA gene. The primers were barcoded with a set of 10 nt MID-barcodes provided by Roche (Roche Applied Science, Mannheim, Ger- many). For each composite DNA extract the amplicon libraries was amplified separately by PCR in triplicate 50 µl reaction mixtures containing 25 µl 2x GoTaq Green Mas- termix (Promega, Madison, WI, USA), 25 µM of each primer and approximately 20 ng template DNA. PCR was performed with an initial denaturation period of 1 min at 98°C followed by 30 cycles of 95°C for 45 s, 57°C for 45 s, and 72°C for 1 min 30 s, then a final elongation step at 72°C for 10 min. After checking the quality of the PCR products by separation on a 1.5% agarose gel, the replicates were pooled and purified by gel ex-

123

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies traction using the QIAquick Gel Extraction Kit (QIAGEN, Hilden, Germany). The purified DNA was quantified using a fluorescence spectrophotometer (Cary Eclipse, Agilent

Technologies, Waldbronn, Germany). An equimolar mixture of each library was subjected to unidirectional pyrosequencing from the 907r ends of the amplicons, using a

454 Titanium amplicon sequencing kit and a Genome Sequencer FLX 454 System (454

Life Sciences/ Roche Applied Biosystems, Mannheim, Germany) at the department of

Soil Ecology, UFZ.

4.2.3 Bioinformatic analysis

We performed multiple levels of sequence quality filtering. The 454 bacterial 16S sequences were extracted based on 100% barcode similarity. Sequences were trimmed from barcodes and sized to a minimum length of 350 nt to cover the V4-V5 region of the 16S rRNA gene using mothur (Schloss et al. 2009). Sequence reads with an average quality score of <20, bases and homo-polymers of > 8 bases were removed. Unique good quality sequences from the dataset were filtered and checked for chimeras using the uchime algorithm, as implemented in mothur. To avoid sampling size effects, the number of reads per sample was normalized to 1,837 for each data set by randomly subsampling to the lowest number of reads among samples.

The revised complete sequence dataset was then clustered and assigned to OTUs using

CD-HIT-EST of CD-HIT version -4.5.4 (Li & Godzik 2006) at a 97% threshold of pairwise sequence similarity. We used GAST (global alignment for sequence taxonomy) (Huse et al. 2008) to taxonomically assign the OTUs against the arb-silva database

114 (Quast et al. 2013) as downloaded in August 2013.

124

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

4.2.4 Statistical analyses

We performed a procrustes analysis (Peres-Neto & Jackson 2001) using the protest function in vegan (Oksanen 2013) to test the impact of excluding the rare taxa (singletons, doubletons, tripletons). Procrustes analysis indicated that removing of rare taxa had no effect on the results obtained (procrustes correlation R = 0.94, P = 0.001), so singletons, doubletons, and tripletons were discarded prior to further analysis. OTUs stemming from Cyanobacteria related to chloroplasts were also discarded (85 OTUs incl. 56 singletons).

The effects of tree species, decay class and forest management type on the 16S OTU community structure in the sampled logs were analyzed by perMANOVA. ANOVA was used to assess the influence of tree species and decay class on bacterial OTU richness and the two richness estimators. We performed one-way analysis of variance

(ANOVA) to identify significant (P < 0.05) differences between mean OTU richness in association with the respective tree species. ANOVA was coupled with Shapiro-Wilk’s

W test for normality and Levene’s test for equality of group variances. In addition, we assessed the impact of forest management intensity on bacterial diversity using two land use intensity indices, SMI (Silvicultural Management Intensity indicator) (Schall &

Ammer 2013) and ForMi (Forest Management Intensity index) (Kahl & Bauhus 2014), that were developed in parallel within the Biodiversity Exploratories research community. Pearson Rank correlations between the relative abundances of dominant bacterial phyla (including proteobacterial subphyla) and selected wood properties were performed in PAST (Hammer et al. 2001). One-way analysis of similarities (ANOSIM) calculations based on four commonly used distance measures in conjunction with data on OTU abundance and presence/absence was performed in PAST (Hammer et al.

2001) to test for significant differences in bacterial community structures and composi- 125

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies tions among different tree species and decay classes, respectively. Assessments of statistical significance were based on 999 permutations and P values were Bonferroni- corrected. NMDS was conducted in R to investigate the influence of the following wood physico-chemical parameters on bacterial community structure: decay classes, the concentrations of the macronutrients C and N, relative wood moisture, wood density, remaining mass, and pH. Goodness-of-fit statistics (R2) for environmental variables fitted to the NMDS ordinations of the bacterial communities were calculated using the envfit function of the “vegan” package, with P values being based on 999 permutations

(Oksanen 2013). Similarity Percentage analysis (SIMPER) was performed in PAST

(Hammer et al. 2001) to identify the important bacterial families responsible for the observed clustering of samples.

4.3 Results

4.3.1 Wood properties

In addition to assigning each log to a decay class, we determined their C and N contents, the concentrations of both elements per unit wood density (g/cm³), and also their relative wood moisture contents and pH values (Tables S4.4 and S4.5). The C/N ratio in

Picea logs ranged from 629.7 ± 48.4 to 422.9 ± 54.4 and was thus substantially greater and more variable than that in Fagus logs (376.9 ± 11.5 to 193.7 ± 15.6). In logs of both tree species, the C/N ratio decreased slightly with increasing decay class. This decrease was largely due to an increase in the logs’ N concentration as they decayed (from 0.13 to 0.25% on F. sylvatica and 0.08 to 0.14% on P. abies, Table S4.4). Due to the significant decrease of wood density (from 0.5 to 0.2 g/cm³ on F. sylvatica and 0.35 to 0.15 g/cm³ on P. abies), we also observed a decrease in total N and C content per density unit

126

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

(Table S4.4). There was no significant difference in wood moisture between the two tree species, but the moisture contents of both species’ logs increased significantly with the extent of decay, from 52.2% ± 5.7 to 155.2% ± 9.1 in F. sylvatica logs and 48.7% ±

11.6 to 163.1% ± 24.6 in P. abies. Finally, the pH of F. sylvatica logs was significantly higher (P < 0.001) than that of P. abies logs; this trend was independent of the logs’ state of decay.

4.3.2 Sequence analyses

In total, 125,183 reads were obtained from the 47 amplicon libraries by 454 pyrosequencing of the deadwood samples. Sequences were initially quality checked, trimmed

(115,750 sequences), normalized per sample (1,837 reads per sample) and screened for potential chimeras. CD-HIT clustering of the remaining 73,099 sequences (discarding potential chimeras) yielded 7,388 OTUs at a 97% cutoff, 5,016 of which were singletons, 807 doubletons, and 368 tripletons. A total of 85 OTUs (among them 56 singletons) stemming from Cyanobacteria related to chloroplasts were also removed from the dataset.

In total, 61,831 sequence reads distributed to 1,180 OTUs were retained for analysis.

Taxonomic assignments were achieved for 99.75% of these filtered OTUs (61,776 sequences (99.9%)) at the phylum level (including proteobacterial classes). 1,056 (58,498 sequences (94.6%)), 901 (53,732 sequences (86.9%)) and 633 (39,645 (64.1%)) OTUs were classified at the level of order, family, and genus level, respectively.

4.3.3 Bacterial 16S rRNA diversity and richness

Species richness across all samples ranged from 105 to 378 OTUs with an average of

258 ± 56. We did not observe any significant differences in mean OTU richness bet-

127

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies ween the two tree species (P = 0.52) (Fig. S4.5). One-way analyses of the variance in mean bacterial species richness for different decay classes revealed an increase from decay class 1 to 3 in Fagus logs (Fig S4.5): the species richness in decay class 1 logs

(191.6 ± 21.1) was significantly lower than in decay class 3 (291.9 ± 19.4). We did not detect any significant variation in OTU richness in Picea logs, although the species richness in the two later stages of decay (271 ± 17) was slightly higher than that in stages 1 and 2 (254 ± 14, Fig. S4.5). This increase is reflected in the observed correlations between OTU richness and the remaining mass of the logs (which provides a measure of their extent of decay). Bacterial OTU richness correlated significantly and positively with the extent of wood decay (P < 0.0005, R² = 0.45) in Fagus logs but not in Picea logs (P = 0.25, R² = 0.06) (Fig. S4.6).

The mean relative abundance of dominant bacterial phyla (including proteobacterial classes) did not differ greatly between the two tree species (Kronafiles S4.11, S4.12 and

S4.13 in supplementary information). Alphaproteobacteria were dominant on both

Fagus (38.9 ± 3.1%) and Picea logs (41.6± 2.1%); other abundant phyla included Ac- idobacteria (14.9 ± 2% and 20.9 ± 0.7%) and Actinobacteria (10.5 ± 0.7% and 11.4 ±

0.1%). However, there were significant differences in the relative abundances of these phyla between decay classes in logs of the same tree species (Fig. 4.1, Table S4.6). The relative contribution of Alphaproteobacteria increased from decay class 1 to 4 in both

Fagus and Picea logs (from 28.1 ± 3% to 50.7 ± 1.9% and 30.4 ± 3.7% to 44.1 ± 1.5%, respectively) (Tab. S4.6). The relative abundances of Acidobacteria were consistent across decay classes but the contribution of Firmicutes decreased significantly from decay class 1 to 4 in both Fagus (9.6 ± 3.2% to 0.6 ± 0.2%) and Picea (5.2 ± 0.8% to

0.4 ± 0.2%) logs.

128

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. 4.1: Relative abundances of phylogenetic groups (bacterial phyla including proteobacterial classes) in deadwood from two species (Fagus sylvatica = FASY, Picea abies = PIAB) in different stages of decay (decay classes 1-4). OTUs that could not be taxonomically assigned at the phylum/subphylum level are reported as “Others” and comprise 0.006% of all sequences. The category “other” also includes all OTUs with <1.5% relative sequence abundance.

These shifts of relative abundances of the dominant phyla have further been observed performing rank abundance correlations (Table S4.7). This analysis revealed that the remaining mass per log had a significantly negative impact on the abundances of Al- phaproteobacteria and Deltaproteobacteria (on F. sylvatica and P. abies), and Cyano- bacteria (on F. sylvatica). In turn, Actinobacteria, Firmicutes (both on F. sylvatica and

P. abies), Gammaproteobacteria (on logs of and F. sylvatica) and Bacteroidetes (on logs of P. abies) significantly decreased during wood decay. Analogous results were observed for the impact of relative wood moisture on the abundances of the respective phyla, since this parameter significantly increases with mass loss (see Table S4.5). The abundances of Alphaproteobacteria (on Fagus sylvatica) and Deltaproteobacteria (on both tree species) correlated negatively and significantly to C/N in contrast to Firmicu- tes. Members of this phylum were also found to be positively and significantly impacted

129

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies by higher pH on logs of Picea abies, whereas Acidobacteria correlated negatively and significantly to pH on logs of Fagus sylvatica (Table S4.7).

At the order level, members of the Rhizobiales were dominant in Fagus logs (22%), followed by Acidobacteriales (13%) and Rhodospirillales (11%) (Kronafiles S4.11 and

S4.12 in supplementary information, Fig. 4.2). In Picea logs, Rhodospirillales (20%) were dominant, followed by Acidobacteriales (18%) and Rhizobiales (17%). As shown in Fig. 4.2, we also observed a significant increase of Rhizobiales from decay class one to three in Fagus logs (15.4 ± 5.8% to 26.9 ± 10.1%). Similarly, for Picea deadwood, the contribution of Rhizobiales went from 14.4 ± 6.4% in decay class 1 to 19.9 ± 6.6% in decay class 4. The most abundant OTU in our dataset appeared to be affiliated with the methanotrophic Methylovirgula genus from the family Beijerinckiaceae. In addition, there were another 5 highly abundant OTUs that were assigned to methanotrophic bacteria of the genera Methyloferula, Methylocella and Methylocystis. Pearson rank correlations revealed that Methylovirgula abundance had a significant negative correlation with the C/N ratio in F. sylvatica logs (R² = 0.38, P = 0.04) (Table S4.8). We also found that the abundance of Methylovirgula and Methyloferula correlated significantly and negatively with the remaining log mass but positively with the logs’ moisture content (Table

S4.8). No such correlations were observed for Picea logs. Fig. S4.7 further clearly illus- trates the increase in the relative abundance of this dominant OTU as decay proceeds.

130

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. 4.2: Relative abundances of three dominant phylogenetic groups (bacterial orders) in deadwood of the two studied tree species (Fagus sylvatica = FASY, Picea abies = PIAB) in different stages of decay (decay classes 1-4).

4.3.4 Effect of forest management on bacterial richness

Our results also revealed a significant impact of forest management regimes on bacterial

OTU richness. The analysis of forest management regimes’ influence was based on separated data for Fagus and Picea logs in the respective forest plots. When considering

Fagus logs in spruce plots, and vice versa (Fig. S.8), we observed that the highest OTU richness in Fagus logs was in the unmanaged plots (P < 0.001). While the Picea logs with the greatest OTU richness were those in managed spruce forests (269.2 ± 14.5), their richness was not significantly different to that for spruce deadwood in managed beech forests (248.2 ± 17.1) or unmanaged beech forests (266.8 ± 12.1). We also found that the richness of bacterial OTUs in Fagus logs was negatively associated with land use intensity (Fig. S4.9) by calculating its correlations with the intensity indices SMI

(R² = 0.31, P = 0.02) and ForMI (R² = 0.21, P = 0.07).

131

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

4.3.5 Bacterial community structure and variation with progressing wood decay

As revealed by perMANOVA and NMDS tree species and decay classes significantly explained the observed variation in bacterial community structure (P = 0.0001), while forest management type did not (Table 4.1, Fig. 4.3). The importance of tree species was further confirmed with one-way ANOSIM using either relative abundance (Bray-

Curtis or Euclidean) or presence/ absence (Jaccard or Sörensen) data (Table S4.9; always P = 0.001). Furthermore, physico-chemical properties of the different types of deadwood were found to correlate significantly with bacterial community structure (as presented by dominant bacterial order; Fig. 4.3). Specifically, the decay class, remaining mass, volume, density, relative wood moisture, pH, C/N ratio and the concentrations and contents of C and N in the logs contributed significantly (P = 0.043-0.0001) to the observed variation in bacterial community structure in both tree species (Table 4.2).

132

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Table 4.1: Results of perMANOVA analysis of the Bray-Curtis dissimilarities for bacterial OTU community structure in relation to tree species, decay class (assigned based on the remaining mass of the log in question), management regime, and their interaction, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseudo-F = F value by permutation. Boldface indicates statistical significance (P < 0.05); P-values are based on 999 permutations (i.e. the lowest possible P-value is 0.001). Df SS F R² P Tree species 1 0.8052 5.6484 0.09982 0.0001 Decay class 1 0.6076 4.2622 0.07532 0.0001 Management type 2 0.3967 1.3914 0.04918 0.0527 Tree species x Decay class 1 0.2377 1.6674 0.02947 0.0315 Tree x Management type 2 0.4011 1.4068 0.04972 0.0466 Decay class x Management type 2 0.3474 1.2186 0.04307 0.1492 Tree species x Decay class x Management type 2 0.2815 0.9872 0.03489 0.4543 Residuals 35 4.9895 0.61853 Total 46 8.0667 1

At the tree species level, bacterial community structure correlated significantly with decay class, remaining mass, wood density, C/N ratio and C content. Wood volume, C

concentrations, N and C content and pH were important in Fagus deadwood, while the

C/N ratio contributed significantly to explaining the variation in bacterial community

structure in Picea deadwood (Table 4.2). Similarity percentage analysis (SIMPER) for

bacterial families revealed contrasting patterns for the two tree species/ decay classes. In

Fagus logs, members of Acidobacteriaceae (Acidobacteria) explained roughly 17% of

the community variation among the 4 decay classes (Fig. S4.10). Burkholderiaceae (Be-

taproteobacteria), which accounted for 5% of the total bacteria on Fagus logs, ex-

plained roughly 9% of the total variation across different decay classes. This family

appeared to be dominant in decay class 1 logs.

133

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Table 4.2: Goodness-of-fit statistics (R2) for parameters fitted to the non-metric multidimensional scaling (NMDS) ordination of bacterial community structure. The significance estimates were based on 999 permutations. Significant factors (Bonferroni corrected P < 0.05) are indicated in bold. Marginally significant variables (Bonferroni corrected P < 0.10) are indicated in italics. Fagus and Picea Fagus Picea R² P R² P R² P decay class 0.5257 0.001 0.6087 0.001 0.5205 0.001 remaining mass% 0.6449 0.001 0.7669 0.001 0.5999 0.001 volume (m³) 0.1519 0.025 0.3149 0.027 0.0806 0.308 wood density (g/cm³) 0.7144 0.001 0.7536 0.001 0.6133 0.001 rel. wood moisture% 0.2959 0.001 0.4438 0.004 0.3654 0.013 C% 0.3011 0.001 0.2357 0.073 0.1944 0.126 N% 0.1352 0.043 0.1989 0.103 0.204 0.112 C/N 0.1932 0.008 0.1917 0.109 0.2071 0.09 C (g/cm³) 0.7175 0.001 0.7569 0.001 0.6174 0.001 N (g/cm³) 0.4622 0.001 0.3728 0.011 0.1601 0.163 pH 0.493 0.001 0.3439 0.015 0.1438 0.197

The opposite pattern was observed for the family Beijerinckiaceae, which explained

8.5% of the community variation and which was almost absent in the earliest stage of decay but became much more abundant as decay progressed (Fig. S4.10). This family is represented by the most abundant OTU, which was assigned to Methylovirgula. In

Picea logs, members of the Acetobacteraceae (Alphaproteobacteria) explained 20% of the total community variation across all decay classes. They were most dominant in decay class 2 logs. Acidobacteriaceae were consistently present at all decay stages and explained 10% of the community variation. They were the dominant bacterial family at all decay stages. Burkholderiaceae and Beijerinckiaceae explained 7 and 6.6% of the community variation, respectively, but there were no clear trends in their abundance comparable to those observed for Fagus logs.

134

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. 4.3: Two-dimensional non-metric multidimensional scaling (NMDS) ordination plots of bacterial community structure across the different tree species at each stage of decay (FASY1-4, PIAB1-4). Plots show centroids within a single decay stage, bars represent one SD along both NMDS axes. Statistical significances (R2 and P-values) are based on Goodness-of-fit statistics for environmental variables and bacterial order abundances per sample.

135

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

4.4 Discussion

We here present one of the first field studies on 16S rDNA bacterial community structure in deadwood of two Central European tree species using 454 pyrosequencing that provides an extensive insight into the diversity and composition of deadwood dwelling microorganisms on a rather underexplored substrate. Using a very comprehensive dataset of physico-chemical wood properties, we were able to identify key wood properties that correlate with bacterial community structure and abundances of dominant bacterial phyla (including proteobacterial classes). In line with the results of our recent study on the distribution of nifH genes (Hoppe et al. 2014), we found that members of the order Rhizobiales became significantly more abundant during the intermediate to advanced stages of decay, indicating that they may play an important ecological role and contribute significantly to N-cycling.

We did not observe any difference in bacterial richness between the two tree species but did find that richness potentially increased as decay progressed (on logs of Fagus sylvatica). It is difficult to put these findings into context due to the absence of comparable data sets. However, the steady increase in bacterial richness as decay proceeds is consistent with results on fungal species richness obtained using the molecular techniques employed in this work (Kubartova et al. 2012; Ovaskainen et al. 2013).

Our study also revealed a significant impact of forest management on bacterial OTU richness, which was significantly higher in Fagus logs in the unmanaged beech forest plots than in Fagus and Picea deadwood in the respective managed plots. This suggests that the bacterial communities in unmanaged forests are more species-rich, which may

136

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies be due to a higher level of substrate continuity arising from the absence of wood extraction (Purahong et al. 2014). When comparing Fagus logs in managed beech stands to their counterparts in managed spruce stands (and vice versa for Picea logs), we also discovered a significant impact of the surrounding stand structures on the bacterial richness. Nacke et al. 2011, who studied soil 16S rDNA diversity in grassland and forest plots using the same experimental platform also detected a significantly higher Shannon diversity index in unmanaged beech plots than managed beech and managed spruce plots. In contrast to Purahong et al. 2014 who observed a higher fungal diversity in

Fagus logs in managed beech stands than in Picea logs in managed spruce stands, we did not find such a clear pattern for bacterial OTU richness.

Alphaproteobacteria, followed by Acidobacteria and Actinobacteria were the dominant phyla in this study. The relative abundances of these phyla are similar to those observed by Nacke et al. 2011 in forest soils and in a separate study on decayed wood samples

(Valaskova et al. 2009). The relative abundance of Acidobacteria in Picea logs (20.9%) was higher than that in Fagus (14.9%). A previous study using a clone library sequencing approach (Zhang et al. 2008) similarly revealed members of the Proteobacteria to be dominant on Keteleeria evelyniana deadwood, followed by Actinobacteria and Ac- idobacteria. In contrast, Bacteroidetes was the second most abundant phylum (rather than Acidobacteria or Actinobacteria) across a range of samples studied in a different set of wood colonization experiments (Sun et al. 2013). The most abundant OTU in this study was assigned to Mucilaginibacter (Bacteroidetes), a genus that has been shown to contain degraders of pectin and xylan (Pankratov et al. 2007). This genus was also represented by the 13th most abundant OTU in our dataset. Its abundance in Fagus logs increased continuously from 14.6% to 26.7% from decay classes 1 to 4 but in Picea logs it decreased from 25.3% to 15.2% with increasing decay stage. In contrast to the find- 137

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies ings of Hervé et al. 2014, the genus Dyella of the Xanthomonadaceae only contributed marginally (0.02%) to the total sequence dataset.

Although Actinobacteria were expected to be among the dominant important early colonizers of deadwood (Clausen 1996), they contributed only 10.5% and 11.4% of the total bacterial richness in logs of Fagus and Picea, respectively. However, we found that their relative abundance decreased significantly with progressing wood decay con- firming their potential role in the early colonization and decomposition of dead wood logs. We assume that their dominant detection in culture-based studies may arise because of culturing conditions favoring them and because of fast germination of the dormant spores under conducive culturing conditions.

In contrast to soils, where pH might serve as predictor for bacterial community structure and as a determinant for relative abundances of dominant bacterial phyla (Baker et al.

2009; Lauber et al. 2009) we did not observe such an important impact of pH in deadwood. In fact, we found negative correlations for Acidobacteria (on P. abies) and Al- phaproteobacteria (on F. sylvatica), which have also been reported for soils (Nacke et al. 2011; Lauber et al. 2009), but the rather small variance of pH between different wood species and its independency from decay classes does not substantially explain the shifts of the relative abundances of the dominant phyla. Our results rather reveal that the remaining mass or the respective decay class could be used as a potential predictor for the shifts in bacterial abundances in deadwood. However, since they are determined by significant variations of the wood physico-chemical properties (e.g. C and N concentration, relative wood moisture, density; compare Tables S4.4 and S4.5), we assume that variations in the abundances of phyla are rather determined by a combination of wood properties than by single parameters, such as pH, alone. 138

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

In line with our assumptions, Rhizobiales was the dominant order in the intermediate and advanced decay classes. In fact, they accounted for almost 25% of the total sequence abundance in F. sylvatica logs of decay class 3. The relative contribution of Rhi- zobiales in the wood block experiments of Folman et al. (2008) and de Boer and van der

Wal (2008) was also around 25% even though the total number of wood-inhabiting bacteria in that study was reduced by the bactericidal effects of the white-rot fungus Hy- pholoma fasciculare. The high abundance of Rhizobiales in that case was attributed to a mutualistic/ predatory interaction since they were not detected on wood blocks without the fungus. The hypothesis that wood-degrading fungi meet their N requirements by associating with N-fixing bacteria was first raised by Cowling and Merrill (1966). This hypothesis has evolved over time, and microbial N-fixation in deadwood has since been demonstrated in several studies conducted over the last few decades (Aho et al. 1974;

Brunner & Kimmins 2003; Seidler et al. 1972; Spano et al. 1982). This hypothesis was further corroborated by the detection of methanotrophic bacteria in deadwood that utilize methane as their sole carbon source (Hanson & Hanson 1996) and possess the nifH gene encoding the enzyme dinitrogenase reductase, which is important for dinitrogen fixation in some taxa (Vorobev et al. 2011; Vorobev et al. 2009). Lenhart et al. (2012) identified eight saprotrophic fungi that produce significant amounts of methane under oxic conditions in the absence of methanogenic archaea, further supporting the possibil- ity of mutualistic interactions between fungi and methanotrophic N-fixers.

Interestingly, the most abundant OTU in our dataset was taxonomically affiliated to the genus Methylovirgula, whose type strains Vorobev et al. (2009) isolated from the beech woodblocks studied by Folman et al. (2008). These methylotrophic bacteria are specialized to utilize methanol as their sole carbon source. Partial sequences of the nifH gene 139

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies from these strains provided evidence of their N-fixing potential. We found that the abundance of Methylovirgula in Fagus logs correlated significantly and negatively with the C/N ratio (Table S4.8). Together with the preferential occurrence of this genus during the advanced stages of decay (Fig. S4.7), this finding supports the hypothesis that wood-decaying fungi interact mutualistically with certain bacterial taxa while suppress- ing or disfavoring others (Clausen 1996; de Boer & van der Wal 2008). This is further supported by the results of Brunner and Kimmins (2003), who studied deadwood from two coniferous tree species in various stages of decay and found that nitrogenase activity was highest in decay classes 3 and 4. Their data indicated that the total rate of N fixation in such logs was up to 2.1 kg N*ha-1*a-1.

Tree species was found as the main factor shaping the bacterial community structure in deadwood. This finding is comparable with the results of previous studies on bacterial community structure in forest soils beneath different tree species (Hackl et al. 2004;

Nacke et al. 2011). We also found that decay class significantly contributed to the shift in the bacterial community structure unlike the forest management regime. This is entirely consistent with our previous observations on the distribution of nifH genes within the bacterial communities in the same logs (Hoppe et al. 2014). The observed effect of the tree species mainly attributed to the wood physico-chemical properties (Table 4.2 and the corresponding attributes in Table S4.4). Among these parameters we were able to identify pH, C and N availability, and wood moisture as the main determinants corresponding to drivers of the bacterial community structure in deadwood. These findings can also be compared to the results of recent studies that verified the impact of soil

140

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies chemical properties (especially pH) on the corresponding bacterial (Hackl et al. 2004;

Lauber et al. 2009; Nacke et al. 2011) and fungal (Wubet et al. 2012) communities.

This paper describes the bacterial diversity and community structure in deadwood of two tree species. The results presented in this study elucidate the functional traits of specific bacterial taxa involved in wood degradation. We have demonstrated that the deadwood bacterial community structures are influenced mainly by the trees species and specifically by wood’s physico-chemical properties, where the most important drivers were remaining mass, density, pH, water content and C/N ratio. We also found that members of the order Rhizobiales were dominant in the studied deadwood logs, accounting for up to 25% of all bacteria present during the intermediate and late stages of decay. The most abundant OTU in our data set was assigned to the genus Methylovirgu- la, which has been shown to contain the nifH gene. This further supports the hypothesis that microbial N-fixation plays an important role in wood decomposition.

Data accessibility: The raw sequence data are available from the NCBI Sequence Read

Archive (http://www.ncbi.nlm.nih.gov/Traces/study/) under experiment SRX589509.

141

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

4.5 References

Aho PE, Seidler RJ, Evans HJ, Raju PN (1974) Distribution, enumeration, and identification of Nitrogen-fixing bacteria associated with decay in living White fir trees. Phytopathology 64, 1413-1420. Baker KL et al. (2009) Environmental and spatial characterisation of bacterial community composition in soil to inform sampling strategies. Soil Biology and. Biochemistry 41, 2292-2298. Blanchette RA, Shaw CG (1978) Associations among bacteria, yeasts, and basidiomycetes during wood decay. Phytopathology 68, 631-637. Brunner A, Kimmins JP (2003) Nitrogen fixation in coarse woody debris of Thuja plicata and Tsuga heterophylla forests on northern Vancouver Island. Canadian Journal of Forest Research 33, 1670-1682. Carlquist S (2001) Comparative wood anatomy: Systematic, ecological, and evolutionary aspects of dicotyledon wood. Springer, Berlin-Heidelberg. Chambers JQ, Higuchi N, Schimel JP, Ferreira LV, Melack JM (2000) Decomposition and carbon cycling of dead trees in tropical forests of the central Amazon. Oecologia 122, 380-388. Clausen CA (1996) Bacterial associations with decaying wood: a review. International Biodeterioration & Biodegradation 37, 101-107. Cornwell WK, Cornelissen JHC, Allison SD, et al. (2009) Plant traits and wood fates across the globe: rotted, burned, or consumed? Global Change Biology 15, 2431-2449. Cowling EB, Merrill W (1966) Nitrogen in wood and its role in wood deterioration. Canadian Journal of Botany 44, 1539-1554. Daniel G, Nilsson T (1998) Developments in the study of soft rot and bacterial decay. In: Forest products biotechnology (eds. Bruce A, Palfreyman JW), pp. 37-62. Taylor & Francis, London. de Boer W, Folman LB, Gunnewiek PJAK, et al. (2010) Mechanism of antibacterial activity of the white-rot fungus Hypholoma fasciculare colonizing wood. Canadian Journal of Microbiology 56, 380-388. de Boer W, van der Wal A (2008) Interactions between saprotrophic basidiomycetes and bacteria. In: Ecology of saprotrophic basidiomycetes (eds. Lynne Boddy JCF, Pieter van W), pp. 143-153. Academic Press. Dedysh SN, Horz H-P, Dunfield PF, Liesack W (2001) A novel pmoA lineage represented by the acidophilic methanotrophic bacterium Methylocapsa acidophila B2. Archives of Microbiology 177, 117-121. Doyle JJ, Doyle JL (1987) A rapid DNA isolation procedure for small quantities of fresh leaf tissue. Phytochemical Bulletin 19. Fengel D, Wegener G (1983) Wood: chemistry, ultrastructure, reactions. Walter de Gruyter, New York. Fischer M, Bossdorf O, Gockel S, et al. (2010) Implementing large-scale and long-term functional biodiversity research: The Biodiversity Exploratories. Basic and Applied Ecology 11, 473-485.

142

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Folman LB, Klein Gunnewiek PJA, Boddy L, De Boer W (2008) Impact of white-rot fungi on numbers and community composition of bacteria colonizing beech wood from forest soil. FEMS Microbiology Ecoloy 63, 181-191. Greaves H (1971)The bacterial factor in wood decay. Wood Science and Technology 5, 6-16. Hackl E, Zechmeister-Boltenstern S, Bodrossy L, Sessitsch A (2004) Comparison of diversities and compositions of bacterial populations inhabiting natural forest soils. Applied and Environmental Microbiology 70, 5057-5065. Hammer Ø, Harper DAT, Ryan PD (2001) PAST: Paleontological statistics software package for education and data analysis. Palaeontologia Electronica 4, 1-9. Hanson RS, Hanson TE (1996) Methanotrophic bacteria. Microbiological Reviews 60, 439-471. Harmon ME, Franklin JF, Swanson FJ, et al. (1986) Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research 15, 133-302. Herrmann S, Bauhus J (2012) Effects of moisture, temperature and decomposition stage on respirational carbon loss from coarse woody debris (CWD) of important European tree species. Scandinavian Journal of Forest Research 28, 346-357. Hervé V, Le Roux X, Uroz S, Gelhaye E, Frey-Klett P (2014) Diversity and structure of bacterial communities associated with Phanerochaete chrysosporium during wood decay. Environmental Microbiology 16, 2238-2252. Hoppe B, Kahl T, Karasch P, et al. (2014) Network analysis reveals ecological links between N-fixing bacteria and wood-decaying fungi. PLoS ONE 9. Huse SM, Dethlefsen L, Huber JA, et al. (2008) Exploring microbial diversity and taxonomy using SSU rRNA hypervariable tag sequencing. Plos Genetics 4. Kahl T, Bauhus J (2014) An index of forest management intensity based on assessment of harvested tree volume, tree species composition and dead wood origin. Nature Conservation 7, 15-27. Kahl T, Mund M, Bauhus J, Schulze ED (2012) Dissolved organic carbon from European beech logs: Patterns of input to and retention by surface soil. Ecoscience 19, 1-10. Kirker GT, Prewitt ML, Diehl WJ, Diehl SV (2012) Community analysis of preservative-treated southern pine (Pinus spp.) using terminal restriction fragment length polymorphism (T-RFLP) analysis. Part 2: Bacteria field study. Holzforschung 66, 529-535. Kubartova A, Ottosson E, Dahlberg A, Stenlid J (2012) Patterns of fungal communities among and within decaying logs, revealed by 454 sequencing. Molecular Ecology 21, 4514-4532. Lane DJ (1991) 16S/23S rRNA sequencing. In: Nucleic acid techniques in bacterial systematics (eds. Stackebrandt E, Goodfellow M), pp. 125-175. Wiley, New York, USA. Lassauce A, Paillet Y, Jactel H, Bouget C (2011) Deadwood as a surrogate for forest biodiversity: Meta-analysis of correlations between deadwood volume and species richness of saproxylic organisms. Ecological Indicators 11, 1027-1039. Lauber CL, Hamady M, Knight R, Fierer N (2009) Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Applied and Environmental Microbiology 75, 5111-5120. Lenhart K, Bunge M, Ratering S, et al. (2012) Evidence for methane production by saprotrophic fungi. Nature Communincations 3.

143

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Li W, Godzik A (2006) Cd-hit: a fast program for clustering and comparing large sets of protein or nucleotide sequences. Bioinformatics 22, 1658-1659. Lindahl BD, Finlay RD (2006) Activities of chitinolytic enzymes during primary and secondary colonization of wood by basidiomycetous fungi. New Phytologist 169, 389-397. Litton CM, Raich JW, Ryan MG (2007) Carbon allocation in forest ecosystems. Global Change Biology 13, 2089-2109. Lynd LR, Weimer PJ, van Zyl WH, Pretorius IS (2002) Microbial cellulose utilization: Fundamentals and biotechnology. Microbiology and Molecular Biology Reviews 66, 506-+. Merrill W, Cowling EB (1966) Role of nitrogen in wood deterioration - Amount and distribution of nitrogen in fungi. Phytopathology 56, 1083-1090. Müller J, Engel H, Blaschke M (2007) Assemblages of wood-inhabiting fungi related to silvicultural management intensity in beech forests in southern Germany. European Journal of Forest Research 126, 513-527. Muyzer G, Dewaal EC, Uitterlinden AG (1993) Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59, 695-700. Nacke H, Thurmer A, Wollherr A, et al. (2011) Pyrosequencing-based assessment of bacterial community structure along different management types in German forest and grassland soils. PLoS ONE 6. Oksanen J (2013) Multivariate analysis of ecological communities in R: vegan tutorial. (accessed: August 20th 2014). Ovaskainen O, Schigel D, Ali-Kovero H, et al. (2013) Combining high-throughput sequencing with fruit body surveys reveals contrasting life-history strategies in fungi. ISME Journal 7, 1696-1709. Pankratov TA, Tindall BJ, Liesack W, Dedysh SN (2007) Mucilaginibacter paludis gen. nov., sp. nov. and Mucilaginibacter gracilis sp. nov., pectin-, xylan- and laminarin-degrading members of the family Sphingobacteriaceae from acidic Sphagnum peat bog. International Journal of Systematic and Evolutionary Microbiology 57, 2349-2354. Peres-Neto PR, Jackson DA (2001) How well do multivariate data sets match? The advantages of a Procrustean superimposition approach over the Mantel test. Oecologia 129, 169-178. Purahong W, Hoppe B, Kahl T, et al. (2014) Changes within a single land-use category alter microbial diversity and community structure: Molecular evidence from wood-inhabiting fungi in forest ecosystems. Journal of Environmental Management 139, 109–119. Quast C, Pruesse E, Yilmaz P, et al. (2013) The SILVA ribosomal RNA gene database project: improved data processing and web-based tools. Nucleic Acids Research 41, 590-596. Rajala T, Peltoniemi M, Pennanen T, Makipaa R (2012) Fungal community dynamics in relation to substrate quality of decaying Norway spruce (Picea abies [L.] Karst.) logs in boreal forests. FEMS Microbiology Ecology 81, 494-505. Schall P, Ammer C (2013) How to quantify forest management intensity in Central European forests. European Journal of Forest Research 132, 379-396.

144

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Schloss PD, Westcott SL, Ryabin T, et al. (2009) Introducing mothur: Open-source, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75, 7537-7541. Seidler RJ, Aho PE, Evans HJ, Raju PN (1972) Nitrogen fixation by bacterial isolates from decay in living White fir trees [Abies Concolor (Gord. and Glend.) Lindl.]. Journal of General Microbiology 73, 413-416. Spano SD, Jurgensen MF, Larsen MJ, Harvey AE (1982) Nitrogen-fixing bacteria in Douglas-fir residue decayed by Fomitopsis pinicola. Plant and Soil 68, 117-123. Stokland JN, Siitonen J, Jonsson BG (2012) Biodiversity in dead wood Cambridge University Press, Cambridge. Sun H, Terhonen E, Kasanen R, Asiegbu FO (2013) Diversity and community structure of primary wood-inhabiting bacteria in boreal forest. Geomicrobiology Journal 31, 315-324. Tenenbaum DJ (2004) Underwater logging: Submarine rediscovers lost wood. Environmental Health Perspective 112, 892-895. Valaskova V, de Boer W, Klein Gunnewiek PJA, Pospisek M, Baldrian P (2009) Phylogenetic composition and properties of bacteria coexisting with the fungus Hypholoma fasciculare in decaying wood. ISME Journal 3, 1218-1221. Volkenant K (2007) Totholz als Lebensraum von Mycozönosen im fortschreitenden Zersetzungsprozess - Eine Chronosequenzstudie an Fagus sylvatica-Totholz im Nationalpark Kellerwald-Edersee, Universität Kassel. Vorobev AV, Baani M, Doronina NV, et al. (2011) Methyloferula stellata gen. nov., sp. nov., an acidophilic, obligately methanotrophic bacterium that possesses only a soluble methane monooxygenase. International Journal of Systematic and Evolutionary Microbiology 61, 2456-2463. Vorobev AV, de Boer W, Folman LB, et al. (2009) Methylovirgula ligni gen. nov., sp nov., an obligately acidophilic, facultatively methylotrophic bacterium with a highly divergent mxaF gene. International Journal of Systematic and Evolutionary Microbiology 59, 2538-2545. Wubet T, Christ S, Schoning I, et al. (2012) Differences in soil fungal communities between European beech (Fagus sylvatica L.) dominated forests are related to soil and understory vegetation. PLoS ONE 7. Zhang HB, Yang MX, Tu R (2008) Unexpectedly high bacterial diversity in decaying wood of a conifer as revealed by a molecular method. International Biodeterioration & Biodegradation 62, 471-474.

145

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies 4.6 Supplemental information

Sampling

In April 2009, the initial experiemtal set-up comprised 48 logs (24 of P. abies and 24 of F. sylvatica) located on the forest floor of nine very intensively investigated 1 ha forest- plots (VIPs) within the “Schwäbische Alb – exploratory”, with three plots representing three different forest types and management regimes, respectively: (i) unmanaged beech forests, where timber harvesting stopped 2 - 7 decades ago, (ii) managed beech forests dominated by Fagus sylvatica and (iii) managed spruce forests dominated by Picea abies, which in both cases are characterized by uniform tree species composition, forest structure and site conditions, were randomly selected and characterized (Fig. S4.4, Ta- ble S4.3). The logs were selected to ensure that some of the Fagus logs were located in Picea-dominated plots and vice versa. In June 2009 3-7 wood samples were collected from each log; the number of wood samples determined according to the formula (with x = the rounded number of drills per log, and y = length of deadwood log). We therefore sampled a minimum of 3 wood samples per deadwood log up to a length of 5m, (additional wood samples were taken if the log was longer) to fully represent the entire log. (compare Pu- rahong et al. 2014). To take the samples, we used a cordless Makita BDF451 drill (Makita, Anja, Japan) equipped with a 2 x 42 cm wood auger as described by Hoppe et al. 2014. The upper surface layer and bark of the deadwood was removed to avoid contamination by bacteria from outside. Each deadwood log was assigned to one of 4 decay classes based on its remaining mass (%) which was determined upon mass and density loss by k-means cluster analysis as described by Kahl et al. 2012. Higher decay classes corresponded to more extensive decay. The point of death of deadwood logs at the date of sampling varied between 3 and 27 years (based on dendro-chronological analyses) and was due to various reasons, ranging from windbreaks to logged and left trees.

146

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.4: Sampling scheme visualized using Treemap v. 3.1.0. (Macrofocus, Zurich, Switzerland) in squarified layout. Items are grouped by management type and corresponding forest plot. Treemap cell size is proportional to remaining mass in% (smaller cells = less decayed logs) Colors represent tree species. (red = Picea abies, green = Fagus sylvatica). Numbers per cell indicate the Tree_ID, Plot_ID, length of deadwood log in (m) and the number of wood subsamples taken per log.

Table S4.3: Sampling design: Distribution of deadwood logs according to tree species and the respective 9 forest plots which were assigned to three different forest management types. Management type

Unmanaged beech managed beech fo- managed spruce fo-

forests (3 plots) rests (3 plots) rests (3 plots)

Fagus sylvatica 8 8 8

Picea abies 8 6 10 Deadwood tree species

147

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Table S4.4: Mean values of wood properties and standard error for each tree species related decay class and ANOVA P values. Differences in C/N, concentrations per wood density and relative wood moisture between decay classes were analyzed by employing one-way analysis of variance and Tukey pair-wise comparisons. Significant ANOVA P values are shown in bold (P < 0.05). Different letters indicate differences among decay classes. (P < 0.05). Density rel. wood C/N C (g/cm³) N (g/cm³) pH (g/cm³) moisture 1 (7 logs) 376.9 ± 11.6 cd 0.25 ± 0.01 a 0.0007 ± 0.00004 a 0.51 ± 0.02 a 52.2 ± 5.8 b 5.3 ± 0.1 a 2 (6 logs) 324 ± 24.3 cd 0.18 ± 0.01 b 0.0006 ± 0.00006 a 0.37 ± 0.01 b 114.2 ± 15.8 ab 4.8 ± 0.1 ab 3 (7 logs) 282.8 ± 29.2 cd 0.15 ± 0 bc 0.0005 ± 0.00004 a 0.31 ± 0.01 bc 138.2 ± 21.2 ab 5.1 ± 0.1 a

Fagus sylvatica 4 (3 logs) 193.7 ± 15.6 d 0.09 ± 0.02 de 0.0005 ± 0.00009 ab 0.2 ± 0.03 de 155.2 ± 9.1 ab 5 ± 0.1 ab 1 (5 logs) 629.7 ± 48.4 a 0.17 ± 0 bc 0.0003 ± 0.00003 bc 0.35 ± 0.01 b 48.7 ± 11.6 b 4.6 ± 0.1 abc 2 (5 logs) 602.6 ± 19.2 ab 0.13 ± 0 cd 0.0002 ± 0.00001 bc 0.27 ± 0.01 cd 65.5 ± 13.4 b 4.3 ± 0.03 bc 3

Picea abies (5 logs) 591.2 ± 38.8 ab 0.11 ± 0.01 de 0.0002 ± 0.00001 c 0.22 ± 0.01 de 95.3 ± 8.9 ab 4.04 ± 0.1 bc 4 (9 logs) 422.8 ± 52.4 bc 0.08 ± 0 e 0.0002 ± 0.00003 c 0.15 ± 0.01 e 163.1 ± 24.6 a 4.3 ± 0.2 c

P <0.001 <0.001 <0.001 <0.001 <0.003 <0.001

Table S4.5: Correlations among assessed wood physico-chemical properties. Significant values (P < 0.05) are given in bold. rel. remaining wood C N mass in% volume density moisture C% N% C/N (g/cm³) (g/cm³) pH remaining mass in% volume -0.16 density 0.89 -0.12 rel. wood moisture% -0.67 0.03 -0.53 C% -0.23 0.11 -0.39 0.25 N% -0.38 -0.05 -0.11 0.6 -0.32 C/N 0.26 0.01 -0.09 -0.53 0.42 -0.92 C (g/cm³) 0.9 -0.13 -0.99 -0.53 -0.34 -0.14 -0.05 N (g/cm³) 0.42 -0.09 0.73 0.02 -0.52 0.56 -0.69 0.7 pH 0.32 0.05 0.55 -0.03 -0.5 0.45 -0.6 0.53 0.71

148

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Table S4.6: Mean relative abundances of dominant bacterial phyla (including proteobacterial subphyla) at different decay stages. Differences of bacterial abundances between decay classes and tree species were analyzed by employing one-way analysis of variance and Fisher’s Least Significant Difference (LSD) post hoc test.

FASY 1 FASY 2 FASY 3 FASY 4 PIAB 1 PIAB 2 PIAB 3 PIAB 4 P

Alphaproteobacteria 28.1 ± 3 c 38.5 ± 1.8 abc 43.5 ± 2.3 ab 50.7 ± 1.9 a 30.4 ± 3.7 bc 45.4 ± 2 ab 42.2 ± 3 ab 44.1 ± 1.5 ab <0.001

Acidobacteria 13.3 ± 3.4 a 16.5 ± 3.7 a 14 ± 3.7 a 15.1 ± 3.2 a 19.5 ± 0.7 a 20.4 ± 0.9 a 22.4 ± 1.9 a 20.7 ± 1.1 a 0.19

Actinobacteria 13.2 ± 1.7 a 11.5 ± 0.7 a 10 ± 1.1 a 6.4 ± 0.5 a 14.2 ± 1.8 a 10.8 ± 1 a 11.9 ± 1.3 a 10.2 ± 0.9 a 0.08

Gammaproteobacteria 12.7 ± 1.5 a 9.7 ± 2.2 ab 8.1 ± 0.4 ab 5.7 ± 1.8 ab 10.8 ± 3.8 ab 5.6 ± 1 ab 5.3 ± 1.1 b 6.8 ± 0.7 ab 0.047

Bacteroidetes 9.1 ± 2.5 a 5.9 ± 0.6 a 10.4 ± 1.7 a 8.7 ± 0.7 a 8.3 ± 1.5 a 5.9 ± 0.3 a 5.8 ± 0.6 a 5.5 ± 0.6 a 0.149

Betaproteobacteria 11.8 ± 2.8 a 7 ± 2.3 a 4.7 ± 0.8 a 7 ± 0.8 a 7.2 ± 1.1 a 7.6 ± 0.9 a 6.8 ± 1.6 a 5.2 ± 0.6 a 0.17

Firmicutes 9.6 ± 3.2 a 6.5 ± 1.6 ab 4.5 ± 1.6 ab 0.6 ± 0.2 b 5.2 ± 0.8 ab 1 ± 0.2 ab 0.9 ± 0.3 b 0.4 ± 0.2 b 0.003

Cyanobacteria 1.3 ± 0.3 a 2.2 ± 0.6 a 1.5 ± 0.3 a 3.1 ± 0.8 a 2.6 ± 1.1 a 1.3 ± 0.2 a 2 ± 0.1 a 2.5 ± 0.2 a 0.23

Deltaproteobacteria 0.6 ± 0.2 b 1.3 ± 0.5 b 1.9 ± 0.7 ab 1.7 ± 1 ab 1.2 ± 0.6 b 1.2 ± 0.6 b 1.6 ± 0.7 ab 2.6 ± 0.9 a 0.001

Table S4.7: Pearson rank correlation table of selected dominant bacterial phyla (including proteobacterial subphyla) with wood physico-chemical properties. Significant correlations (P < 0.05), are displayed in bold. Calculations were performed separately for the respective tree species. Fagus Picea Fagus Picea Fagus Picea Fagus Picea C/N remaining mass% rel. wood moisture% pH

Alphaproteobacteria -0.44 -0.03 -0.75 -0.49 0.52 0.24 0.05 -0.4 Acidobacteria 0.13 -0.02 -0.12 -0.1 0.2 0.07 -0.46 -0.13 Actinobacteria 0.27 0.31 0.54 0.4 -0.42 -0.43 0.26 0.33 Gammaproteobacteria -0.03 -0.21 0.52 0.25 -0.33 0.15 0.21 0.39 Bacteroidetes -0.05 0.02 0.02 0.43 -0.25 -0.37 0.28 0.18 Betaproteobacteria 0.12 0.19 0.39 0.27 -0.14 -0.45 0.06 -0.14 Firmicutes 0.4 0.3 0.45 0.7 -0.43 -0.35 0.04 0.41 Cyanobacteria -0.14 -0.06 -0.38 -0.18 0.23 0.16 -0.27 -0.18 Deltaproteobacteria -0.41 -0.53 -0.64 -0.56 0.52 0.65 0.04 0.21

Table S4.8: Pearson Rank correlation table of selected methylo- and methanotrophic bacteria with wood physico-chemical properties. Significant correlations (P < 0.05) are displayed in bold. Fagus Picea Fagus Picea Fagus Picea Fagus Picea Genus C/N remaining mass% rel. wood moisture% pH Methylovirgula -0.38 -0.17 -0.7 -0.24 0.69 -0.28 -0.23 -0.04 Methyloferula -0.29 0.1 -0.31 -0.02 0.56 -0.01 -0.3 -0.06 Methylocella 0.18 -0.32 -0.02 -0.08 -0.23 -0.07 0.26 -0.1 Methylocystis -0.02 -0.22 -0.06 -0.03 0.24 -0.33 -0.22 -0.29

149

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Table S4.9: One-way analysis of similarity (ANOSIM) based on each two distance measures using relative abundance and presence/ absence data for different wood-inhabiting bacterial communities on two different tree species (Fagus sylvativca and Picea abies). Bray-Curtis Euclidean Tree species R P R P Rel. Abundance 0.3098 0.001 0.327 0.001 Fagus sylvatica vs. Jaccard Sörenson Picea abies R P R P Presence/ absence 0.3141 0.001 0.3141 0.001

150

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.5: Bacterial richness of the two deadwood species Fagus sylvatica and Picea abies distinguished upon the 4 different decay classes (e.g. FASY1 = Fagus sylvatica, decay class 1) at a genetic cutoff of 3%. Richness is displayed as the mean number of observed sequences per tree decay classes. Differences of bacterial species richness between decay classes and tree species were analyzed by employing one-way analysis of variance and Fisher’s Least Significant Difference (LSD) post hoc test.

Fig. S4.6: Correlation between bacterial OTU richness and remaining mass in%. Green triangles represent data on FASY (Fagus sylvatica) and red diamonds on PIAB (Picea abies).

151

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.7: Relative abundances of four selected bacterial OTUs (methano- and methylotrophic genera) per wood tree species (Fagus sylvatica = FASY, Picea abies = PIAB) and connected decay classes (1-4).

Fig. S4.8: Mean bacterial OTU richness according to differently managed forest types. (A) Fagus sylvatica logs in each of the three differently managed forest plots and (B) vice versa of Picea abies logs. Fish- er’s Least Significant Difference (LSD) post hoc were calculated separately for (A) and (B).

152

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.9: Correlation between bacterial OTU richness and the two land use intensity indices ForMI (light grey squares) and SMI (dark grey diamonds).

Fig. S4.10: Relative abundances and their contribution to total community variation of bacterial families which represent > 0.75 of the sequences of the whole dataset as calculated by SIMPER. Both, Fagus (FASY) and Picea (PIAB) were separated in four wood decay classes. Heatmaps (white = zero, red = maximum abundance) were overlaid separately for Fagus and Picea.

153

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.11: Kronafile SK1: Interactive web-visualization of taxonomic information on all deadwood logs using Krona (http://sourceforge.net/projects/krona/).

Fig. S4.12: Kronafile SK2: Interactive web-visualization of taxonomic information on Fagus sylvatica deadwood logs using Krona (http://sourceforge.net/projects/krona/).

154

4 A pyrosequencing insight into sprawling bacterial diversity and community dynamics in decaying deadwood logs of Fagus sylvatica and Picea abies

Fig. S4.13: Kronafile SK3: Interactive web-visualization of taxonomic information on Picea abies deadwood logs using Krona (http://sourceforge.net/projects/krona/).

155

5 Synthesis and discussion

5.1 General rationale and achievement

The presented thesis provides novel information on biological and in parts functional diversity of deadwood-inhabiting microbiota.

By using a high throughput sequencing technique, together with a very comprehensive dataset on wood physico-chemical properties, which was provided by cooperating working groups, I was able to disentangle some parameters that correlate with taxonomically identified microorganisms which provided insights into some functional mechanisms.

The initial goal of this thesis was to observe and describe the impact of different forest management intensities and practices on diversity and community structure of deadwood-inhabiting fungi. The study was conducted under the overarching experimental platform of the German Biodiversity Exploratories, which provided comprehensive and descriptive information on the forest management type of the respective study plot. To capture most information on fungal diversity, we performed a parallel methodical approach by using molecular tools on the one hand, and also intensive monitoring of sporocarps (at three different seasonal occasions in 2009 and 2010) on the other hand. At the beginning ARISA (Boddy & Watkinson 1995; Fisher & Triplett 1999) was the molecular method chosen, to observe shifts in microbial community structure (abundance of the respective community members) and composition (presence/ absence of the community members) on the investigated deadwood logs without any further identification of the observed organisms. However I realized that some scientific questions that

156

5 Synthesis and discussion should have been easily answered with the experimental set-up could not be sufficiently resolved by the application of ARISA. Due to rapid advancements of techniques and the parallel installation of a 454 pyrosequencing platform (Margulies et al. 2005) at the De- partment of Soil Ecology, I was able to change methodology and thus gain taxonomically resolved insights on bacteria and fungi, so that besides an analysis of the community richness (number of OTUs), the community composition (identity of the OTUs) became accessible. In addition, the depth of 454 pyrosequencing and the amount of samples that could be analyzed also enabled me to precise the community structure and a wealth of functional ecological patterns such as co-occurrence. Results of these different approaches are presented in chapters 2 and 4. The third chapter however is based on a traditional Sanger sequencing approach, which does not enable the analyzing depth of 454, but provides longer fragments, by a reduced sequencing error rate, which is useful to insure correct identification of OTUs. At the end, combining the three approaches: (1) monitoring of sporocarps, (2) clone library sequencing and (3) 454 pyrosequencing, appeared to be optimal for the kind of study made in this thesis.

5.2 The potential and limits of Next-generation sequencing as the

method of choice

ARISA was developed as an automated molecular technique to accurately and cost- efficiently monitor and detect changes in diversity and community structures of microorganisms (Fisher & Triplett 1999; Popa et al. 2009). It identifies so-called phylotypes

(OTU is also an appropriate terminology) upon length polymorphisms of the intergenic or intragenic transcribed spacer of bacterial or fungal rRNA, respectively. It has been routinely applied in various studies observing temporal and spatial variation of microbi-

157

5 Synthesis and discussion al community structures (Carson et al. 2010; Ranjard et al. 2001; Torzilli et al. 2006).

However in contrast to these obvious benefits of a technique which is easy applicable and cost-efficient, ARISA suffers from crucial methodological discrepancies. It has been shown, that an operational taxonomic unit (OTU) in an ARISA community profile may represent more than one taxon (Gillevet et al. 2009). As a consequence of this, it is not possible to treat or handle an OTU as one representative taxon, since it technically may represent multiple taxa. However, for studying biodiversity and related ecosystem processes, it is necessary to have good predictions or estimations of species richness or richness indices. A second crucial methodological drawback is based on the fact that

ARISA fingerprint profiles do not allow for any taxonomical assignment, which could also be important according to the scientific question/ hypothesis addressed. In the case of this thesis, it occurred to be essential to gain taxonomically resolved information on the captured organisms, in order to correlate this information with microbial-mediated ecosystem functions (lignin-modifying enzyme activity) and processes (wood decomposition), as demonstrated in chapters 2 and 4. The application of solely a DNA-based fingerprint method could have not provided details on the identity of the key species of fungal and bacterial communities that explain the variation in community structure on the two different deadwood substrates/ species. ARISA may only be used to provide a quick snapshot of diversity or to observe temporal or spatial variation of community structure which can be further refined using sequencing methods (Gillevet et al. 2009;

Weig et al. 2013).

Though many studies investigating deadwood fungal diversity have conducted intensive monitoring of sporocarps (Bässler et al. 2014; Bässler et al. 2010; Blaser et al. 2013;

Heilmann-Clausen & Christensen 2003), these approaches strongly rely on the expertise of individual mycologist. Microscopically and macroscopically identifications can be 158

5 Synthesis and discussion quit subjective and may be restricted to knowledge of species and technical equipment

(literature, identification keys, microscope, binocular) (Schmidt & Moreth 2006). This problem is further aggravated by the fact that, owing to the reduction in taxonomical training at universities, the number of people, who can conduct such scientific inventories, is restricted to few and most likely amateur mycologists. Due to the seasonal fructification potentials of different fungal species/ groups, it has also been shown that one single fruiting body inventory survey is not sufficient (Halme & Kotiaho 2012). How- ever, as presented in chapter 3, I intended to correlate the presence and findings of diverse nifH genes in deadwood with sporocarps data, since sporocarps present the actively growing fungi and the availability and quality of N determines fruiting body production (Moore et al. 2008). In fact, I also observed significant and positive correlations between molecular fungal OTU richness and nifH OTU richness, but the application of sophisticated network analyses revealed certain fungal sporocarps only appeared or were present in correlation to the occurrence of certain nifH genes.

I further performed, together with Dr. Tiemo Kahl, an initial approach to compare the outcomes of a DNA-based fingerprint (ARISA) (which present the “hidden”, not visual- ly observable diversity form inside the wood logs) with sporocarp data, to assess fungal diversity and community structure. For this preliminary analysis, we considered tree species, decay classes, volume of the deadwood logs and the Biodiversity Exploratory as factors, correlating to either community structure or fungal species richness (as determined by sporocarps or ARISA-OTUs). We observed comparable results for the fungal community structure using both datasets (Table 5.1). In both cases deadwood tree species and decay classes significantly explained most variation of the fungal community structure on the investigated logs. In addition also the results for exploratory and deadwood volume significantly explained parts of the variation. 159

5 Synthesis and discussion

I further observed very different correlations between these above mentioned environmental parameters and fungal richness (OTU richness and sporocarps richness) (Fig.

5.1). Highest species richness was observed in logs of the “Schwäbische Alb” - exploratory (Alb) when applying ARISA, whereas sporocarp sampling resulted in highest species richness on logs in the “Hainich” - exploratory (Hai) (Fig. 5.1a).

Table 5.1: Results of a simplified perMANOVA (interactions are excluded) analysis of Bray-Curtis dissimilarities in ARISA OTU (left) and sporocarp community structure in relation to tree species, exploratory decay class and and volume of the deadwood log, Df = degrees of freedom; SS = sum of squares; MS = mean sum of squares; Pseudo-F = F value by permutation, boldface indicates statistical significance at P < 0.05, P values based on 999 permutations (lowest P-value possible is 0.001).

ARISA-OTUs F R² P Sporocarps F R² P

Tree species 6.8808 0.06267 0.001 Tree species 13.8662 0.112 0.001

Exploratory 2.1003 0.01913 0.001 Exploratory 5.1322 0.04145 0.001

Decay class 6.4238 0.02926 0.001 Decay class 7.7143 0.03115 0.001

Volume 2.6609 0.01212 0.001 Volume 4.6244 0.01868 0.001

Residuals 0.77422 Residuals 0.68653

Total 1 Total 1

160

5 Synthesis and discussion

Fig. 5.1: Comparison of correlations between sporocarps richness (fungi)/ ARISA-based OTU richness (arisa) as the chosen method (meth) and environmental data: (upper left – a) exploratory (exp = Alb – Schwäbische Alb, Hai - Hainich, Sch – Schorfheide), (upper right – b) tree species (Fasy – Fagus sylvatica, Pcab – Picea abies, Pisy – Pinus sylvestris, (lower left – c) remaining mass (mass.perc), and (lower right – d) volume of the log (v.log).

Another contrasting result was detected, when focusing on species richness per tree species. Sporocarp sampling revealed highest species richness on Fagus sylvatica (Fasy) logs, while ARISA revealed more species in logs of the coniferous species Picea abies

(Pcab) and Pinus sylvestris (Pisy) (Fig. 5.1b). These preliminary comparisons indicate that the results are highly dependent on the methodological approach to the scientific questions that were addressed.

The rapid development of molecular techniques has reached the level of so-called

“Next-generation sequencing”. This metagenomic revolution in molecular ecology means unprecedented amounts of nucleic acid sequence data that are accumulated and allow researchers to test many hypotheses on the distribution and functions of microor-

161

5 Synthesis and discussion ganisms in many environments. The 454-platform (Rothberg & Leamon 2008) that I used to generate the data presented in chapters 2 and 4 was launched in 2004 and is based on pyrosequencing in microreactors on a picotiter plate (Margulies et al. 2005).

By applying this technique to a preselected set of samples from deadwood logs, I was able to gain encompassing taxonomically resolved identities of fungi and bacteria. For both datasets I obtained a comparable amount of rarified sequence reads per sample

(1837 for fungi and 1555 for bacteria), which is in line to recent literature (Christ et al.

2011; Nacke et al. 2011).

Even though pyrosequencing on the ITS barcode marker approach avoids the cloning biases (Liang et al. 2011) and allows for the simultaneous multiplexed analysis of large numbers of samples, which dramatically advanced molecular environmental studies, it has some considerable difficulties to handle this huge amount of sequence data. A large proportion of potentially undescribed fungi are being discovered at an astounding rate based on sequence data yet lacking any confirmation of fruiting bodies (Hibbett &

Taylor 2013), even leading to the suggestion of one author to abolish formal naming of fungi by Linnean-based nomenclature codes (Money 2013). This may not necessarily be the case for bacteria, due to the fact that a tremendous proportion of them can be culti- vated and morphologically described anyways. Moreover, the fungal ITS locus has many caveats such as copy number variability (Ganley & Kobayashi 2007; Herrera et al. 2009), copy-to-copy intranuclear sequence heterogeneity (Lindner et al. 2013) and a lack of phylogenetic resolution (Krüger et al. 2012) especially when moving outside closely related taxa as normally is done in metagenomic diversity studies. In general, the dereplication of environmental sequence data and building of operational taxonomic units (OTUs or phylotypes) follows methods involving multiple-sequence alignment or only pairwise comparison especially the first of these being negatively impacted by ITS 162

5 Synthesis and discussion heterogeneity. This resulted for example in the detection of 68 different OTUs that were taxonomically assigned to one species Resinicium bicolor. In this study I solved this problem by individually re-blasting all representative sequences per OTU against Gen-

Bank and assigning all OTUs to a so called cOTU (cumulative OTU) by fixing an ITS homology threshold of 97% to 100% maximum identity in order to define a “taxon”.

Regardless of some technical obstacles, Next-generation sequencing analyses can potentially detect most fungal species present in wood and, if linked to taxonomically resolved identities, this provides a tremendous advancement in our understanding of spatiotemporal fungal community patterns but also provides fundamental insights into fungal community ecology and fungal-mediated processes (Lindahl et al. 2013).

For further investigations, I suggest to consider the applications of precursor rRNA- based investigations to combine advantages of molecular techniques and sporocarps monitoring, by only detecting the active fungal community in decomposing substrates.

Though only a few studies are published to date (Baldrian et al. 2012; Rajala et al.

2011), the using this molecular marker could substantially improve our understanding of the relationships between microorganisms and microbial-mediated ecosystem processes.

5.3 Impact of land use intensity on deadwood volume and species

richness

I observed conditionally comparable patterns of fungal and bacterial richness (as determined by OTU richness and the richness estimators Chao1 and/or ACE) in accordance with the forest management type. I therefore extracted subsets of the whole data, e.g. only considering Fagus sylvatica logs in managed and extensively managed beech

163

5 Synthesis and discussion stands in comparison to Picea abies logs in managed spruce stands. This filtering occurred to be necessary, since an impact of the surrounding forest type on the microbial community composition of the respective deadwood item was expected (Purahong et al.

2014). For the fungal OTU richness based on Chao1 and ACE Picea deadwood appeared to harbor was significantly lower fungal diversity than Fagus deadwood

(P = 0.01-0.02, Fig. 2.10), which is consistent with a study conducted in the same area that compared the fungal OTU richness in Picea deadwood in converted forests (age- class spruce forest converted from beech forest) to that of Fagus deadwood in a semi- natural age-class beech forest (Purahong et al. 2014). Bacterial OTU richness did not vary significantly between these two management types, but was highest in Fagus logs in extensively managed beech stands. Especially for the initial phases of wood decay I observed a significantly higher bacterial OTU richness on Fagus logs in extensively managed beech plots compared to Fagus logs in managed beech stands and Picea logs in managed spruce stands. As expected due to prior studies on fungal diversity in deadwood under the same overarching experimental design of the Biodiversity Exploratories

(Blaser et al. 2013; Purahong et al. 2014), I also observed a significant impact of the management type on the fungal community structure, as revealed by the Goodness-of- fit statistics. This is likely due to the fact that fungi are not only strongly dependent on certain substrates (Kebli et al. 2011; Volkenant 2007), but also are known to act and serve as “indicators of nature value” (Blaschke et al. 2009). This may be also depending to the fact that deadwood in natural and extensively managed is continuously available and therefore could allow for respective adaption mechanisms of certain fungi that rely on continuity and are rather inflexible to populate new substrates.

164

5 Synthesis and discussion

In any case, the results concerning management type and their interpretation towards impact of land use intensity on biodiversity have to be viewed with caution. The fact that Blaser et al. (2013), in contrast to our study and that of others (Lindblad et al. 1998;

Müller et al. 2007), did not observe higher fungal species richness in deadwood logs in unmanaged beech forests emphasizes the necessity to set-up an appropriate experimental design. In contrast to the design presented in this thesis (directed selection of deadwood logs on the whole 100m x 100m plot), Blaser et al. (2013) sampled all deadwood items on a 20m x 20m square on the same plots. This contrarily resulted in sampled deadwood logs with big volumes in the extensively managed plots of my studies, while Blaser et al. (2013) sampled on larger deadwood items (probably by chance) in the managed plots. Comparing the results of these two investigations simply displays the important impact of deadwood volume of the logs on species richness, when sampling sporocarps, which has been shown in several studies (Bader et al. 1995; Lindblad

1998; Ódor et al. 2006) and is considered as the species-area/ species-diameter relationship (Sippola et al. 1998; Stokland et al. 2012). Heilmann-Clausen and Christensen

(2004), who studied the impact of volume or different deadwood fractions/ items on fungal diversity discovered a positive correlation between deadwood diameter and fungal species richness (sporocarps), when single deadwood samples/ items were compared. Rarefaction curves, based on volume, however, displayed higher species richness per unit of log volume on small trees and rarefaction curves based on log surface-based curves revealed that species density (number of species per area) slightly decreased with tree size. This “surface area factor”, as Heilmann-Clausen and Christensen (2004) described it, explains and emphasizes the fact that small diameter wood has a larger surface area per volume and therefore can also harbor more species per unit of log volume

(when calculating rarefaction curves). Interestingly I did not per se observe such an in- 165

5 Synthesis and discussion crease of fungal species/ OTU- richness on larger deadwood logs, which is in line with other studies applying molecular techniques (Kebli et al. 2011; Rajala et al. 2012).

Controlled deadwood experiments like BELongDead4 or LOGLIFE (Cornelissen et al.

2012) provide a standardized set-up, by observing deadwood items of comparable sizes.

Especially the BELongDead experiment was installed to study the impact of different forest management intensities and the surrounding habitat on wood decomposition.

A further point of criticism regarding the impact of forest management on microbial diversity in deadwood (as shown in the results of this thesis and the two comparable studies of Blaser et al. (2013) and Purahong et al. (2014)), concerns the fact that forest management type was only classified into three categories, even though stand structure

(number of trees per ha, stand age, “stocking degree” etc.) significantly varied within each category. The initial design of the Biodiversity Exploratories did not consider forest management intensity in different forest plots as quantifiable continuous variable.

Over the course of years three different indices to assess forest management intensity were developed for the Biodiversity Exploratories research platform. The LUDI (land use and disturbance intensity index) describes the “relationship between stand density and diameter at breast height for a relatively unmanaged and pristine baseline forest and different management schemes, in conjunction with the self-thinning relationship, to calculate the difference between potential and actual biomass storage” (Luyssaert et al.

2011). The SMI index (silvicultural management intensity indicator) developed by

Schall & Ammer (2013) considers tree density, stand age, total woody biomass and a tree species component that is related to the risk of management with a certain tree spe-

4 The BELongDead experiment was conceived in 2008 under the direction of Prof. E. D. Schulze (MPI Biogeochem- ie Jena) with the intention to monitor the impact of habitats on the deadwood decomposition processes and to observe colonization patterns of various saproxylic organisms over long time. He and colleagues therefore put 12-13 different tree species (three replicates = subplots) in all three exploratories (1140 logs in total). Its central goals are i) how does the surrounding ecosystem influence wood decomposition, ii) how does succession of the substrates proceed and iii) how do microorganisms mediate wood decay and subsequently impact ecosystem processes like element turnover and storage. 166

5 Synthesis and discussion cies. The ForMI (forest management intensity Index) developed by Kahl & Bauhus

(2014) is based on information of harvested tree volume (in relation to the volume that would be present without harvesting), the proportion of species, that are not naturally occurring in that forest ecosystem (potentially natural vegetation) and the proportion of deadwood (logs, crown parts and stumps) showing signs of saw cuts. Both ForMI and

SMI have been tested on a broad set of forest plots of the “exploratories” and have demonstrated to significantly correlate with each other. I initially tested for correlations between the two indices ForMI and SMI with bacterial OTU richness and observed a significant impact of land use intensity on deadwood logs of Fagus sylvatica (Fig.

S4.9). Indeed, this occurs to be in line with the results of calculating ANOVAs of species richness according to categorized management types. However, I suggest to rather test for hypotheses regarding the impact of forest management intensity on biodiversity using the land use intensity indices as continuous variables (that reflects statistical dependencies).

5.4 Does the substrate control community structure and diversity?

The fact or the assumption, that deadwood of different tree species harbors distinctly different fungal (microbial) communities appears to be trivial at first sight. Indeed, there are recent investigations that observed community variations across deadwood of different tree species (Blaser et al. 2013; Kebli et al. 2011) but most European-based studies were objected to either Fagus sylvatica (Bässler et al. 2010; Blaschke et al. 2009;

Heilmann-Clausen & Christensen 2003, 2005; Ódor et al. 2006) or Picea abies (Allmer et al. 2006; Kubartova et al. 2012; Ovaskainen et al. 2013; Rajala et al. 2012), whereas molecular studies on European beech deadwood are lacking or were rather addressed to

167

5 Synthesis and discussion wood-decaying fungi or endophytes in living trees (Schwarze & Baum 2000;

Unterseher et al. 2013).

The experimental setup, as displayed in chapter 2, is the first to really allow for cross- comparisons of a very comprehensive set of deadwood properties (physico-chemical parameters) related to the two silviculturally important tree species of Central European forests. Regarding the fact, that we sampled logs from only one experimental site, we may exclude spatial and geographic effects on the variation of the microbial communities. Furthermore, the use of pyrosequencing allowed to link data on microbial diversity and taxonomically resolved community composition, with these wood physico-chemical properties, enzyme activity and decomposition rates, which has been rarely investigated under natural conditions. Especially regarding the fungal dataset, I intended to disentangle the ecological and environmental factors that correlate with fungal community structure and therefore hypothesized that different wood physico-chemical properties of the two model tree species could give rise to significant differences in fungal diversity and community structure.

The results presented in chapter 2 clearly indicate a substrate dependency of the fungal communities in deadwood, which has not been particularly clarified by applying molecular techniques. By performing various multivariate analyses, I was able to demonstrate, that the presence and abundances of particular fungi (fungal OTUs) significantly corresponded to variations in the physico-chemical properties in the course of wood decomposition (Figs. 2.1 and 2.3). Fig. 2.1 displays the substantial differences in wood properties between the two tree species. Especially decay class, dry mass based water content, remaining mass, wood density, C/N ratio, total lignin, and C concentration explain a significantly portion of the variation in community structure in both tree species, which is in accordance to the reports on Picea abies of (Rajala et al. 2012). I also observed 168

5 Synthesis and discussion distinct patterns of community dynamics in both types of deadwood substrates. The white-rot fungus Resinicium bicolor was dominant across all decay stages in logs of

Picea abies and was accompanied by (in lower frequencies) by Amylostereum areolatum, Heterobasidion sp. and Fomitopsis pinicola. In contrast, no such dominant fungus was observed in Fagus sylvatica logs. The Xylariceae species Annulohypoxylon cohaerens dominated decay class 1, which is characterized by high wood density, low water content and a broader C/N ratio and is then substituted by white-rot fungus Fomes formentarius. These results were also consistent with sporocarps inventories on the same logs.

I did not observe significant differences of fungal OTU richness on both deadwood substrates. The initial hypothesis to observe positive correlations between fungal diversity and ecosystem functions (here measured as lignin-modifying enzyme activity) and processes (wood decomposition rates) could also not been verified. The results of chapter 2 rather indicate some functional redundancies within the investigated fungal communities. Evidence for this redundancy comes also from laboratory studies, which showed, that certain species interactions are rather responsible for higher measured enzyme activities (Baldrian 2004; Snajdr et al. 2011). In turn, Fukami et al. (2010) and Dickie et al. (2012) observed a high degree of interaction between fungi, implying, a high individual investment into competition rather than into producing wood-degrading enzymes, which could serve as an explanation, why higher fungal species richness did not correlate with higher enzyme activity in our study.

Regarding the role of bacteria in deadwood, I hypothesized that the community structure will be significantly correlated to substrate qualities and tree species. This assumption was based upon the recent knowledge of wood-inhabiting fungi and the preliminary 169

5 Synthesis and discussion results of the distribution of nifH genes, presented in chapter 4. Results of multivariate analyses also revealed a significant variation of the bacterial community between the tree species and related decay stages. However, these results are difficult to discuss, since there are no comparable observations or field studies to date. However, the results are in line to related investigations of Nacke et al. (2011), who showed distinct bacterial community structures in soils of forests consisting of different tree species compositions. As revealed by the Goodness-of-fit statistics (Table 4.2) the factors contributing to the variations in community structure between tree species were decay class, remaining mass, volume, density, relative wood moisture, pH, mass, C/N ratio, and the concentrations and contents of C and N.

Within Fagus logs of different decay stages the wood volume, C concentrations, N content and pH were important factors that correspond to differences in bacterial community structure, while C/N explained the community variations in Picea deadwood.

These findings on microbial community structure in relation to certain investigated wood properties are of great importance to understand microbial-mediated mechanisms in deadwood, however the important question arose, whether these microbiota determine substrate qualities or whether in turn the substrate controls the structure and dynamics of this community.

The results of the all chapters are based on a sampling design for wood which pooled all samples from the same log. As introduced in Fig. 1.1 deadwood serves as a microhabitat and is characterized by small scale heterogeneity. As shown in the results of chapter 3, most wood physico-chemical properties crucially altered in the course of wood decay.

Furthermore the heterogeneity aspect can be explained by the fact that deadwood logs are not decomposed homogeniously along log length and volume. It therefore would be

170

5 Synthesis and discussion very interesting to investigate how microbial communities spatially vary within deadwood logs.

5.5 The role of N-fixation in deadwood decomposition

The methodological approach of this thesis, to combine novel molecular techniques with “classical” monitoring of sporocarps developed the idea to further investigate potential sources of N which are essential for fungi to form fruiting bodies. By analyzing data on C and N availability (own data and literature surveys), it was realized, that N in deadwood is very limited. The hypothesis, that fungi meet their requirement in N by potential interaction with free living diazotrophs was first proposed by Merrill and

Cowling (1966). In the 1970s and 1980s, several studies had confirmed biological N- fixation by performing acetylene reduction assays (ARA) (Aho et al. 1974; Seidler et al.

1972; Silvester et al. 1982; Spano et al. 1982). Due to the complexness of cultivating bacteria, little information on taxonomic identification was available at that time; e.g. bacteria of the genera Enterobacter and Klebsiella were demonstrated to be able to fix atmospheric N (Aho et al. 1974). With the development of molecular techniques Zehr and McReynolds (1989) were the first to design primer to amplify the nifH gene encoding for dinitrogenase reductase. Ever since diazotroph (N-fixing bacterial) communities have been described and investigated in various environments and ecosystems around the globe, using molecular techniques. However, the study, presented in chapter 3, was the first to investigate and verify the presence of a highly diverse N-fixing bacterial community in deadwood. By amplifying the nifH gene from wood samples of Fagus sylvatica and Picea abies, we were able to demonstrate that both tree species harbor a distinct community structure and that a considerable proportion of these genes (95%) had not been detected in any other environment and can therefore considered to being 171

5 Synthesis and discussion specific to deadwood (Fig. 3.2). The study also revealed, that nifH OTU richness significantly correlated with remaining mass% and dry mass based water content in% (on logs of Fagus silvatica), which demonstrated substrate dependencies. This is in accordance with results of Brunner and Kimmins (2003) and Hicks et al. (2003) who showed that N-fixation rates are related to wood properties and that in turn, these properties may serve as a predictor of N-fixation (Hicks 2000).

Through the performance of an extended study, targeting the 16S rRNA gene of bacteria, as presented in chapter 4, I further was able observe the presence of potential N- fixers in the same subset of deadwood logs. The significantly higher proportion of members of the Rhizobiales in the intermediate decay stage 3 (Fig. 4.2) can further serve as an explanation for the higher observed nifH OTU richness (Fig. S3.9) on Fagus sylvatica logs.

I hypothesized that nifH diversity will positively correlate with sporocarp richness. By performing a network analysis, which was based on calculations of non-random co- occurrence patterns, I further gained, together with significant correlations of sporocarp and nifH OTU richness, initial evidence that at least certain wood-inhabiting fungi are related to potential N-fixing bacteria. The finding that fungi are capable to produce methane during wood decomposition (Lenhart et al. 2012; Mukhin & Voronin 2007), together with the fact that methanotrophic bacteria rely on the released methane as their carbon source further provides indications of mutualistic interactions between fungi and bacteria in the processes of wood decomposition. As discussed in chapter 4 of this thesis, the most abundant bacterial OTU that was identified on the investigated deadwood logs was taxonomically assigned to the genus Methylovirgula. Interestingly, this bacterium has been isolated by Vorobev et al. (2009) in parallel to an inoculation experiment

172

5 Synthesis and discussion conducted on Fagus sylvatica woodblocks by Folman et al. (2008), which validated the findings presented in chapter 4.

It must be pointed out that the investigations on the distribution of nifH genes in deadwood are still lacking of any information and verification of actual N-fixation. This has also been objected by Mäkipää et al. (2014), who pointed out, that both analytical

(measurement of N-fixation rates) and molecular (composition of diazotrophs as assessed by the nifH gene) have to be performed in parallel. This could be for example realized in terms of a 15N-labeling microcosm experiment (Fig. 5.2), which combines

NGS metagenomics and bioanalytics (acetylene reduction tests and stable isotope probing SIP). The experimental setup could follow the procedures described in (Pinto-

Tomás et al. 2009).

Homogenized woodchips of the respective deadwood tree species should be placed into a microcosm pot experiment composed of an airtight container system. N could then be

15 14 amended by pulse labeling using a mixture of 80% N2 and 20% O2 or 80% N2 (as control). One, two and three weeks after the first enrichment (t1, t2, t3) respectively, wood samples could be taken for subsequent 15N-analyses of acetylene reduction

(Brunner & Kimmins 2003) to assess the actual N-fixation rates and for further DNA extractions. SIP-analyses separating 15N-labeled DNA can be performed by density centrifugation as described in (Lueders 2010; Neufeld et al. 2007). Subsequently PCRs on all resulting fractions can be performed to amplifying fungal ITS, bacterial 16S and nifH. A real-time PCR approach should be conducted on all samples to quantify total light and heavy DNA for these three markers. As a result of this sophisticated methodological approach, one could (1) identify the respective organisms that benefit from the N amendment and (2) if the nifH gets fractionated, which means that 15N is incorporated, fixation of dinitrogen would also be confirmed. The study on N-fixing bacteria present- 173

5 Synthesis and discussion ed in chapter 4 could serve as a prelude to further investigate the interactions among different microbial organisms which could contribute to better understand one substantial ecological mechanism in deadwood.

Fig. 5.2: Workflow of an SIP- (stable isotope probing) microcosm experiement. Environmental wood 15 samples are incubated with a N2-labeled substrate, and microorganisms that actively assimilate this 15 15 substrate will incorporate the N2 into their DNA. The “heavy” N2-labeled nucleic acids can then be separated from the “light” unlabeled nucleic acids by an ultracentrifugation. Microbial member then will be identified in the different fractions by applying molecular tools.

174

5 Synthesis and discussion

5.6 References

Adair S, Kim SH, Breuil C (2002) A molecular approach for early monitoring of decay basidiomycetes in wood chips. FEMS Microbiology Letters 211, 117-122. Aho PE, Seidler RJ, Evans HJ, Raju PN (1974) Distribution, enumeration, and identification of Nitrogen-fixing bacteria associated with decay in living White fir trees. Phytopathology 64, 1413-1420. Allmer J, Vasiliauskas R, Ihrmark K, Stenlid J, Dahlberg A (2006) Wood-inhabiting fungal communities in woody debris of Norway spruce (Picea abies (L.) Karst.), as reflected by sporocarps, mycelial isolations and T-RFLP identification. FEMS Microbiology Letters 55, 57-67. Anderson IC, Cairney JWG (2004) Diversity and ecology of soil fungal communities: increased understanding through the application of molecular techniques. Environmental Microbiology 6, 769-779. Arantes V, Jellison J, Goodell B (2012) Peculiarities of brown-rot fungi and biochemical Fenton reaction with regard to their potential as a model for bioprocessing biomass. Applied Microbiology and Biotechnology 94, 323-338. Bader P, Jansson S, Jonsson BG (1995) Wood-inhabiting fungi and substratum decline in selectively logged boreal spruce forests. Biological Conservation 72, 355- 362. Baldrian P (2004) Increase of laccase activity during interspecific interactions of white- rot fungi. FEMS Microbiology Ecology 50, 245-253. Baldrian P (2008) Enzymes of saprotrophic basidiomycetes. In: Ecology of saprotrophic basidiomycetes (eds. Lynne Boddy, Frankland JC, Pieter van W), pp. 19-41. Academic Press, Amsterdam. Baldrian P, Kolarik M, Stursova M, et al. (2012) Active and total microbial communities in forest soil are largely different and highly stratified during decomposition. ISME Journal 6, 248-258. Barron GL (2003) Predatory fungi, wood decay, and the carbon cycle. Biodiversity 4, 3- 9. Bässler C, Ernst R, Cadotte M, Heibl C, Müller J (2014) Near-to-nature logging influences fungal community assembly processes in a temperate forest. Journal of Applied Ecology, 939-948. Bässler C, Müller J (2010) Importance of natural disturbance for recovery of the rare polypore Antrodiella citrinella Niemelä & Ryvarden. Fungal Biology 114, 129- 133. Bässler C, Müller J, Dziock F, Brandl R (2010) Effects of resource availability and climate on the diversity of wood-decaying fungi. Journal of Ecology 98, 822- 832. Baum S, Sieber TN, Schwarze FWMR, Fink S (2003) Latent infections of Fomes fomentarius in the xylem of European beech (Fagus sylvatica). Mycological Progress 2, 141-148. Blaschke M, Helfer W, Ostrow H, et al. (2009) Naturnähezeiger–Holz bewohnende Pilze als Indikatoren für Strukturqualität im Wald. Natur und Landschaft 84, 560-566. 175

5 Synthesis and discussion

Blaser S, Prati D, Senn-Irlet B, Fischer M (2013) Effects of forest management on the diversity of deadwood-inhabiting fungi in Central European forests. Forest Ecology and Management 304, 42-48. BMELV (2011) German Forests – Nature and Economic Factor (ed. Bundesministerium für Ernährung LuVFMoF, Agriculture and Consumer Protection), Berlin. Boddy L, Heilmann-Clausen J (2008) Basidiomycete community development in temperate angiosperm wood. In: Ecology of saprotrophic basidiomycetes (eds. Lynne Boddy JCF, Pieter van W), pp. 211-237. Academic Press. Boddy L, Watkinson SC (1995) Wood decomposition, higher fungi, and their role in nutrient redistribution. Canadian Journal of Botany 73, 1377-1383. Brunner A, Kimmins JP (2003) Nitrogen fixation in coarse woody debris of Thuja plicata and Tsuga heterophylla forests on northern Vancouver Island. Canadian Journal of Forest Research 33, 1670-1682. Buée M, Reich M, Murat C, et al. (2009) 454 Pyrosequencing analyses of forest soils reveal an unexpectedly high fungal diversity. New Phytologist 184, 449-456. Carson JK, Gleeson DB, Clipson N, Murphy DV (2010) Afforestation alters community structure of soil fungi. Fungal Biology 114, 580-584. Christ S, Wubet T, Theuerl S, Herold N, Buscot F (2011) Fungal communities in bulk soil and stone compartments of different forest and soil types as revealed by a barcoding ITS rDNA and a functional laccase encoding gene marker. Soil Biology and Biochemistry 43, 1292-1299. Clausen CA (1996) Bacterial associations with decaying wood: a review. International Biodeterioration & Biodegradation 37, 101-107. Cornelissen JHC, Sass-Klaassen U, Poorter L, et al. (2012) Controls on coarse wood decay in temperate tree species: Birth of the LOGLIFE experiment. AMBIO 41, 231-245. Cowling EB, Merrill W (1966) Nitrogen in wood and its role in wood deterioration. Canadian Journal of Botany 44, 1539-1554. Daniel G, Nilsson T (1998) Developments in the study of soft rot and bacterial decay. In: Forest products biotechnology (eds. Bruce A, Palfreyman JW), pp. 37-62. Taylor & Francis, London. de Boer W, Folman LB, Klein Gunnewiek PJA, et al. (2010) Mechanism of antibacterial activity of the white-rot fungus Hypholoma fasciculare colonizing wood. Canadian Journal of Microbiology 56, 380-388. de Boer W, van der Wal A (2008) Interactions between saprotrophic basidiomycetes and bacteria. In: Ecology of saprotrophic basidiomycetes (eds. Lynne Boddy JCF, Pieter van W), pp. 143-153. Academic Press. Deacon JW (2009) Fungal biology Blackwell Publishing Ltd, Malden. Dickie IA, Fukami T, Wilkie JP, Allen RB, Buchanan PK (2012) Do assembly history effects attenuate from species to ecosystem properties? A field test with wood- inhabiting fungi. Ecology Letters 15, 133-141. Eaton RA (1994) Bacterial decay of ACQ-treated wood in a water cooling tower. International Biodeterioration & Biodegradation 33, 197-207. Edwards IP, Zak DR, Kellner H, Eisenlord SD, Pregitzer KS (2011) Simulated atmospheric N deposition alters fungal community composition and suppresses ligninolytic gene expression in a northern hardwood forest. PLoS ONE 6, e20421.

176

5 Synthesis and discussion

Fischer M, Bossdorf O, Gockel S, et al. (2010) Implementing large-scale and long-term functional biodiversity research: The Biodiversity Exploratories. Basic and Applied Ecology 11, 473-485. Fisher MM, Triplett EW (1999) Automated approach for ribosomal intergenic spacer analysis of microbial diversity and its application to freshwater bacterial communities. Applied and Environmental Microbiology 65, 4630-4636. Foley JA, DeFries R, Asner GP, et al. (2005) Global consequences of land use. Science 309, 570-574. Folman LB, Klein Gunnewiek PJA, Boddy L, de Boer W (2008) Impact of white-rot fungi on numbers and community composition of bacteria colonizing beech wood from forest soil. FEMS Microbiology Ecology 63, 181-191. Frey-Klett P, Burlinson P, Deveau A, et al. (2011) Bacterial-fungal interactions: hyphens between agricultural, clinical, environmental, and food microbiologists. Microbiology and Molecular Biology Reviews 75, 583-609. Fukami T, Dickie IA, Wilkie JP, et al. (2010) Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. Ecology Letters 13, 675-684. Ganley ARD, Kobayashi T (2007) Highly efficient concerted evolution in the ribosomal DNA repeats: Total rDNA repeat variation revealed by whole-genome shotgun sequence data. Genome Research 17, 184-191. Gillevet PM, Sikaroodi M, Torzilli AP (2009) Analyzing salt-marsh fungal diversity: comparing ARISA fingerprinting with clone sequencing and pyrosequencing. Fungal Ecology 2, 160-167. Gossner MM, Lachat T, Brunet J, et al. (2013) Current near-to-nature forest management effects on functional trait composition of saproxylic beetles in beech forests. Conservation Biology 27, 605-614. Greaves H (1971) The bacterial factor in wood decay. Wood Science and Technology 5, 6-16. Hall N (2007) Advanced sequencing technologies and their wider impact in microbiology. Journal of Experimental Biology 210, 1518-1525. Halme P, Kotiaho JS (2012) The importance of timing and number of surveys in fungal biodiversity research. Biodiversity and Conservation 21, 205-219. Harmon ME, Franklin JF, Swanson FJ, et al. (1986) Ecology of coarse woody debris in temperate ecosystems. Advances in Ecological Research 15, 133-302. Hatakka A, Hammel KE (2011) Fungal biodegradation of lignocelluloses. In: The Mycota: A comprehensive treatise on fungi as experimental systems for basic and applied research. Industrial Applications (ed. Hofrichter M), pp. 319-340. Springer, Heidelberg. Heilmann-Clausen J, Christensen M (2003) Fungal diversity on decaying beech logs - implications for sustainable forestry. Biodiversity and Conservation 12, 953- 973. Heilmann-Clausen J, Christensen M (2004) Does size matter? On the importance of various dead wood fractions for fungal diversity in Danish beech forests. Forest Ecology and Management 201, 105-119. Heilmann-Clausen J, Christensen M (2005) Wood-inhabiting macrofungi in Danish beech-forests ? conflicting diversity patterns and their implications in a conservation perspective. Biological Conservation 122, 633-642. Hendry SJ, Boddy L, Lonsdale D (2002) Abiotic variables effect differential expression of latent infections in beech (Fagus sylvatica). New Phytologist 155, 449-460. 177

5 Synthesis and discussion

Herrera ML, Vallor AC, Gelfond JA, Patterson TF, Wickes BL (2009) Strain-dependent variation in 18S ribosomal DNA copy numbers in Aspergillus fumigatus. Journal of Clinical Microbiology 47, 1325-1332. Hibbett DS, Donoghue MJ (2001) Analysis of character correlations among wood decay mechanisms, mating systems, and substrate ranges in homobasidiomycetes. Systematic Biology 50, 215-242. Hibbett DS, Taylor JW (2013) Fungal systematics: is a new age of enlightenment at hand? Nature Reviews Microbiology 11, 129-133. Hicks WT (2000) Modeling nitrogen fixation in dead wood Dissertation, Oregon State University. Hicks WT, Harmon ME, Myrold DD (2003) Substrate controls on nitrogen fixation and respiration in woody debris from the Pacific Northwest, USA. Forest Ecology and Management 176, 25-35. Hobi ML (2013) Structure and disturbance patterns of the largest European primeval beech forest revealed by terrestrial and remote sensing data, Dissertation, ETH Zürich. Jasalavich CA, Ostrofsky A, Jellison J (2000) Detection and identification of decay fungi in spruce wood by restriction fragment length polymorphism analysis of amplified genes encoding rRNA. Applied and Environmental Microbiology 66, 4725-4734. Jellison J, Connolly J, Goodell B, et al. (1997) The role of cations in the biodegradation of wood by the brown rot fungi. International Biodeterioration & Biodegradation 39, 165-179. Kahl T, Bauhus J (2014) An index of forest management intensity based on assessment of harvested tree volume, tree species composition and dead wood origin. Nature Conservation 7, 15-27. Kebli H, Drouin P, Brais S, Kernaghan G (2011) Species composition of saproxylic fungal communities on decaying logs in the boreal forest. Microbial Ecology 61, 898-910. Kellner H, Luis P, Pecyna MJ, et al. (2014) Widespread occurrence of expressed fungal secretory peroxidases in sorest soils. PLoS ONE 9. Kellner H, Luis P, Schlitt B, Buscot F (2009) Temporal changes in diversity and expression patterns of fungal laccase genes within the organic horizon of a brown forest soil. Soil Biology & Biochemistry 41, 1380-1389. Kellner H, Zak DR, Vandenbol M (2010) Fungi unearthed: Transcripts encoding lignocellulolytic and chitinolytic enzymes in forest soil. PLoS ONE 5. Kim YS, Singh AP, Nilsson T (1996) Bacteria as important degraders in waterlogged archaeological woods. Holzforschung 50, 389-392. Kreisel H (1961) Die Entwicklung der Mycozönose an Fagus-Stubben auf norddeutschen Kahlschlägen. Feddes Repertorium 139, 227-232. Krüger D, Kapturska D, Fischer C, Daniel R, Wubet T (2012) Diversity measures in environmental sequences are highly dependent on alignment quality - data from ITS and new LSU primers targeting basidiomycetes. PLoS ONE 7. Kubartova A, Ottosson E, Dahlberg A, Stenlid J (2012) Patterns of fungal communities among and within decaying logs, revealed by 454 sequencing. Molecular Ecology 21, 4514-4532. Lachat T, Bouget C, Bütler R, Müller J (2013) Deadwood: quantitative and qualitative requirements for the conservation of saproxylic biodiversity European Forest Institute. 178

5 Synthesis and discussion

Lambin EF, Turner BL, Geist HJ, et al. (2001) The causes of land-use and land-cover change: moving beyond the myths. Global Environmental Change 11, 261-269. Lenhart K, Bunge M, Ratering S, et al. (2012) Evidence for methane production by saprotrophic fungi. Nature Communincations 3. Liang B, Luo M, Scott-Herridge J, et al. (2011) A Comparison of Parallel Pyrosequencing and Sanger Clone-Based Sequencing and Its Impact on the Characterization of the Genetic Diversity of HIV-1. PLoS ONE 6, e26745. Lindahl BD, Nilsson RH, Tedersoo L, et al. (2013) Fungal community analysis by high- throughput sequencing of amplified markers – a user's guide. New Phytologist 199, 288-299. Lindblad I (1998) Wood-inhabiting fungi on fallen logs of Norway spruce: relations to forest management and substrate quality. Nordic Journal of Botany 18, 243-255. Lindner DL, Carlsen T, Nilsson HR, et al. (2013) Employing 454 amplicon pyrosequencing to reveal intragenomic divergence in the internal transcribed spacer rDNA region in fungi. Ecology and Evolution 3, 1751-1764. Lueders T (2010) Stable isotope probing of hydrocarbon-degraders. In: Handbook of Hydrocarbon and Lipid Microbiology (ed. Timmis KN), pp. 4011-4026. Springer Berlin Heidelberg. Luyssaert S, Hessenmöller D, von Lüpke N, Kaiser S, Schulze ED (2011) Quantifying land use and disturbance intensity in forestry, based on the self-thinning relationship. Ecological Applications 21, 3272-3284. Mäkipää R, Smolander A, Fritze H, et al. (2014) N fixation in decaying wood. Ecosystem services from dead wood in North European forests - Workshop on 7th-8th April 2014, Helsinki Mardis ER (2008) Next-generation DNA sequencing methods. Annual Review of Genomics and Human Genetics 9, 387-402. Margulies M, Egholm M, Altman WE, et al. (2005) Genome sequencing in microfabricated high-density picolitre reactors. Nature 437, 376-380. Merrill W, Cowling EB (1966) Role of nitrogen in wood deterioration - Amount and distribution of nitrogen in fungi. Phytopathology 56, 1083-1090. Money NP (2013) Against the naming of fungi. Fungal Biology 117, 463-465. Moore D, Gange AC, Gange EG, Boddy L (2008) Fruit bodies: Their production and development in relation to environment. In: Ecology of saprotrophic basidiomycetes (eds. Lynne Boddy JCF, Pieter van W), pp. 79-103. Academic Press. Moreth U, Schmidt O (2000) Identification of indoor rot fungi by taxon-specific priming polymerase chain reaction. Holzforschung 54, 1-8. Mouillot D, Graham NAJ, Villéger S, Mason NWH, Bellwood DR (2013) A functional approach reveals community responses to disturbances. Trends in Ecology & Evolution 28, 167-177. Mukhin VA, Voronin PY (2007) Methane emission during wood fungal decomposition. Doklady Biological Sciences 413, 159-160. Müller J, Engel H, Blaschke M (2007) Assemblages of wood-inhabiting fungi related to silvicultural management intensity in beech forests in southern Germany. European Journal of Forest Research 126, 513-527. Nacke H, Thurmer A, Wollherr A, et al. (2011) Pyrosequencing-based assessment of bacterial community structure along different management types in German forest and grassland soils. PLoS ONE6.

179

5 Synthesis and discussion

Neufeld JD, Vohra J, Dumont MG, et al. (2007) DNA stable-isotope probing. Nature Protocols 2, 860-866. Nilsson T, Björdal C (2008) Culturing wood-degrading erosion bacteria. International Biodeterioration & Biodegradation 61, 3-10. Nilsson T, Daniel G (1983) Tunneling bacteria. Ódor P, Heilmann-Clausen J, Christensen M, et al. (2006) Diversity of dead wood inhabiting fungi and bryophytes in semi-natural beech forests in Europe. Biological Conservation 131, 58-71. Ovaskainen O, Schigel D, Ali-Kovero H, et al. (2013) Combining high-throughput sequencing with fruit body surveys reveals contrasting life-history strategies in fungi. ISME Journal 7, 1696-1709. Paillet Y, Bergès L, Hjälten J, et al. (2010) Biodiversity differences between managed and unmanaged forests: Meta-Analysis of species richness in Europe. Conservation Biology 24, 101-112. Pinto-Tomás AA, Anderson MA, Suen G, et al. (2009) Symbiotic nitrogen fixation in the fungus gardens of leaf-cutter ants. Science 326, 1120-1123. Popa R, Popa R, Mashall MJ, et al. (2009) Limitations and benefits of ARISA intragenomic diversity fingerprinting. Journal of Microbiological Methods 78, 111- 118. Purahong W, Hoppe B, Kahl T, et al. (2014) Changes within a single land-use category alter microbial diversity and community structure: Molecular evidence from wood-inhabiting fungi in forest ecosystems. Journal of Environmental Management 139, 109–119. Raberg U, Hogberg NOS, Land CJ (2005) Detection and species discrimination using rDNA T-RFLP for identification of wood decay fungi. Holzforschung 59, 696- 702. 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 Ecology 4, 437-448. Rajala T, Peltoniemi M, Pennanen T, Makipaa R (2012) Fungal community dynamics in relation to substrate quality of decaying Norway spruce (Picea abies [L.] Karst.) logs in boreal forests. FEMS Microbiology Ecology 81, 494-505. Ranjard L, Poly F, Lata JC, et al. (2001) Characterization of bacterial and fungal soil communities by automated ribosomal intergenic spacer analysis fingerprints: Biological and methodological variability. Applied and Environmental Microbiology 67, 4479-4487. Rayner ADM, Todd NK (1979) Population and community structure and dynamics of fungi in decaying wood Academic Press. Rothberg JM, Leamon JH (2008) The development and impact of 454 sequencing. Nature Biotechnology 26, 1117-1124. Sala OE, Chapin III SF, Armesto JJ, et al. (2000) Global biodiversity scenarios for the year 2100. Science 287, 1770-1774. Schall P, Ammer C (2013) How to quantify forest management intensity in Central European forests. European Journal of Forest Research 132, 379-396. Schmidt O, Liese W (1994) Occurrence and significance of bacteria in wood. Holzforschung 48, 271-277. Schmidt O, Moreth U (2006) Molekulare Untersuchungen an Hausfaulepilzen. Zeitschrift für Mykologie 72, 137-152.

180

5 Synthesis and discussion

Schwarze FWMR, Baum S (2000) Mechanisms of reaction zone penetration by decay fungi in wood of beech (Fagus sylvatica). New Phytologist 146, 129-140. Schwarze FWMR, Engels J, Mattheck C (2000) Fungal strategies of wood decay in trees Springer-Verlag, Berlin, Heidelberg. Seidler RJ, Aho PE, Evans HJ, Raju PN (1972) Nitrogen fixation by bacterial isolates from decay in living White fir trees [Abies Concolor (Gord. and Glend.) Lindl.]. Journal of General Microbiology 73, 413-416. Seifert KA (2009) Progress towards DNA barcoding of fungi. Molecular Ecology Resources 9, 83-89. Siitonen J (2001) Forest management, coarse woody debris and saproxylic organisms: Fennoscandian boreal forests as an example. Ecological Bulletins, 11-41. Silvester WB, Sollins P, Verhoeven T, Cline SP (1982) Nitrogen-fixation and acetylene-reduction in decaying conifer boles - Effects of incubation-time, aeration, and moisture-content. Canadian Journal of Forest Research 12, 646- 652. Singh AP (2012) A review of microbial decay types found in wooden objects of cultural heritage recovered from buried and waterlogged environments. Journal of Cultural Heritage 13, 16-20. Singh BK, Campbell CD, Sorenson SJ, Zhou J (2009) Soil genomics. Nature Reviews Microbiology 7, 756-756. Sippola AL, Siitonen J, Kallio R (1998) Amount and quality of coarse woody debris in natural and managed coniferous forests near the timberline in Finnish Lapland. Scandinavian Journal of Forest Research 13, 204-214. Snajdr J, Dobiasova P, Vetrovsky T, et al. (2011) Saprotrophic basidiomycete mycelia and their interspecific interactions affect the spatial distribution of extracellular enzymes in soil. FEMS Microbiology Ecology 78, 80-90. Spano SD, Jurgensen MF, Larsen MJ, Harvey AE (1982) Nitrogen-fixing bacteria in Douglas-fir residue decayed by Fomitopsis pinicola. Plant and Soil 68, 117-123. Stokland JN, Siitonen J, Jonsson BG (2012) Biodiversity in dead wood Cambridge University Press, Cambridge. Strickland MS, Rousk J (2010) Considering fungal:bacterial dominance in soils – Methods, controls, and ecosystem implications. Soil Biology & Biochemistry 42, 1385-1395. Theuerl S, Buscot F (2010) Laccases: toward disentangling their diversity and functions in relation to soil organic matter cycling. Biology and Fertility of Soils 46, 215- 225. Torzilli AP, Sikaroodi M, Chalkley D, Gillevet PM (2006) A comparison of fungal communities from four salt marsh plants using automated ribosomal intergenic spacer analysis (ARISA). Mycologia 98, 690-698. Unterseher M, Peršoh D, Schnittler M (2013) Leaf-inhabiting endophytic fungi of European Beech (Fagus sylvatica L.) co-occur in leaf litter but are rare on decaying wood of the same host. Fungal Diversity 60, 43-54. Vainio EJ, Hantula J (2000) Direct analysis of wood-inhabiting fungi using denaturing gradient gel electrophoresis of amplified ribosomal DNA. Mycological Research 104, 927-936. Valaskova V, de Boer W, Klein Gunnewiek PJA, Pospisek M, Baldrian P (2009) Phylogenetic composition and properties of bacteria coexisting with the fungus Hypholoma fasciculare in decaying wood. ISME Journal 3, 1218-1221.

181

5 Synthesis and discussion

Volkenant K (2007) Totholz als Lebensraum von Mycozönosen im fortschreitenden Zersetzungsprozess - Eine Chronosequenzstudie an Fagus sylvatica-Totholz im Nationalpark Kellerwald-Edersee, Dissertation, Universität Kassel. Vorobev AV, de Boer W, Folman LB, et al. (2009) Methylovirgula ligni gen. nov., sp nov., an obligately acidophilic, facultatively methylotrophic bacterium with a highly divergent mxaF gene. International Journal of Systematic and Evolutionary Microbiology 59, 2538-2545. Weig AR, Peršoh D, Werner S, Betzlbacher A, Rambold G (2013) Diagnostic assessment of mycodiversity in environmental samples by fungal ITS1 rDNA length polymorphism. Mycological Progress 12, 719-725. Will C, Thurmer A, Wollherr A, et al. (2010) Horizon-specific bacterial community composition of German grassland soils, as revealed by pyrosequencing-based analysis of 16S rRNA genes. Applied and Environmental Microbiology 76, 6751-6759. Wubet T, Christ S, Schoning I, et al. (2012) Differences in soil fungal communities between European beech (Fagus sylvatica L.) dominated forests are related to soil and understory vegetation. PLoS ONE 7. Zehr JP, McReynolds LA (1989) Use of degenerate oligonucleotides for amplification of the nifH gene from the marine Cyanobacterium Trichodesmium thiebautii. Applied and Environmental Microbiology 55, 2522-2526. Zhang HB, Yang MX, Tu R (2008) Unexpectedly high bacterial diversity in decaying wood of a conifer as revealed by a molecular method. International Biodeterio- ration & Biodegradation 62, 471-474.

182

Acknowledgements

Ich widme diese Arbeit meinen geliebten Großeltern Gertrud und Wilfried Hoppe. Nach euerm leisen Gehen hab ich verstanden, dass ich außer bei Euch, nirgendwo in dieser

Welt und diesem Leben so sein darf, wie ich will und kann. Kein Rechtfertigen, keine unnützen Worte. Stattdessen pure Liebe. Das fehlt mir so sehr.

Ich hab nie gedacht, dies zu schreiben, aber ich widme diese Arbeit ebenso sehr Peter

Moll. Ich bin froh darüber Teil deiner Familie zu sein und ich habe viel von Dir gelernt, bezüglich ein guter Vater und Mann zu sein. Ich danke Dir für deine zauberhafte Toch- ter und ich werde auf sie aufpassen.

Dear Prof. François Buscot, I am so thankful for what I personally and professionally received from you. I can be talking to you anytime and anywhere. You are way more than a head of a department. The day that I had the interview in Halle and turned back to Dresden, home at that time, checking mails and receiving a positive reply immediately, was one of the happiest moments of my live. I was dancing, I can tell you.

Dear Prof. Jürgen Bauhus, I also have to express my gratefulness towards you. I really enjoy the discussions we have and your patience of course. It’s no coincidence that you are among the best and pleasant Professors in Germany. And I knew that, even before I got know you.

183

Acknowledgements

Dear Prof. Siegfried Fink, I gratefully appreciate you reviewing my thesis. I went to your office and you immediately aggred. Thank you very much.

Dear Dr. Dirk Krüger, my pleasure to be part of your working group. Even though it took quite a while to get this thesis finished, I really learned a lot from you and most of it all, I think lots of the ideas I’m now graduating on, are due to a really good and in- formal communication we share. And whenever you had money left to spend on my scientific career, you were there for me. This is not for granted and I really appreciate that.

Dear Dr. Tiemo Kahl, it’s been hard time discussing with you every once in a while, but truly, fairly and delightful. I honestly wish that there would be more of you in the science “business”. And of course, I can be talking to you about anything that comes along, except soccer. And I have to point out, that I would have never had the chance to work in this project without your initiation. And the fact, that we all made it into a second prolongation of this research endevour, is just fantastic. Thanks for that.

Dear Dr. Tesfaye Wubet, your door is always open, you always smooth and calm things down. And you always seriously pay attention to what I’m asking you. I really made big progress by the time that we collaborated.

Dear Witoon Purahong, it’s just a pleasure to work with you and I’m glad, that we really got to collaborate now.

184

Acknowledgements

Lieber Theo, lieber Hannes, Ihr seid das größte Geschenk, was mir das Leben bescheren konnte. Ihr seid so tolle Jungs und ich liebe Euch über alles.

Julichen;) Du bist ne tolle Frau, mit Allem, was dazu gehört. Und ich bin der Glücks- pilz, der an deiner Seite durchs Leben schlendern darf.

Mama und Papa, ihr seid großartig. Absolute Vorbilder, zielstrebig, liebevoll, lebensbe- jahend und fröhlich, umtriebig und aufmerksam. Ich bin total dankbar und glücklich darüber Euer Sohn zu sein.

Es sind weit über 20 Jahre inniger Freundschaft Fabian, und ich will keinen gemeinsa- men Moment missen. Auch Euch, Malte, Veiko und Flo danke ich dafür, dass ihr immer da seid und wart. Kezia, du bist großartig und ich verbringe sehr gern Zeit mit dir. Du hilfst mit stets aus der Patsche. Annette, Ralf und Tobi, ihr Drei habt Halle zur coolsten

Stadt Mitteldeutschlands gemacht.

Many thanks to Sigrid Härtling and Bea Schnabel, it would have not been possible to present this wonderful data without your big contributions. Renate Rudloff, you’re great and always there for me. Thanks for that.

Ulf Mallast is the creative art director of this thesis.

The Department of soil ecology in Halle is a wonderful place to work at. I’m thankful to be a part of this working group.

I have to thank the DFG for giving me and some more people the opportunity to work in this wonderful project. It’s named “FunWood”. 185