Research Collection

Doctoral Thesis

Stability of soil microbial communities to applications of the fungal biological control agent Metarhizium brunneum

Author(s): Mayerhofer, Johanna

Publication Date: 2017

Permanent Link: https://doi.org/10.3929/ethz-b-000244602

Rights / License: In Copyright - Non-Commercial Use Permitted

This page was generated automatically upon download from the ETH Zurich Research Collection. For more information please consult the Terms of use.

ETH Library

DISS. ETH Nr. 24581

STABILITY OF SOIL MICROBIAL COMMUNITIES TO APPLICATIONS OF THE FUNGAL BIOLOGICAL CONTROL AGENT METARHIZIUM BRUNNEUM

a dissertation submitted to attain the degree of

DOCTOR OF SCIENCES of ETH ZURICH

(Dr. sc. ETH Zurich)

presented by

JOHANNA MAYERHOFER

MSc Mikrobiologie, Karl-Franzens Universität Innsbruck

born on 03.04.1988

citizen of Austria

accepted on the recommendation of

Prof. Dr. Adrian Leuchtmann

Prof. Dr. Bruce McDonald

Dr. Jürg Enkerli

Prof. Dr. Stefan Vidal

2017

CONTENT

Content

Content ...... 3 Summary ...... 7 Zusammenfassung ...... 9 General introduction ...... 11 1.1 The click beetles Agriotes spp...... 11

Distribution and damage potential of Agriotes spp...... 11

Life cycle of Agriotes spp...... 12

Control options against Agriotes spp...... 13

1.2 The Western corn rootworm Diabrotica virgifera virgifera ...... 14

Distribution and damage potential of D. v. virgifera ...... 14

Life cycle of D. v. virgifera ...... 15

Control options against D. v. virgifera ...... 15

1.3 Soil microorganisms ...... 16

Diversity and function of soil microorganisms ...... 16

Assessing soil microbial diversity ...... 18

Response of soil microbial communities to disturbances ...... 19

1.4 The biocontrol agent and entomopathogenic fungus Metarhizium ...... 20

Biological control using entomopathogenic fungi and its history ...... 20

The entomopathogenic fungus Metarhizium ...... 21

1.4.2.1 Ecology and life cycle of the entomopathogenic fungus Metarhizium ...... 21

1.4.2.2 Taxonomy and species concept in fungi with focus on Metarhizium ...... 22

1.4.2.3 Phylogeny of Metarhizium ...... 23

1.4.2.4 Detection and identification of Metarhizium strains ...... 24

Non-target effects of Metarhizium ...... 25

1.5 Objectives ...... 27

1.6 Outline of the thesis ...... 28

Multiplexed microsatellite markers for seven Metarhizium species ...... 29 2.1 Abstract ...... 29

2.2 Introduction ...... 29

2.3 Material and methods ...... 30 3

CONTENT

2.4 Results and discussion ...... 30

2.5 Acknowledgements ...... 33

Assessing effects of the entomopathogenic fungus Metarhizium brunneum on soil microbial communities in Agriotes spp. biological pest control ...... 35 3.1 Abstract ...... 35

3.2 Introduction ...... 35

3.3 Material and methods ...... 37

Rearing of Agriotes obscurus larvae ...... 37

Treatments ...... 37

Set-up of the pot experiment ...... 38

Set-up of the field experiment ...... 38

Processing of soil samples, isolation of Metarhizium CFU and identification of applied strain ..... 39

DNA extraction from soil, PCR and Illumina sequencing ...... 40

ITS2 sequence of the applied strain ...... 41

Sequence processing and taxonomic classification ...... 41

Statistical analyses ...... 41

3.4 Results ...... 42

Abundance of the applied Metarhizium strain and efficacy of biocontrol treatments in pots ...... 42

Abundance of the applied Metarhizium strain and efficacy of biocontrol treatments in the field ... 44

Soil microbial communities of the pots ...... 45

Abundance of the applied strain and effects of treatments on soil microorganisms in pots ...... 45

Changes of the microbial communities over time in pots ...... 48

Soil microbial communities in the field ...... 48

Abundance of the applied strain and effects of treatments on microbial communities in the field . 49

Changes of the microbial communities over time and space in the field ...... 49

3.5 Discussion ...... 51

3.6 Funding ...... 54

3.7 Acknowledgments ...... 54

Response of soil microbial communities to the application of a formulated Metarhizium brunneum biocontrol strain ...... 55 4.1 Abstract ...... 55

4.2 Introduction ...... 55

4.3 Material and methods ...... 57 4

CONTENT

Set-up of the pot experiment ...... 57

Treatments ...... 57

Application and assessment of D. v. virgifera and root damage of maize ...... 57

Soil sampling ...... 58

Monitoring of the applied strain ...... 58

Determination of the ITS2 sequence of the applied strain ...... 58

Analyses of microbial community composition ...... 58

Sequence processing and taxonomic classification ...... 59

Statistical analyses ...... 59

4.4 Results ...... 60

Abundance of the applied strain, number of adult beetles and damage assessment ...... 60

Soil microbial diversity ...... 61

Effects of treatments on microbial diversity ...... 61

Temporal changes of the microbial diversity ...... 63

Comparison of effects on soil microbial communities to a similar pot experiment ...... 65

4.5 Discussion ...... 66

4.6 Acknowledgements ...... 68

Evaluating PCR biases of one dominant target on the assessment of community structures of a soil microbiome ...... 69 5.1 Introduction ...... 69

5.2 Material and methods ...... 70

Soil sample ...... 70

Plasmid construction and addition ...... 70

Amplicon sequencing ...... 71

Sequence processing, taxonomic classification and statistical analyses ...... 71

5.3 Results ...... 71

Assessment of the sequence data obtained ...... 71

The OTU containing the sequence of pITS2-ART2825 ...... 71

Potential effects of spiking with plasmid pITS2-ART2825 ...... 72

5.4 Discussion ...... 74

General discussion ...... 77 6.1 Quantifying exposure of soil microorganisms to applied biocontrol strains ...... 77

5

CONTENT

Abundance of the applied strains using a cultivation-dependent approach ...... 77

Detection of the applied strains using amplicon sequencing ...... 78

ITS2 for phylogenetic resolution of Metarhizium ...... 79

Alternative approaches for detection of the applied strain ...... 80

6.2 Potential effects of M. brunneum strains on soil microorganisms ...... 81

Do inundative applications of M. brunneum affect soil microbial communities? ...... 81

Sequencing errors and limitations of amplicon sequencing ...... 85

Alternative future approaches ...... 86

Comparison to other hypocralean entomopathogenic fungi ...... 86

6.3 Efficacy of biological control using M. brunneum ...... 86

6.4 Overall conclusions ...... 87

6.5 Perspectives ...... 88

Identification of applied fungal strains using SSR analyses ...... 88

Methodological aspects regarding effects on soil microbial communities ...... 88

Experimental aspects regarding effects on soil microbial communities ...... 89

Appendix ...... 91 A) Supporting information to Chapter 2 ...... 91

B) Supporting information to Chapter 3 ...... 98

C) Supporting information to Chapter 4 ...... 121

D) Supporting information to General discussion ...... 124

References ...... 132 Acknowledgements ...... 149

6

SUMMARY

Summary

Naturally occurring entomopathogens, including fungi, bacteria, protozoa, nematodes and viruses, are important for natural regulation of insect populations and therefore used in biological control of insect pests. Biological control using entomopathogens is an alternative pest management strategy to chemical control, is considered to be more environmental friendly than chemical control and is often part of integrated pest management (IPM). Beneficial effects of entomopathogens as compared to chemicals may include conservation of other natural enemies, reduction of pesticide residues in soil and plants as well as safety for non-target organisms. The development of existing biological control for important soil dwelling pests in Europe by exploiting novel application strategies and synergistic effects of entomopathogenic fungi, entomopathogenic nematodes and natural substances, has been the major aim of the EU-project supporting this thesis (INBIOSOIL). Biological control using entomopathogenic fungi may include soil applications of large quantities of fungal propagules, often resulting in densities of up to 10 14 propagules per ha. Such mass applications and introductions of microorganisms to soil may affect indigenous soil microbial communities and the ecosystem functions they fulfil. Assessment of potential effects of applied microorganisms on native soil microbial communities is therefore important and also required in the registration process of novel products by European and Swiss regulations. Species of the entomopathogenic fungal genus Metarhizium are widely used in biological control and potential effects of Metarhizium on native soil microbial communities have barely been investigated. As a basic principal of risk assessment of mass applications it is mandatory to verify that soil microbial communities are exposed to the applied fungal biological control agent (BCA). It is important to be able to determine both exposure (presence and abundance) as well as effects, e.g., changes in microbial community structures, of a BCA in soil as basis of a risk assessment. Therefore, the aims of this thesis were I) to improve molecular tools to assess presence and abundance of the BCA in soil in order to track exposure of soil microbial communities and II) to assess potential effects of applications of two strains of the BCA on soil microbial communities in pot and field experiments and III) to investigate potential analytical constraints of the presence of one highly abundant microbial strain on the assessment of changes in soil microbial community structures. Exposure of soil microbial communities to the BCA was assessed by isolation of Metarhizium spp. from soil on selective medium and subsequently identifying the genotype of the applied strains using simple sequence repeat (SSR) marker analyses. To optimize the typing tool the transferability of 41 existing SSR markers to different Metarhizium spp. were investigated (Chapter 2). This was particularly important since structure and taxonomy within the genus Metarhizium have been drastically revised in the recent past. Among other changes, the species M. anisopliae has been recognized as a species complex and has been divided into 10 species. Suitable markers were selected for genotyping of the applied strains and discriminating them from native Metarhizium strains present in soils. Biocontrol experiments were performed to control Agriotes obscurus larvae in a pot and a field experiment and Diabrotica virgifera virgifera larvae in a pot experiment using the BCA M. brunneum ART2825 and M. brunneum EAMa01/58-Su over a period of four months (Chapter 3 and 4). Treatments included three different formulations of the BCA, i.e. fungus colonized barley kernels (FCBK), fungal capsules (F cap ) and fungal granules (F gran ), unformulated fungal spores, combinations of the BCA and garlic extract, garlic extract alone and controls. Potential effects of M. brunneum on fungal and prokaryotic soil communities were assessed using next 7

SUMMARY generation sequencing (NGS) of ribosomal marker regions, i.e., the internal transcribed spacer 2 (ITS2) for fungal soil communities and part of the small subunit of the ribosomal RNA gene (16S V3-V4) for prokaryotic communities. One of the biases occurring in NGS is preferential amplification of target sequences in the PCR due to the occurrence of one highly abundant sequence. In order to assess if the sequence of the applied strain, which was highly abundant, will affect the assessment of changes of soil microbial communities, an experiment was performed in which different concentration of a plasmid containing the target sequence of the BCA was added to a soil DNA extract (Chapter 5). Results revealed that M. brunneum ART2825 formulated as FCBK was the most efficient treatment against A. obscurus larvae in the pot experiment, which resulted in 77 % reduction of damaged potato tubers compared to the untreated control (Chapter 3). No biocontrol effect was detected in the Agriotes -field experiment as well as in the Diabrotica -pot experiment (Chapter 3 and 4). Selective plating and SSR marker analyses confirmed exposure of soil microbial communities to the applied strains in the pot and field experiments and built the basis for effect analyses (Chapter 3 and 4). In the Agriotes -pot experiment, M. brunneum ART2825 affected the fungal community structures only slightly and only if formulated. These effects were in the same range as effects caused by the carrier material only and no effects were detected if pure fungal spores were applied to soil. In contrast, prokaryotic community structures were not affected by any fungal applications. Also, fungal and prokaryotic communities did not change upon the application of the BCA in the Diabrotica -pot and the Agriotes -field experiment. Finally, assessment of effects, i.e., changes in soil microbial community structures, were not impeded by the presence of one highly abundant sequence or sequencing errors that potentially corroborate the reliability of the results (Chapter 5). In conclusion, genotyping using SSR marker analyses was a valuable tool for confirming inoculation success and exposure of soil microbial communities to the applied strains. Although, constraints of amplicon sequencing, such as PCR bias, erroneous sequences and limitations in species identification, may occur, it is currently the most suitable technique, for the sequencing depth and the large number of samples and replicates required in experiments as performed in this study. Natural fluctuations including spatial and temporal differences of soil microbial community structures were similar or greater than any treatment-effects detected by applying different formulations of the BCA or unformulated fungal spores. Therefore, it is unlikely that applications of M. brunneum ART2825 and M. brunneum EAMa01/58-Su adversely affect soil microbial communities beyond the extent induced by fluctuations occurring in natural habitats. However, the findings of this study are restricted to specific BCA strains, soils, experimental designs, and analytical procedures and may require confirmation in other systems or if new analytical tools become available.

8

ZUSAMMENFASSUNG

Zusammenfassung

Die Infektion von Insekten mit natürlich vorkommenden Pilzen, Bakterien, Protozoen, Nematoden und Viren ist die Grundlage der mikrobiellen biologischen Schädlingsbekämpfung von Insekten in der Landwirtschaft. Biologische Schädlingsbekämpfung ist in vielen Fällen Teil eines integrierten Schädlingsbekämpfungsprogramms und stellt eine umweltfreundliche Alternative zu chemischen Insektiziden dar. Durch gezieltere Wirkung auf den Schädling werden dabei natürliche Antagonisten der Schädlinge und andere Bodenorganismen geschont und es gelangen weniger belastende Pestizide in die Umwelt und in Lebensmittel. Diese Doktorarbeit ist Teil des EU- Projekts INBIOSOIL, das sich zum Ziel gesetzt hat, Schädlinge, die in Europa grosse Schäden verursachen und einen Teil ihres Lebenszyklus im Boden verbringen, mit biologischen Mitteln zu bekämpfen. Im Rahmen des EU- Projekts wurden neue Applikationsstrategien entwickelt und synergistische Wirkungen zwischen entomopathogenen Pilzen, Nematoden und natürlichen Substanzen untersucht. Bei der Massenapplikation von entomopathogenen Pilzen werden grosse Mengen an Pilzsporen und Myzel appliziert. Das kann zu hohen Pilzdichten von bis zu 10 14 koloniebildenden Einheiten pro Hektar im Boden führen. Solch hohe Pilzdichten könnten Bodenmikroorganismen sowie ihre ökologische Funktionen im Boden möglicherweise beeinträchtigen. Deswegen ist es wichtig, Untersuchungen über potentielle Einflüsse von entomopathogenen Pilzen auf Bodenmikroorganismen durchzuführen. Obwohl der entomopathogene Pilz Metarhizium in der biologischen Schädlingsbekämpfung häufig eingesetzt wird, sind potentielle Effekte von Metarhizium auf die Gemeinschaftsstruktur von Bodenmikroorganismen bis jetzt selten untersucht. Diese Untersuchungen sind Teil der Risikobewertung einer Massenapplikation von Pilzsporen und Myzel, werden auch für die Registrierung von neuen Produkten verlangt und sind in der EU und der Schweiz durch gesetzliche Verordnungen geregelt. Grundlagen dafür sind einerseits die Sicherstellung der Exposition der Bodenmikroorganismen durch den applizierten Pilz und andererseits die Erfassung der Veränderungen der mikrobiellen Bodengemeinschaften. Daraus ergaben sich die folgenden drei Ziele dieser Doktorarbeit: I) Die Verbesserung molekularer Methoden um den applizierten Pilzstamm im Boden nachweisen zu können, II) Untersuchungen von Effekten der applizierten Pilzstämme auf natürlich vorkommende Bodenmikroorganismen in Topfversuchen wie auch im Feld und III) Bestimmung möglicher Einschränkungen der molekularen Analysen der mikrobiellen Gemeinschaftsstruktur, die durch die Präsenz einer sehr abundanten Sequenz hervorgerufen werden könnten. Im vorliegenden Projekt wurde zuerst die Pilzdichte von Metarhizium im Boden wurde auf Selektivnährmedium bestimmt. Darauf folgte die Genotypisierung der Metarhizium -Isolate mittels genetischer Marker (Mikrosatelliten), um den applizierten Stamm zu identifizieren und von natürlich vorkommenden Metarhizium - Stämmen zu unterscheiden (Kapitel 3 und 4). Da die Taxonomie des Genus Metarhizium vor kurzem grundlegend verändert wurde (z. B. entstanden zehn Arten aus der davor als M. anisopliae benannten Art), musste die Übertragbarkeit von 41 genetischen Markern zwischen verschiedenen Metarhizium -Arten getestet werden (Kapitel 2). Von den 41 Mikrosatellitenmarkern wurden sechs hochauflösende Marker für die Identifikation der applizierten Metarhizium -Stämme ausgewählt. Topf- und Feldversuche zur Bekämpfung von Agriotes obscurus und Diabrotica virgifera virgifera Larven mit den Pilzstämmen M. brunneum ART2825 und M. brunneum EAMa01/58-Su wurden über einen Zeitraum von 4 Monaten durchgeführt (Kapitel 3 und 4). Folgende Behandlungen wurden appliziert: Drei verschiedene Formulierungen des Pilzes, und zwar Pilzgerste (FCBK),

9

ZUSAMMENFASSUNG

Pilzkapseln (F cap ) und Pilzgranulat (F gran ), des weiteren Pilzsporen ohne Formulierung, Kombinationen des Pilzes mit Knoblauchextrakt, Knoblauchextrakt allein, ein Insektizid und unbehandelte Kontrollen. Effekte auf die mikrobielle Gemeinschaftsstruktur im Boden wurden mit Hilfe der Sequenzierung von ribosomalen Markerregionen (ITS2 für die Gemeinschaftsstrukturanalyse der Pilze und 16S V3-V4 für die Gemeinschaftsstrukturanalyse der Prokaryoten) mittels Next-Generation Sequencing (NGS) untersucht. Eine der bekannten Fehlerquellen in der NGS Analyse ist die präferentielle Amplifikation von Sequenzen in der PCR, ausgelöst durch das Vorkommen einer Sequenz in sehr hoher Abundanz. Um herauszufinden, ob diese Fehlerquelle die Analysen der mikrobiellen Gemeinschaftsstrukturen beeinflusst hat, wurde in einem Experiment die Sequenz des applizierten Stammes in ein Plasmid integriert und dann in unterschiedlichen Mengen zu dem gleichen DNA-Extrakt hinzugefügt (Kapitel 5). Die Ergebnisse zeigten die erfolgreichste Applikation, und zwar 77 % Reduktion beschädigter Kartoffeln im Vergleich zu der Kontrolle, konnte mit M. brunneum ART2825 formuliert in FCBK erzielt werden (Kapitel 3). In dem Topfversuch gegen D. v. virgifera und dem Feldversuch gegen Agriotes spp. wurden keine Kontrolleffekte gemessen (Kapitel 3 und 4). Die Mikrosatellitenanalysen bestätigten jedoch die Präsenz der applizierten Pilzstämme im Boden. Somit war die Exposition der Bodenmikroorganismen bestätigt und die Grundvoraussetzung für die Effektstudien gewährleistet. Im Topfversuch zur Bekämpfung von Agriotes spp. konnten geringfügige Effekte von M. brunneum ART2825 in Formulierung aber nicht als reine Pilzsporen auf die Gemeinschaftsstruktur der Bodenpilze detektiert werden (Kapitel 3). Zudem war das Ausmass der gemessenen Effekte vergleichbar mit dem Ausmass an Effekten durch die Applikation der Formulierung allein, ohne dem Pilz. Im Gegensatz dazu wurden gar keine Veränderungen in der Gemeinschaftsstruktur der Prokaryoten nach der Applikation von M. brunneum gefunden (Kapitel 3 und 4). Im Feldversuch gegen Agriotes spp. und im Topfversuch gegen D. v. virgifera wurden keine Effekte durch die zwei M. brunneum Stämme detektiert. Ausserdem ergab die Untersuchung über den potentiellen Einfluss einer sehr abundanten Sequenz auf die Erfassbarkeit der Veränderungen der Gemeinschaftsstrukturen, dass mit grosser Wahrscheinlichkeit keine Einschränkungen vorhanden waren (Kapitel 5). Zusammenfassend konnten folgende Schlussfolgerungen aus dieser Arbeit gezogen werden: Die Isolation von Metarhizium auf Selektivnährmedium gefolgt von der Analyse der Genotypen mittels Mikrosatellitenanalyse ist eine effiziente Methode um den Inokulationserfolg und die Exposition der Bodenmikroorganismen mit den applizierten Stämmen nachzuweisen. Obwohl verschiedene Einschränkungen der NGS Technik, wie z. B. fehlerhafte Sequenzen und limitierte Bestimmung auf Artniveau möglicherweise vorkommen, ist diese Methode zurzeit und in Anbetracht der Sequenziertiefe und der grossen Anzahl an Proben, die für Effektstudien nötig sind, die geeignetste. Natürliche Fluktuationen sowie Veränderungen der Gemeinschaftsstruktur der Bodenmikroorganismen über Zeit und Raum waren ähnlich gross oder weitreichender als Effekte ausgelöst durch die Applikationen der formulierten Pilzstämme. Deswegen ist es unwahrscheinlich, dass Applikationen von M. brunneum ART2825 und M. brunneum EAMa01/58-Su einen negativen Effekt auf die Gemeinschaftsstrukturen der Bodenmikroorganismen, die über natürlich vorkommende Schwankungen hinausgehen, haben. Anzumerken ist jedoch, dass die Resultate, die aus diesen Studien hervorgingen, spezifisch für die Pilzstämme, den Boden, das experimentelle Design und die verwendeten Analysen sind und daher möglicherweise nicht übertragbar in andere Systeme sind. Zur Sicherung der Erkenntnisse könnten Untersuchungen mit neu entwickelten Methoden wiederholt werden.

10

GENERAL INTRODUCTION

General introduction

Biological control is an important alternative pest management strategy to chemical control and is defined as ‘the use of living organisms to suppress the population density or impact of a specific pest organism, making it less abundant or less damaging than it would otherwise be’ (Eilenberg et al., 2001). Living organisms used in biological pest control are predators, parasitoids and pathogens, including bacteria, fungi, protista, viruses and nematodes (Eilenberg et al., 2001; Lacey et al., 2001). Biological control of insect species is often used in combination with other control measures in integrated pest management. Alternatives to chemical pesticides are recommended by EU law, for instance the EU Council Directive 2009/128/EC deals with the sustainable use of pesticides and includes the statement that ‘Member States shall take all necessary measures to promote low pesticide-input pest management, giving wherever possible priority to non-chemical methods’ (Council Directive, 2009). The advantage of insect pest management using entomopathogenic organisms instead of chemical insecticides may result from reduced chemical input into the environment and therefore less pesticide residues in the food, increased safety for humans and other non-target organisms such as natural enemies of pests and increased biodiversity in agriculturally managed areas (Lacey et al., 2001). Before a new biological control agent is commercialized potential effects on non-target organisms have to be evaluated and this is required by law in Europe (European Commission, 2011) by executive order in Switzerland in the “Verordnung über das Inverkehrbringen von Pflanzenschutzmitteln 916.161” from 12 May 2010. Soil microorganisms constitute a major part in the ecosystem soil and therefore potential effects of biological control agents on soil microorganisms have to be assessed. The major aim of the thesis was to investigate potential effects of the biological control agent and entomopathogenic fungus Metarhizium brunneum Petch (Hypocreales: Clavicipitaceae) on soil microbial communities. Potential effects were studied in two pot experiment and one field experiment. In order to conduct such studies the exposure of soil microorganisms to M. brunneum has to be ensured. For this purpose, a DNA based tool was improved. The general introduction of this thesis starts with information about each of the two pest organisms, i.e. Agriotes spp. and Diabrotica virgifera virgifera , which were targeted in the experiments to assess potential effects on soil microbial communities. This is followed by an introduction to soil microbial communities and potential disturbances of their community structures, as well as methods to assess them. The introduction is concluded with a chapter about general information about M. brunneum, its phylogeny, methods to identify Metarhizium at species and strain level and knowledge about potential effects on non-target organisms.

1.1 The click beetles Agriotes spp.

Distribution and damage potential of Agriotes spp.

Agriotes spp. Eschscholtz (Coleoptera: Elateridae) are polyphagous insect pests recognized as major and global insect pests for centuries (Traugott et al., 2015; Vernon and Van Herk, 2012). The genus Agriotes includes 274 species of which eight ( A. lineatus, A. obscurus, A. sputator, A. sordidus, A. rufipalpis, A. brevis, A. litigiosus and A. ustulatus ) cause economically relevant damage in Europe (Furlan et al., 2001). Species composition differs among regions and species differ in biology, ecology and crop preference which has important implications for control options (Ritter and Richter, 2013). The most common species in Europe are A. lineatus followed by A.

11

GENERAL INTRODUCTION obscurus and A. sputator and in Switzerland A. ustulatus in addition (Ritter and Richter, 2013). Agriotes spp. are generalists and their larvae (also called wireworms) damage lower stem parts and roots of many different plants such as maize, potato, oil crops and vegetables (Hill, 1987). In potatoes larval feeding results in holes and tunnels, which may increase the occurrence of fungal infection of the potato tubers with Rhizoctonia solani (Keiser et al., 2012) . Already low abundance of Agriotes spp. larvae (100’000 larvae / ha) may cause economic damage because they reduce potato tuber quality rather than yield (Parker and Howard, 2001). Usually one hole per potato tuber is enough to render it unmarketable for human consumption. In the USA and UK typically 5 – 25 % of potato tuber yield are lost due to feeding by Agriotes spp. larvae, even if insecticides were used (Parker and Howard, 2001) and in Switzerland damage of up to 60 % of the potato tubers has been reported (Keiser, 2007).

Life cycle of Agriotes spp.

The life cycles of different Agriotes spp. have been shown to vary among species, and not the same degree of information is available for all economically relevant species (Barsics et al., 2013; Ritter and Richter, 2013). Furthermore, the duration of the life cycle may vary within individuals of the same species. Life cycles of A. obscurus (Figure 1.1) and A. ustulatus , which have been studied thoroughly, may last between one to six years (Ford, 1917; Furlan, 1996; Furlan, 1998; Sufyan et al., 2014). A. obscurus adults (Figure 1.1) overwinter buried in the soil while A. ustulatus overwinter as larvae (Furlan, 1998). Eggs with a diameter of approximately 0.5 mm are laid individually or in clusters into the upper layer of soil between May and June for A. obscurus and between July and August for A. ustulatus . Larvae (Figure 1.1) hatch after three to six weeks but time points of oviposition and hatching differs depending on temperature and if data were assessed in a field experiment or in a laboratory assay (Furlan, 1996). After hatching larvae may pass through up to 13 larval stages (Furlan, 1998; Sufyan et al., 2014). Usually larvae show two intense feeding periods per season, i.e. from March to May and September to October (Parker and Howard, 2001). These feeding periods coincide with periods of favourable soil conditions (moisture and temperature) for the larvae. Low moisture and high temperature has shown to induces downward movement into deeper soil layers (Furlan, 1998; Lafrance, 1968), where it is very difficult to apply any control measures. Agriotes larvae usually pupate between May and August and hatching occurs within approximately two weeks for A. obscurus and within a few days for A. ustulatus . The among species variation of life cycles of Agriotes spp., the within species variability of the duration of life cycles, the moving behaviour within soil layers, the differences in susceptibility to control measures at different developmental stages and the co-occurrence of different species in the same field make control of Agriotes spp. very challenging (Ritter and Richter, 2013; Traugott et al., 2015).

12

GENERAL INTRODUCTION

Figure 1.1: A. obscurus adult (A; Image from G. Brändle, Agroscope, Switzerland) and larva (B; Image from C. Schweizer, Agroscope, Switzerland). Control options against Agriotes spp.

Control options include insecticides, management practices such as tillage and crop rotation, mating disruption with male sex pheromones and biological control using entomopathogenic fungi or nematodes. Several different insecticides have been used for control of Agriotes larvae. In the past effective insecticides included persistent organochlorides, such as aldrin and lindane, and carbamates such as carbofuran (Maskell, 1958; Merrill, 1952; Parker and Howard, 2001). But these have been prohibited in the European Union due to environmental concerns. Newly developed insecticides such as chlorpyriphos (organophosphate) have shown inconsistent efficacies (Parker and Howard, 2001; Vernon et al., 2008) and neonictonioids such as clothianidin, thiamethoxam and imidacloprid as well as pyrethroids such as tefluthrin have shown a repellent but no toxic effect which resulted in crop protection but no reduction of number of larvae (van Herk et al., 2008b; van Herk et al., 2008c; Vernon et al., 2009). The widely used phenyl pyrazol fipronil proved to be toxic to Agriotes larvae (Furlan and Toffanin, 1998; Vernon et al., 2009; Vernon et al., 2013), however, this compound has also been found to be toxic for bees (Zaluski et al., 2015), and it is not registered in Switzerland. Besides insecticides, different soil management strategies such as tillage, crop rotations and trap crops may result in a control effect on Agriotes spp. The effectivity of tillage is under debate (Ritter and Richter, 2013). Nevertheless, because eggs and first instar larvae are the most susceptible life stages to tillage, and A. ustulatus and A. litigiosus have short and concerted oviposition in spring, tillage may be applicable for these species (Furlan, 2005). Abundance of Agriotes spp. larvae has been shown to increase or decrease when certain plants are grown in a field. For instance, crops increasing the density of larvae such as wheat and peas ( Pisum sativum ) may be useful for trap crops (Landl and Glauninger, 2013; Parker, 1996). In contrast, crop rotations including, for instance, buckwheat or brown mustard resulted in decreased density of Agriotes larvae (Noronha, 2011), however, they might not be applicable because these crops may not be in demand. Another control option is biological control. For control of Agriotes spp. entomopathogenic nematodes and fungi have shown the highest potential compared to other biological control agents such as predators and parasitoids (Bognar, 1955; Miles and Cohen, 1941; Ritter and Richter, 2013). For instance, the entomopathogenic nematode Heterorhabditis bacteriophora caused 67% mortality of A. lineatus larvae in a laboratory assay (Ansari et al., 2009), however, a very limited control effect was observed in a laboratory assay to control A. obscurus at the

13

GENERAL INTRODUCTION

Institute for Sustainability Sciences, Agroscope, (personal communication G. Grabenweger). Certain strains of entomopathogenic fungi have shown great potential to infect Agriotes larvae, i.e., Beauveria bassiana (Ansari et al., 2009; Ester and Huiting, 2007; Ladurner et al., 2009) and Metarhizium strains belonging to the former species complex Metarhizium anisopliae (Kabaluk et al., 2005; Kabaluk et al., 2007; Kölliker et al., 2011). Furthermore, combining these control measures with other approaches has resulted in successful control. For instance, the application of M. anisopliae and sex pheromones have been applied to attract Agriotes beetles to spots of high fungal density (Kabaluk et al., 2015). Sex pheromones applied alone in order to perform mating disruption may also provide potential to control Agriotes spp. (Reddy and Tangtrakulwanich, 2014). Male sex pheromones have been developed for the eight most important species in Europe, but they have been mostly used for monitoring purpose (Ritter and Richter, 2013; Tóth and Furlan, 2005; Tóth et al., 2003). Costs for mating disruption are high and differ among species due to differences in the required number of traps. Therefore, the applicability of mating disruption depends on species, crop value and pest pressure (Hicks and Blackshaw, 2008; Ritter and Richter, 2013). Another possible combination of different control measures is the application of Metarhizium together with botanicals such as garlic (Eckard et al., 2017) or neem cake, a natural pesticide from neem tree (Ritter and Richter, 2013). Most of the above mentioned non-chemical measures provide insufficient control, however, integrating and combining different control options and strategies is a promising approach for the control of Agriotes spp.

1.2 The Western corn rootworm Diabrotica virgifera virgifera

Distribution and damage potential of D. v. virgifera

Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae) is an important pest species predominantly in maize and an invasive species in Europe. Its success as a pest can be attributed to its specificity on maize (Zea mays L.) and the small number of naturally occurring enemies in Europe because of its recent invasion. The genus Diabrotica probably originated from Central America. Subsequently, it moved together with its host to southwestern USA about 3000 years ago. In the 1950s and 1980s D. v. virgifera spread throughout the corn belt and to the east coast of North America, respectively (Krysan and Smith, 1987). Then it was introduced to Europe during several occasions (Ciosi et al., 2008; Miller et al., 2005). First it was detected close to Belgrade airport in 1992 (Bača, 1994). Since then, it has been observed in many European countries and it has successfully invaded (i.e., continuously spread in) central and south-eastern Europe and north-western Italy. Moreover, models based on bioclimatic parameters, physiological information on D. v. virgifera the range of maize cultivation areas predict potential expansion to great parts of Asia (Aragón et al., 2010). Due to its economic importance D. v. virgifera is known as “the billion dollar bug” causing a billion dollar damage per year in the USA (Metcalf, 1986). In Europe, scenarios on implementing measures to control the further spread of D. v. virgifera revealed economic benefits between 143 to 1739 million Euro (Wesseler and Fall, 2010). Maize plants are damaged by D. v. virgifera larvae as well as by the adults. Pruning of maize roots by the larvae leads to a declined nutrient and water uptake. This causes in dislodging of the maize plants which makes it difficult to harvest the plants. Furthermore it results in lower grain yields as well as in a lower nutrient content of the plants (Kahler et al., 1985). Adults feed on maize pollen, kernels, foliage and silk, which results in maize ears with low numbers of kernels and in higher risk of fungal infections of the ears (Chiang, 1973).

14

GENERAL INTRODUCTION

Life cycle of D. v. virgifera

The life cycle of D. v. virgifera (Figure 1.2) is univoltine with overwintering eggs (Chiang, 1973). Larvae hatch depending on temperature in May or June and then pass through three larval instars (Figure 1.2) until they pupate and emerge as adult beetles (Figure 1.2) between mid-June and early-August. Oviposition into soil takes place in maize fields between August and mid-September (Branson and Krysan, 1981). In the field females laid on average 150 eggs, but numbers of up to 500 eggs were observed (Toepfer and Kuhlmann, 2006). Although 99 % natural mortality was observed from the egg stage to hatching of adults, with eggs and the first instar larvae being the most vulnerable developmental stages, 500 eggs would still lead to a strong population increase (Toepfer and Kuhlmann, 2006).

Figure 1.2: D. v. virgifera adult on maize leaf (A) and larva (B; Image from H. Strasser, University of Innsbruck, Austria). Control options against D. v. virgifera

Control options for D. v. virgifera larvae and adults are needed and they comprise applications of chemical and biological insecticides, implementation of crop rotations and use of resistant maize varieties obtained by breeding as well as genetically modified maize plants. Aboveground foliar applications target D. v. virgifera adults and belowground soil applications target the larvae. One of the oldest and still most applied means of control is crop rotations to interrupt the life cycle of D. v. virgifera , which has a very narrow host range (Hill et al., 1948; Kiss et al., 2005). D. v. virgifera females lay eggs preferably in maize fields and larvae have a strong preference for maize roots (Branson and Krysan, 1981). Adaption to crop rotation has been shown for some D. v. virgifera populations to crops like soybean ( Glycine max L.), pumpkin ( Curcubita pepo L.) and Miscanthus grass (Hummel et al., 2008; Levine et al., 2002; Spencer and Raghu, 2009). A prolonged diapause or the presence of volunteer maize in the field in a 1-year rotation scheme may have been the reason for the adaption (Krysan et al., 1984; Levine and Oloumi-Sadeghi, 1991). In the USA chemical control with chlorinated insecticide (DDT and benzene hexachloride) was recommended in the 1940s (Hill et al., 1948) and later cyclodienes (aldrin and dieldrin) were used but D. v. virgifera developed 15

GENERAL INTRODUCTION resistances and the insecticides were banned due to environmental concern (Metcalf, 1983). Carbamates and organophosphates were used as replacement, however, resistance to these insecticides was reported in the mid- 1990s (Meinke et al., 1998). Today, insecticides based on neonicotinoids and pyrethroids are used in the USA and so far no resistance has been detected (van Rozen and Ester, 2010). In the EU neonicotinoids are banned (European Commission, 2013) and pyrethroids require approval in some European countries (CABI, 2017). Soil applications of insecticides targeting larvae are usually preferred because they are less expensive than foliar applications and because of the negative impact of foliar applications on beneficial insects and predatory mites (Gerber et al., 2005). A possibility to decrease insecticidal input in aboveground applications is the combination of insecticides with an attractant. For instance, application of the attractant cucurbitacin allowed to reduce the dosage of the insecticide by 95 to 98 %, and this strategies was successful for some D. v. virgifera populations in the USA (Chandler, 2003). Chemical control options are often not satisfactory, not applicable due to environmental concerns and / or resistances have developed and therefore alternatives such as biological control with predator, parasitoids and pathogens are of great interest. Parasitoids, e.g., Celatoria compressa , entomopathogenic nematodes, e.g., Heterorhabditis bacteriophora and entomopathogenic fungi, e.g., Beauveria spp. and Metarhizium spp. are potentially useful for biological control of D. v. virgifera (Kuhlmann et al., 2005; Pilz et al., 2009; Pilz et al., 2008; Toepfer et al., 2010). In a survey it has been shown that D. v. virgifera were naturally infected with Metarhizium spp. and Beauveria spp. in fields of some European countries including Hungary, Romania, Serbia, Austria and Italy (Pilz et al., 2008). The application of M. brunneum resulted in a 31 % reduction in abundance of D. v. virgifera adults (Pilz et al., 2009). Whereas Rauch et al. (2017) did not find a control effect when a combination of M. brunneum and the nematode H. bacteriophora was applied. Conventional or molecular assisted breeding, or genetic modification may provide alternative control options for D. v. virgifera . In the USA maize varieties with quick root regeneration were developed and provided tolerance against D. v. virgifera (Jenison et al., 1981). Molecular assisted breeding programs aim at developing resistant maize varieties in the USA (Gray et al., 2009). Transgenic maize varieties expressing one or more toxins of Bacillus thuringiensis (Bt-maize) proved to be very efficient in controlling D. v. virgifera in the USA, however, resistances to some of them have developed (Devos et al., 2013; Jakka et al., 2016; Narva et al., 2013) and in the European Union Bt-maize against D. v. virgifera is not approved. Optimizations and combinations of the above discussed control options within integrated pest management strategies are needed to control D. v. virgifera (Toepfer and Kuhlmann, 2004). EU authorities recommend the preferable use of non-chemical methods by using crop rotation and biological control agents and by changing maize sowing dates to avoid coincidence of hatching of the larvae and germination of the maize seeds, as well as removal of volunteer plants and cleansing of machinery (European Commission, 2014).

1.3 Soil microorganisms

Diversity and function of soil microorganisms

Soil microorganisms include prokaryotic (bacteria, archaea) and eukaryotic organisms (fungi, protista, algae). However, in this thesis the term soil microorganisms refers to archaea, bacteria and fungi. Soils are characterized by complex structures based on aggregation of differently sized particles. The resulting pores provide niches for soil microorganisms (Lennon et al., 2012). Microbial communities are very diverse; abundances of 4 x 10 3 to 5 x 10 4 bacterial species per g -1 dry weight soil, more than 10 4 soil fungal species (total number of known soil fungi), up to 10 10 bacterial and about 10 6 fungal cells per g -1 dry weight soil (Bridge and Spooner, 2001; Roesch et al.,

16

GENERAL INTRODUCTION

2007; Torsvik et al., 1990; Trevors, 2009). Soil microorganisms are influenced by their immediate chemical, physical and biological environment but they also affect their environment by degrading (Ritz, 2004). The abundance of soil microorganisms varies among different soil compartments. For instance, the rhizosphere, which is defined as the soil compartment influenced by the roots of growing plants (Hartmann et al., 2008; Hiltner, 1904), provides nutrients in form of root exudates for soil microorganisms which results in a 10 – 100 fold higher abundances of soil microorganisms in the rhizosphere compared to the surrounding rather oligotroph bulk soil (Lugtenberg, 2015a). Another highly complex microenvironment for soil microorganisms are biofilms. Biofilms contain soil microorganisms within extracellular polysaccharide structures (Burns and Stach, 2002). These biofilms often form at the interface of soil and water on surfaces of soil particles where nutrients are bound. Biofilms provide a matrix in which microorganisms exchange genetic material or communicate via signalling molecules known as quorum sensing (Dessaux et al., 2011). Quorum sensing involves the production of small signalling molecules (such as N-acyl homoserine lactone) by bacterial populations until a threshold or “quorum” is reached which results in the regulation of expression of specific quorum sensing genes. In addition to the spatial heterogeneity of soil and the resulting variability in abundance and composition of soil microorganisms, soil conditions also fluctuate over time. Soil microorganisms may adapt to these fluctuations by changing between active and non-active states such as dormancy and thereby maintaining high soil biodiversity (Jones and Lennon, 2010). Microbial cells become dormant by entering a resting state or by forming resting spores. It has been shown that depending on nutrient availability a great proportion of microbial cells in soil are dormant (may become active within hours to days), and only 0.1 – 5 % are active microorganisms which contribute to ongoing processes (Blagodatskaya and Kuzyakov, 2013). If there is a high substrate input the active part may become the most abundant fraction of soil microorganisms. However, it is still experimentally challenging to differentiate between active, dormant and dead microbial cells in soil (Blagodatskaya and Kuzyakov, 2013). The wealth of different habitats within soil and temporal changes result in a high diversity of soil microorganisms. Diversity of soil microorganisms is based on the presence and abundance of certain taxa but alternatively diversity measures may be based on functions provided by soil microorganisms. Soil microorganisms play a crucial role in many different soil functions such as nutrient cycling, degradation of organic and xenobiotic material and formation of soil structure, and especially important for agricultural soils, plant growth promotion as well as pest control (Kennedy, 1999; Treseder and Lennon, 2015). Soil microorganisms are at the base of global nutrient cycling (Price, 1988). One key aspect in nutrient cycling is the decomposition of organic material and xenobiotic compounds leading to the formation of soil humus, providing substrate for other organisms and improving soil physical properties, such as water holding capacity (Kennedy, 1999). For instance, some fungi are capable of degrading cellulose and lignin which are major components of the cell walls of plants and therefor belong to the most abundant biopolymers on earth (Klemm et al., 2005). Another example for the role of soil microorganisms in nutrient cycling is the well- studied group of nitrogen fixing bacteria which are able to transform atmospheric nitrogen to plant available nitrogen (Beijerinck, 1901), and therefor they constitute an important step in the nitrogen cycle. Another soil ecosystem function provided by soil microorganisms is the contribution to soil structure by binding soil aggregates with extracellular polysaccharides or fungal hyphae (Lynch and Bragg, 1985; Ritz and Young, 2004) which leads to aeration and water infiltration into the soil. Furthermore, fungi also directly interact with plants e.g. in form of mycorrhiza. These fungi provide nutrients to plants, prevent invasion of pathogens by niche exclusion and may induce seed germination (Bridge and Spooner, 2001). Soil microorganisms also play a role in plant health by

17

GENERAL INTRODUCTION controlling pests, such as insects and nematodes, weeds and pathogens, such as bacteria, fungi, oomycetes and bacteria (Whipps, 2001).

Assessing soil microbial diversity

There is a great variety of techniques to determine the diversity of microorganisms in soils and these methods can be divided into cultivation-dependent or cultivation-independent methods (Agrawal et al., 2015; Hill et al., 2000; Rastogi and Sani, 2011). Cultivation-dependent methods based on plating soil suspensions on nutrient media have been standard methods until recently (Hill et al., 2000). The greatest drawback of plating on nutrient media is that only a small fraction of different soil microbes are detectable, i.e., only less than 0.1 percent of soil microbes were culturable (Torsvik et al., 1990). Although several different nutrient media have been developed in order to increase recovery of different microbes there is still a great bias towards fast growing microbes (e.g. Vieira and Nahas, 2005). Advantages of cultivation-dependent methods are that single isolates are available for subsequently studying their ecology, function and interaction with other microbes. Also, it is relatively cheap compared to cultivation-independent methods. Another cultivation-dependent approach is community level physiological profiling, e.g., based on carbon utilization patterns which allows discrimination of communities based on differences in carbon utilization. Although today cultivation-dependent methods are mostly replaced by cultivation-independent methods to study soil microorganisms, they remain valuable tools for the assessment of abundance of specific taxa, e.g. for entomopathogenic fungi (Strasser et al., 1996). Cultivation-independent methods assess diversity of microorganisms based on molecules such as phospholipid fatty acids (PFLA) or nucleic acids, or based on microscopic evaluation of fluorescently labelled cells (reviewed in Hill et al., 2000; Rastogi and Sani, 2011). PFLA profiling takes advantage of differences of PFLAs in the membrane of microorganisms which allows the discrimination of specific groups at different taxonomic or functional levels, e.g., fungi, protozoa, microalgae, aerobic bacteria, anaerobic bacteria, sulphate-reducing bacteria, Bacillus pp. or Flavobacterium balustinum (Hill et al., 2000). An advantage of this method is the ability to assess the living fraction of soil microorganisms because of the rapid degradation of cell membranes after cell death. Disadvantages comprise the inability to discriminate low taxonomic ranks for most groups, the dependence of PFLA profiles on growth conditions and environmental factors and the discrepancy in abundance of PFLAs between groups, for instance between fungi and bacteria (Haack et al., 1994). Compared to the assessment of PFLAs for soil microbial diversity, techniques based on differences in DNA and RNA composition provide higher resolution and may be used to classify down to species. Methods based on RNA are able to assess living cells, while DNA based methods may include both living, dead and dormant cells (Wellington et al., 2003). Nucleic acid based methods comprise profiling techniques such as denaturing gradient gel electrophoresis (DGGE), single strand conformation polymorphism (SSCP), terminal restriction fragment length polymorphism (TRFLP) or automated ribosomal intergenic spacer analysis (ARISA), sequencing techniques of barcodes, whole genomes or transcriptomes and quantification of specific taxonomic or functional taxa by qPCR, microarrays or fluorescence in situ hybridization (Rastogi and Sani, 2011). Profiling methods provide quantitative differences in diversity of microorganisms and parallel analysis of multiple samples, but no taxonomic information unless individual bands (DNA sections) are excised and sequenced (Muyzer, 1999). In contrast to profiling methods, sequencing approaches allow the taxonomic classification down to species or even sub-species level depending on which region in the genome is targeted. Initial sequencing approaches involved construction of clone libraries (plasmids containing one sequence each) and subsequently sequencing individual clones. Later this technique was replaced 18

GENERAL INTRODUCTION by next-generation sequencing (NGS) allowing assessment of tens of thousands of different sequences in one sample and therefore increasing resolution by far. NGS is used for approaches such as amplicon sequencing, also known as barcoding approach, metagenomics and metatranscriptomics (Rastogi and Sani, 2011). Amplicon sequencing involves a PCR based amplification of a marker region, e.g., ribosomal RNA gene cluster, from soil DNA extracts (Lindahl et al., 2013). PCR products obtained from such amplifications represent mixtures of marker sequences from all microorganisms present in the soil sample. Amplicon sequencing may be hampered by DNA extraction and PCR biases. DNAs yield differ between DNA extraction protocols, therefore we used the same DNA extraction protocol for all samples. Also, in order to increase DNA extraction efficiency the mechanical disruption of cells was repeated three times (Bürgmann et al., 2001). DNA extracts of soil samples may contain co-extracts, such as humic acids, which may hamper PCR. Therefore DNA extracts were cleaned using the NucleoSpin® gNDA clean-up kit (Machery-Nagel, Germany). Another known bias which may occur during PCR is preferential amplification of certain taxa (Lindahl et al., 2013). To ameliorate this bias PCR was repeated four times for each sample and resulting PCR products were pooled. Also, degenerate primer pairs were used to include many different taxa during PCR amplification. Several different sequencing platforms exist (i.e., Illumina, 454 Pyrosequencing, Iron Torrent) and provide sequences with different lengths (Liu et al., 2012; Quail et al., 2012). We chose Illumina MiSeq v3 which provides sequencing of about 300 bp length to cover the ITS2 and the V3-V4 of the 16S subunit for fungal and prokaryotic sequences. It is known that NGS is error prone, and for Illumina MiSeq an error rate of 0.8 % was observed which requires thorough quality filtering. High quality sequences with 97 % identity are grouped to operational taxonomic units (OTU) which are used as equivalents to species. Occurrence and abundance of OTUs are compared among samples. Diversity indices are used to describe the OTU composition of samples, e.g. OTU richness per sample or among samples, e.g. Bray Curtis dissimilarity (Bray and Curtis, 1957; Legendre and Legendre, 2012). In order to deduce the taxonomic affiliation of each OTU its sequence is compared to a curated databases. The most frequently used region with the most comprehensive database available for taxonomic classification is the ribosomal gene cluster, which is ubiquitous in all organisms, is structurally and functionally conserved and includes conserved and highly variable regions (Pace et al., 1986; Woese, 1987). In contrast to amplicon sequencing, metagenomics and metatranscriptomics aim at sequencing the whole genome or transcriptome ideally of all organisms in a soil sample which includes a tremendous amount of taxonomic and functional information. However, for a high number of samples in one sequencing run, metagenomics and metatranscriptomics will be inferior because resolution will be lost due to a limited sequencing depth (Rastogi and Sani, 2011). Amplicon sequencing, however, offers a high throughput application and therefore it is state of the art for soil microbial diversity studies (Lindahl et al., 2013).

Response of soil microbial communities to disturbances

Due to the diverse function of soil microorganisms a major focus in microbiological research is the response of soil microorganisms to disturbances (Shade et al., 2012). Disturbance is a key concept in ecology and it refers to a causal event that either impacts a community directly or its immediate environment which in turn affects the community (Glasby and Underwood, 1996; Rykiel, 1985). The stability of a community is defined as its response to disturbance and one approach to measure stability is to use resistance, i.e. the degree of withstanding to a disturbance, or resilience, i.e. the rate at which a microbial community returns to its original composition (Allison and Martiny, 2008; Pimm, 1984; Shade et al., 2012). In order to assess stability it is necessary to define the natural fluctuations of a community (Shade et al., 2012). Natural fluctuations may develop as a result of environmental 19

GENERAL INTRODUCTION impacts but also by endogenous dynamics, i.e., changes that are caused by interactions among species that occur even under constant environmental conditions (Konopka et al., 2015). Another important issue in stability of soil microbial communities is the debate if microbial communities reach equilibria or if they undergo a continuous flux (Konopka et al., 2015; Shade et al., 2013; Shade et al., 2012). The topic becomes even more complex if one considers that soil microbial communities have different compositions and functions, and stability of composition and function may differ for the same community, i.e., a change in taxonomy might not be coupled with a change in function because the same function may be provided by different species (functional redundancy). Therefore both compositional and functional stability have to be assessed to comprehensively describe a community.

1.4 The biocontrol agent and entomopathogenic fungus Metarhizium

Biological control using entomopathogenic fungi and its history

Pest management is an integral part of agriculture and probably evolved from the beginning of agriculture about 10’000 B.C. For centuries people have developed different pest management strategies ranging from banning pests from the field by religious leaders to the use of high-tech products such as genetically modified organisms (Boulter, 1993; Orlob, 1973). In the late 19 th century chemical pest management was developed and is still widely used, however, due to environmental concerns many chemicals have been banned whereupon research focused on the development of more environmental friendly alternatives, such as biological control agents. Fungal infections are often found in and represent important antagonists of arthropod populations which make entomopathogenic fungi potential candidates for biological pest control (reviewed in Hajek and St Leger, 1994; Lacey et al., 2015). In comparison to biological control using arthropods, entomopathogenic fungi can be applied with conventional agricultural machines and their production and storage is relatively cheap and easy (Lacey et al., 2001). An advantage of fungal entomopathogens compared to entomopathogenic bacteria, nematodes or viruses is their ability to penetrate their host through the exoskeleton and therefore oral ingestion is not a requirement for infection (Hajek and St Leger, 1994). Outbreaks of infections in an arthropod population with entomopathogenic fungi are called epizootics and have been observed frequently in nature, for instance, in populations of the aphid Diuraphis noxia Kurdjumov, the moth Lymantria dispar L. populations or the maybeetle Melolontha melolontha (Hajek et al., 1990; Keller and Zimmermann, 1989; Knudsen et al., 1994). Taking advantage of the occurrence of natural epizootics is the aim of conservation biological control, which relies only on manipulating the environment in order to favour natural outbreaks (Eilenberg et al., 2001). Often this may not be sufficient, because natural outbreaks are difficult to predict and manipulate, and they often depend on high densities of the pest above the economic threshold (Lacey et al., 2001). In order to induce or accelerate epizootics, entomopathogenic fungi are usually applied using inundative application, which includes application of high densities of infective propagules (Eilenberg et al., 2001). Factors which might impede the development of epizootics include abiotic factors, such as sensitivity to solar radiation, presence of pesticides, temperature and humidity, and biotic factors, such as microbial antagonists, host behaviour, physiological conditions of the host, pathogen vigour and inoculum threshold. In order to understand these factors and their potential interactions basic knowledge about the ecology of entomopathogenic fungal strains is crucial (Hajek and St Leger, 1994; Lacey et al., 2001). Entomopathogenic fungi were already studied in the early 19 th century in order to protect silkworm production and honey bees (Hall and Papierok, 1982; Lacey et al., 2001) by the well-known pioneers, Agostino Bassi, Luis Pasteur and John Lawrence Le Conte. The first mass production of Metarhizium anisopliae for biological control was

20

GENERAL INTRODUCTION undertaken by the Russian biologists Ilja Iljitsch Metchnikoff and Isaac Krassilstschik in the late 19th century (Krassilstschik, 1888; Metchnikoff, 1879). Approximately 700 – 1000 entomopathogenic fungal species are known (Hajek and St Leger, 1994; St Leger and Wang, 2010) and found within four fungal phyla, i.e., Ascomycota, Basidiomycota, Zygomycota and Chytridiomycota (Rehner, 2009). Today, the most commonly commercially applied fungal entomopathogens belong to the genera Beauveria spp. and Metarhizium spp. (Figure 1.3) followed by Isaria spp. (formerly Paecilomyces spp.) and Lecanicillium spp. (Faria and Wraight, 2007). In 2007 products containing a strain of the species complex Metarhizium anisopliae were available to control species of Acari (Ixodidae, Tetranychidae), Blattodea (Blattidae, Blattellidae), Coleoptera (Curculionidae, Nitidulidae, Scarabidae), Diptera (Ephydridae, Mycetophilidae, Sciaridae, Tipulidae), Hemiptera (Aphididae, Cercopidae, Cicadellidae, Delphacidae, Miridae, Pentatomidae), Isoptera (Kalotermitidae, Rhinotermitidae, Termopsidae), Hymenoptera (Formicidae), Lepidoptera (Crambidae, Noctuidae), Orthoptera, Siphonaptera (Pulicidae), Thysanoptera (Thripidae) in various different countries worldwide (Faria and Wraight, 2007). Challenges in the use of entomopathogens in biological control are often linked to low persistence, low speed of kill, too narrow or too broad host range and high costs linked to shelf-life if compared with chemical pesticides (Lacey et al., 2001). To overcome these challenges research focuses on novel formulations, optimization of application (timing, etc.) and combination with other chemical or botanical agents and strategies, such as mating disruption using pheromones or attraction to traps (Brandl et al., 2017; Eckard et al., 2017; Kabaluk et al., 2015; Schumann et al., 2014).

The entomopathogenic fungus Metarhizium

1.4.2.1 Ecology and life cycle of the entomopathogenic fungus Metarhizium

Metarhizium spp. grow and produce high amounts of conidia on their arthropod host (Figure 1.3), however, they also occur in soil and recently their potential to interact with plants has been realized (Meyling and Eilenberg, 2007). Different strains of Metarhizium spp. are able to infect different arthropod species (204 species listed by Veen, 1968; Zimmermann, 2007) of the classes Insecta and Arachnida. With its host, Metarhizium spp. form a hemibiotrophic / saprotrophic relationship in which the fungus feeds on dead and living or only on dead hosts (Shah and Pell, 2003). Its life cycle in insects has been described in several different hosts (e.g. Hänel, 1982; McCauley et al., 1968; Schabel, 1978) and recently research has focused on molecular mechanisms of and genes involved in the infection process (Schrank and Vainstein, 2010; St Leger and Wang, 2010; Wang and Feng, 2014). Research on genes involved in the infection process is also driven by the endeavour to develop genetically modified Metarhizium strains with increased virulence (St Leger and Wang, 2010; St. Leger et al., 1996). The first step in the infection process consists of Metarhizium spp. conidia attaching to the host cuticle via adhesins and hydrophobins (Fang et al., 2007; Li et al., 2010; Wang and St Leger, 2007). Recognition of the host is complex and probably induces the expression of host specific proteins (Schrank and Vainstein, 2010). After attachment conidia germinate and degrade the lipid layer of the epicuticle (Beys da Silva et al., 2010). Then germination tubes differentiate to appressoria. The production of proteases, lipases, and chitinases are involved in the penetration of the fungus through the cuticle (Beys da Silva et al., 2010; Boldo et al., 2009; Wang et al., 2002). The active penetration through the cuticle is a special trait of entomopathogenic fungi compared to other entomopathogens such as bacteria or viruses which have to be ingested (Hajek and St Leger, 1994). The fungus grows further into the body cavity of the host and produces blastospores which are distributed throughout the haemolymph and allow

21

GENERAL INTRODUCTION infection of the whole arthropod body. Metarhizium spp. produce a cocktail of insecticidal and pharmaceutical compounds including destruxins (Wang et al., 2012). The ability of producing destruxins and also other secondary metabolites involved in virulence seems to be a trait of Metarhizium spp. with a broad host range, such as M. brunneum, M. robertsii and M. anisopliae as compared to those with a narrow one, such as M. acridum and M. majus (Hu et al., 2014; Wang et al., 2012). Also, Metarhizium spp. produces trehalases to utilize trehalose the main sugar found in the haemolymph (Zhao et al., 2007). Once the insect is dead and nutrients are depleted Metarhizium spp. grows out of the insect body and produce high numbers of conidia on the surface of the cadaver for dispersal. Entomopathogenic fungi are dispersed by abiotic factors, such as water or air, or by biotic factors, such as migrating host and non-host arthropods (Dromph, 2003; Dromph and Vestergaard, 2002; Hajek, 1997).

Figure 1.3: Mycelium and conidia of M. brunneum infecting and growing out of a Tenebrio molitor larva on a nutrient medium agar plate (A) and mycelium of Metarhizium spp. growing out through grooves between segments of a dead A. obscurus larva after infection (B; G. Grabenweger, Agroscope, Switzerland). Metarhizium is part of the soil microbial community (Domsch et al., 1980) and is frequently found in agricultural soil (e.g. Bidochka et al., 1998; Meyling and Eilenberg, 2007; Steinwender et al., 2014). Compared to opportunistic saprotrophic fungi in soil Metarhizium is a poor competitor for nutrients (Hajek, 1997; Keller and Zimmermann, 1989), therefore, more likely Metarhizium rests in soil until it comes in contact with its host (Meyling and Eilenberg, 2007). Recently, interaction of Metarhizium with plants have been reported. Possible interaction include plant growth promotion, induction of seed germination, translocation of insect nitrogen to plants, antagonism to plant pathogens and induction of plant resistance to osmotic stress (Behie et al., 2012; Khan et al., 2012; Sasan and Bidochka, 2012). In the rhizosphere, Metarhizium was found in higher densities than in the bulk soil (Hu and St Leger, 2002; St. Leger, 2008; Wang et al., 2005), and recently endophytic growth has been reported (Barelli et al., 2016; Behie et al., 2015; Vega et al., 2009). Molecular evidence suggests that Metarhizium is related to the fungal endosymbionts Claviceps and Epichloë (Spatafora et al., 2007) and more closely related to endophytes and plant pathogens than to animal pathogens, which was also observed for other entomopathogenic fungi (Gao et al., 2011; Shang et al., 2016). However, the ecological role of Metarhizium spp. in plant associations and a resulting potential for biological control still needs to be fully elucidated.

1.4.2.2 Taxonomy and species concept in fungi with focus on Metarhizium

The Clavicipitaceae recently underwent several taxonomic reclassifications and changes based on molecular multi locus phylogenetic studies. Additionally, some species within Clavicipitaceae were renamed in order to fulfil the

22

GENERAL INTRODUCTION recent endeavour to agree on one name for sexual and asexual forms of the same fungus, which was proposed by the in the International Code of Nomenclature for algae, fungi and plants (ICN).(Kepler et al., 2014; McNeill et al., 2012; Taylor, 2011). The genus Metarhizium was formerly described as asexual, however, using molecular tool it was linked to the sexual fungus Metacordyceps (Kepler et al., 2014; Kepler et al., 2012; Sung et al., 2007). However, other asexual fungal species and members of the genera Pochonia and Nomuraea were also linked to Metacordyceps . Species concepts are necessary to accurately define species which is a prerequisite to study their characteristics, such as biogeography, genetic structure and functional ecology (Rehner, 2009). Different species concepts have been developed and out of 22 different ones the theoretical Evolutionary Species Concept was evaluated as the most appropriate and therefore primary concept by Mayden (1997). The Evolutionary Species concept defines a species as “… a single lineage of ancestor-descendent populations which maintains its identity from other such lineages and which has its own evolutionary tendencies and historical fate” (Taylor et al., 2000; Wiley, 1978). Secondary operational species concepts are useful for practical study of species and therefore termed as “Species Recognition” by Taylor et al. (2000). Examples are the Biological, Morphological and Phylogenetic Species Recognition. The Morphological Species Recognition for entomopathogenic fungi is limited due to phenotypic plasticity and simple morphological features (Rehner, 2009). The Biological Species Recognition, which is based on reproductively isolated groups, is constrained by the lack of sexual reproduction or experimental challenges to prove sexual reproduction within entomopathogenic fungi (Rehner, 2009). However, the exclusive presence of the mating types in some Metarhizium strains in most species suggests sexual recombination (Pattemore et al., 2014; Rehner and Kepler, 2017). The Phylogenetic Species Recognition, based on shared feature from ancestors (typically nucleotide sequences), is probably most suitable to the Evolutionary Species Concept in fungi (Eldredge and Cracraft, 1980; Rehner, 2009; Taylor et al., 2000). The most recent phylogenetic delineation of Metarhizium spp. is based on phylogenetic consensus of multiple nucleotide loci (Bischoff et al., 2009; Kepler et al., 2014; Rehner and Kepler, 2017).

1.4.2.3 Phylogeny of Metarhizium

Morphological and ecological characteristics are not informative enough to generate monophyletic groups within neither the family Clavicipitaceae nor the genus Metarhizium , therefore, molecular tools are necessary to disentangle their phylogeny (Kepler et al., 2014). The first marker gene used for taxonomic classification of Metarhizium was the internal transcribed spacer region (ITS) which discriminated M. anisopliae , M. flavoviridae and M. album, and multiple varieties were suggested within these species (Driver et al., 2000). The most recent taxonomy is based on sequences of four different genes, i.e., beta tubulin, RNA polymerase II largest subunit, RNA polymerase II second largest subunit and translation elongation factor 1 alpha (tef1a ) (Bischoff et al., 2009; Kepler et al., 2014), however, these genes may underrepresent the diversity of the M. anisopliae species complex (Kepler and Rehner, 2013; Rocha et al., 2013). The search for more informative loci revealed seven nuclear intergenic sequence markers (Kepler and Rehner, 2013), which are now available for species identification. The recent taxonomical revision of Metarhizium defined 30 species (Bischoff et al., 2009; Kepler et al., 2014), however, new species are still described and added (Chu et al., 2016; Montalva et al., 2016) and Mycobank includes 64 Metarhizium spp. and varieties at the moment (3 July 2017; Crous et al., 2004). The former M. anisopliae species complex was divided into the well-defined PARB clade, including M. pingshaense , M. anisopliae , M. robertsii a nd M. brunneum , a clade consisting of M. majus , M. guizhouense and 23

GENERAL INTRODUCTION

M. indigotica and three additional species, i.e., M. acridum , M. lepidiotae , and M. globosum (Kepler et al., 2016; Kepler et al., 2014) (Figure 1.4). However, cryptic lineages within these clades have been found and suggest further changes in the Metarhizium phylogeny (Rehner and Kepler, 2017).

Figure 1.4: Unrooted phylogenetic tree of 24 partial sequences of the tef1a representing ten species of the former M. anisopliae species complex using Maximum likelihood method based on the Jukes-Cantor model (Jukes and Cantor, 1969), 1000 times bootstrapping and standard settings in MEGA 6 (Tamura et al., 2013).

Previously, most Metarhizium spp. used as biological control agents belonged to M. anisopliae or M. anisopliae var. acridum (Faria and Wraight, 2007). However, due to the phylogenetic revision species identity of these biological control strains has to be re-assessed. Furthermore, studies on the ecology, e.g., occurrence, host specificity, reproductive biology or habitat adaption of Metarhizium species, have to be re-evaluated.

1.4.2.4 Detection and identification of Metarhizium strains

Tools and techniques that allow identification of Metarhizium species and strains are important to study ecological aspects, evolution and community structure of Metarhizium spp. Furthermore, detection and identification of Metarhizium strains enables discrimination of applied BCA strains from naturally occurring strains and therefore research on the behaviour as a biological control agent. Detection and quantification of Metarhizium is performed using cultivation-dependent and / or cultivation-independent methods (Inglis et al., 2012). The traditional cultivation-dependent approach is based on isolation of Metarhizium colonies from substrates such as soil, plants roots, or host species. From soil, Metarhizium is isolated using either insect baiting with Galleria mellonella (Zimmermann, 1986) or isolation using selective medium agar plates (Inglis et al., 2012; Strasser et al., 1996). Subsequently, Metarhizium colonies are identified at genus level based on morphological characters (Humber, 2012). Fungi of the genus Metarhizium are characterized by their ability to fully cover their insect host. The conidia of Metarhizium spp. are pale to bright green to yellow-green, olivaceouse, sepia or white and on nutrient plates conidia are produced in dense patches (Figure 1.3). The conidiophores are broadly branched and conidiogenous

24

GENERAL INTRODUCTION cells have rounded to conical apices and produce long chains of cylindrical or ovoid conidia. Morphological characteristics do not provide sufficient resolution below genus level, therefore, species identification is performed using molecular tools (Bischoff et al., 2009; Kepler et al., 2014). Most tools to identify Metarhizium species are based on DNA, however, an approach based on protein profiling, i.e. matrix-assisted laser desorption/ionization time-of-flight mass spectrometry (MALDI-TOF MS), has been developed recently (Lopes et al., 2014). DNA based methods for identification of Metarhizium include a variety of methods which are based either on sequence characterized DNA or anonymous DNA sequences (Inglis et al., 2012). They include methods, such as PCR restriction fragment length polymorphism (PCR-RFLP) (Enkerli et al., 2009; Leal et al., 1997; Tiago et al., 2011), simple sequence repeats (SSR) (Enkerli et al., 2005; Oulevey et al., 2009), random amplified polymorphic DNA (RAPD) (Velásquez et al., 2007), amplified fragment length polymorphism (AFLP) (Inglis et al., 2008), group I intron insertion analyses (Marquez et al., 2006) and sequence analyses (Kepler and Rehner, 2013). These different methods enable the discrimination of Metarhizium at different taxonomic level. The most frequently used method for determination of Metarhizium species comprise sequence analyses of tef1a (Castro et al., 2016a; Fisher et al., 2011; Kepler et al., 2015; Nishi et al., 2011; Steinwender et al., 2014; Wyrebek et al., 2011), and SSR analyses for genotyping of Metarhizium isolates (Castro et al., 2016a; Rogge et al., 2017; Steinwender et al., 2015). Sequence analyses have become more efficient by implementing NGS approaches, e.g. seven loci were analysed simultaneously using PacBio RS II sequencing platform (Kepler et al., 2016). For genotyping of the former Metarhizium anisopliae species complex 41 SSR markers are available (Enkerli et al., 2005; Oulevey et al., 2009). These SSR markers were developed from three isolates each affiliated to different species according to the current taxonomy ( M. anisopliae ART2062, M. brunneum ARSEF7524 and M. robertsii ARSEF7532). The among- species-transferability of the 41 SSR markers and the degree of polymorphism of each marker in different species is not known.

Non-target effects of Metarhizium

Commercialisation of a biological control agent requires registration of the product for which relevant data on its safe use has to be compiled. Requirements for registration of chemical and microbial products are published in the Official Journal of the European Union (European Commission, 2011) and in the Swiss legislation, i.e. in the “Verordnung über das Inverkehrbringen von Pflanzenschutzmitteln 916.161” from 12 May 2010. Concerning criteria for authorization of active ingredients (annex 2, paragraph 3), the Swiss regulation refers to the European regulation, which includes the assessment of “non-target species likely to be at risk from exposure to the active substance, its metabolites, degradation and reaction products” (L 155/31-32). Potential effects of Metarhizium on non-target species have been tested for various species from following taxa: mammals (including humans), vertebrates (fish, amphibia, reptiles, birds), plants, aquatic organisms (i.e. shrimp larvae), non-target arthropods (predators, parasitoids, mites, collembolans and honey bees), earth worms and microorganisms (reviewed by Zimmermann, 2007). In most cases no effects of Metarhizium on non-target species was observed, however, some non-target arthropods, e.g. the bumblebee Bombus terrestris (Hokkanen et al., 2003) and ladybug Hippodamia convergens (Ginsberg et al., 2002) were affected by certain Metarhizium strains. Challenges in the assessment of non-target effects include the selection of relevant and exposed non-target species, of appropriate experimental approaches due to the discrepancy between effects measured in the laboratory and the field (Vestergaard et al., 2003) and of the applied form of Metarhizium (conidia vs. culture extract) (Zimmermann, 2007). Metarhizium is often applied in soil and therefore studies on potential non-target effects on soil organisms, such as arthropods, 25

GENERAL INTRODUCTION earthworms and microorganisms, are relevant. Negligible effects of Metarhizium brunneum BIPESCO 5 on population levels of non-target insects in a maize field have been found (Babendreier et al., 2015) and no effects on predatory beetles and spiders (Rauch et al., 2017), as well as no effects on earthworms (Hozzank et al., 2003) have been detected. Studies on effects of microbial products on indigenous soil microbial populations are explicitly required for product registration in the EU and Switzerland (L155/66). Interestingly, the US Environmental Protection Agency (EPA) does not require such test as their evaluation revealed “no significant or lasting impact on ecosystems from introduction of pesticidal microbes even where changes to these populations can be meaningfully tracked” (EPA, 2007). Studies on potential effects of Metarhizium spp. on indigenous soil microorganisms revealed no effects (Hu and St Leger, 2002; Kirchmair et al., 2008). However, in these studies cultivation-dependent approaches were used, which are known for assessing only a small fraction of the soil microbial populations and to be less sensitive than cultivation-independent NGS approaches. NGS sequencing has been used to assess potential effects of B. bassiana on soil fungal communities and no effects were detected in the bulk soil of a chili field seven weeks after application of the fungus (Hirsch et al., 2013) but such a study has not been conducted for Metarhizium spp. Furthermore, up to date studies on potential effects of Metarhizium have assessed only soil fungal communities and studies on soil prokaryotic communities are lacking.

26

GENERAL INTRODUCTION

1.5 Objectives

Soils are densely populated by microorganisms, which fulfil a plethora of different ecosystem functions. These soil microbial communities undergo changes due to different biotic and abiotic factors. A biotic factor potentially affecting microbial communities, is the addition of a microbial strain applied for biocontrol. Assessing this aspect for biological control agents, such as entomopathogenic fungi, is an important safety issue and by law (in form of a regulation) it is required to provide information on potential effects of a microbial biological control agent on soil microbial communities for registration. When performing effect analyses it is essential to ensure that soil microbial communities are exposed to the applied strain and therefore tools to monitor the abundance of the applied strain are required. This thesis has two major goals: A) Improving a DNA based method to identify Metarhizium at strain level B) Assessing potential effects of applied M. brunneum strains on soil microbial communities Potential effects on native soil microbial communities were studied in two pot and one field experiment targeting either Agriotes spp. or D. v. virgifera in potato or maize using M. brunneum ART2825 or M. brunneum EAMa01- Su. The applied strains were monitored using isolation on selective medium followed by genotyping using SSR markers, and soil microbial communities were assessed using amplicon sequencing of ribosomal marker genes. The following objectives were defined for this thesis: 1) Determining the transferability of 41 SSR markers to different Metarhizium spp. and assessing the degree of polymorphism of the markers in different Metarhizium spp. 2) Assessing potential effects of unformulated and formulated M. brunneum ART2825 or M. brunneum EAMa01- Su, efficacy-enhancing agents and insecticides on native soil microbial communities 3) Evaluating the potential of amplicon sequencing to simultaneously assess both the abundance of an applied strain and soil fungal and prokaryotic communities 4) Investigating potential analytical constraints of the presence of one highly abundant sequence on the assessment of changes in soil microbial community structures, due to biases in NGS analyses.

27

GENERAL INTRODUCTION

1.6 Outline of the thesis

This thesis includes 6 chapters:

Chapter I is the general introduction.

Chapter II deals with the transferability of 41 SSR markers to seven different Metarhizium spp. This resulted in the compilation of five sets of three markers each which were evaluated for their applicability for genotyping of different Metarhizium spp. including the applied strains.

Chapter III includes a pot and field experiment to assess potential effects of M. brunneum ART2825 on soil microbial community structures in biological control of Agriotes spp. in potato. M. brunneum was applied in form of FCBK, fungal capsules and spore powder, and it was combined with the efficacy enhancing agent garlic.

Chapter IV reports on a pot experiment in which potential effects of M. brunneum EAMa01-Su formulated as FCBK on soil microbial community structures were assessed in a biological control experiment aimed at controlling D. v. virgifera in maize.

Chapter V addresses the question whether one dominant OTU infers with the assessment of changes in soil microbial communities structures due to biases in NGS.

Chapter VI is the overall discussion including conclusions and perspectives.

28

MULTIPLEXED MICROSATELLITE MARKERS FOR SEVEN METARHIZIUM SPECIES

Multiplexed microsatellite markers for seven Metarhizium species

A modified version of this chapter was published as: Mayerhofer, J., Lutz, A., Widmer, F., Rehner, S. A., Leuchtmann, A., Enkerli, J. (2015). Multiplexed microsatellite markers for seven Metarhizium species. Journal of Invertebrate Pathology 132: 132-134.

2.1 Abstract

Cross-species transferability of 41 previously published simple sequence repeat (SSR) markers was assessed for 11 species of the entomopathogenic fungus Metarhizium . A collection of 65 Metarhizium strains including all 54 used in a recent phylogenetic revision of the genus were characterized. Between 15 and 34 polymorphic SSR markers produced scorable PCR amplicons in seven species, including M. anisopliae , M. brunneum , M. guizhouense , M. lepidiotae , M. majus , M. pingshaense , and M. robertsii . To provide genotyping tools for concurrent analysis of these seven species fifteen markers grouped in five multiplex pools were selected based on high allelic diversity and easy scorability of SSR chromatograms.

2.2 Introduction

Insect pathogenic species of the fungal genus Metarhizium Sorokin (Hypocreales, Clavicipitaceae) are widely used in biological control of arthropod pests (Faria and Wraight, 2007). Recent multilocus phylogenetic analyses of the genus resulted in delineation of a complex of nine species within the species Metarhizium anisopliae (Bischoff et al., 2009). The complex comprises a well-defined inner core, the PARB clade, including M. pingshaense , M. anisopliae , M. robertsii and M. brunneum (Bischoff et al., 2009). Furthermore, it includes a clade consisting of M. majus and M. guizhouense and three additional species, i.e., M. acridum , M. lepidiotae , and M. globosum . In order to further improve our understanding of genotypic diversity and population genetic structures within Metarhizium species, to track introduced strains in the environment, to assess their possible effects on indigenous Metarhizium populations, and / or to characterize cultivars, highly resolving genetic markers are required (Enkerli and Widmer, 2010). Microsatellites, also known as simple sequence repeat (SSR) markers, have proven to be an ideal tool for such purposes (Taylor and Fisher, 2003). Forty-one SSR markers have been isolated from three different Metarhizium strains originally identified as M. anisopliae but now recognized as M. anisopliae (strain ART 2062), M. brunneum (strain ARSEF 7524) and M. robertsii (ARSEF 7532) (Enkerli et al., 2005; Oulevey et al., 2009). Selected subsets of these SSR markers have been used to characterize genotypic diversity of Metarhizium isolates from different environments (Freed et al., 2011; Kepler et al., 2015; Steinwender et al., 2014; Velásquez et al., 2007). The goal of this study was to test the applicability of the 41 SSR markers for multilocus genotyping in different Metarhizium species, taking into account the recent taxonomic refinements in this genus. This information was used to compile SSR marker sets applicable to several different Metarhizium species.

29

MULTIPLEXED MICROSATELLITE MARKERS FOR SEVEN METARHIZIUM SPECIES

2.3 Material and methods

A collection of 65 Metarhizium strains representing all nine Metarhizium species of the M. anisopliae species complex and two species of the M. flavoviride species complex (outgroup) was genotyped using 41 SSR markers ( Table 2.1 and Table Appendix A 1). BLAST searches with reference sequences for these SSR markers (Enkerli et al., 2005; Oulevey et al., 2009) demonstrated their presence and broad distribution in the genomes of M. anisopliae , M. brunneum and M. robertsii (Hu et al. 2014; Pattemore et al. 2014; Table Appendix A 2). Fifty-four of the strains used in this study were included in the most recent revision of Metarhizium (Bischoff et al., 2009) and were obtained from the USDA-ARS Collection of Entomopathogenic Fungal Cultures (ARSEF, Ithaca, NY, USA) or the Centraalbureau voor Schimmelcultures collection (CBS, Utrecht, Netherlands). The remaining eleven strains, among them the frequently used biological control agent M. brunneum strain BIPESCO 5 (also known as F52, M43, ATCC 90448), were obtained from other culture collections (Table Appendix A 1). Species affiliation of eleven strains not included in the study by Bischoff et al. (2009) was verified by sequencing and comparing the 5’ end of elongation factor 1 alpha (EF1alpha) as described by Bischoff et al. (2009). GenBank accession numbers are provided in Table Appendix A 1. Each species was represented by at least 4 strains except for M. frigidum and M. globosum with only one strain each. The strains derived from insects or soils and originated from 29 countries representing all continents except Africa and Antarctica (Bischoff et al. 2009; Table Appendix A 1). Cultures were maintained and fungal mycelia were produced as previously described (Oulevey et al., 2009). Genomic DNA was extracted using NucleoSpin Plant II (Machery & Nagel, Germany) DNA extraction kit. Forty-one SSR primer pairs (Enkerli et al., 2005; Oulevey et al., 2009) were combined in sets of two or three pairs to perform multiplex touchdown polymerase chain reactions (PCR, Table 2.2 and Table Appendix A 3). PCR was conducted in 20 µl reaction volumes containing 10 ng genomic DNA, 0.2 µM of each primer, 0.2 mM dNTPs, 1 x GoTaq ® Flexi

® Reaction buffer, 0.25 U of GoTaq Flexi DNA Polymerase (Promega, WI, USA) and 3 or 4 mM MgCl 2 (Table 2.2 and Table Appendix A 3). One primer of each pair was labelled with NED, HEX or FAM (Applied Biosystems, CA, USA), respectively. Touchdown PCR conditions consisted of 2 min initial denaturation at 94 °C, followed by

12 cycles of 30 sec denaturation at 94 °C, 30 sec annealing at T a + 12 °C, (with 1 °C decrease per cycle) and 40 sec extension at 72 °C followed by n (Table 2.2 and Table Appendix A 3) cycles of 30 sec denaturation at 94 °C,

30 sec annealing at T a and 40 sec extension at 72 °C. PCR was terminated with a final elongation step of 15 min at 72 °C. PCR fragment sizes were analysed using capillary electrophoresis as described previously (Oulevey et al., 2009). For each species and SSR marker the percentages of strains that produced a scorable PCR amplicon were determined. Markers revealing PCR products for ≥ 75 % of the strains of a species from which the marker was not isolated ( M. anisopliae (8 markers), M. robertsii (6 markers) or M. brunneum (27 markers)) were considered as transferable to the respective species (cross-species transferability). Nei’s unbiased genetic diversity ( H = (1 − ∑ p); Nei and Roychoudhury 1974), was calculated for each marker and species.

2.4 Results and discussion

In seven species ( M. anisopliae, M. brunneum, M. guizhouense, M. lepidiotae, M. majus, M. pingshaense , and M. robertsii ;

30

MULTIPLEXED MICROSATELLITE MARKERS FOR SEVEN METARHIZIUM SPECIES

Table 2.1) at least 21 (15 polymorphic) markers were amplified from ≥ 75 % of the strains of a species. For the remaining species only 2 - 12 markers were amplified of which only one marker was polymorphic for M. acridum . For M. globosum and M. frigidum only one strain per species was included, thus results for these two species are considered tentative. The highest percentages of cross-species transferability and the highest numbers of polymorphic markers were obtained for species in the PARB clade ( Table 2.1). All SSR markers isolated from M. anisopliae were transferable to M. brunneum and M. pingshaense , and all markers obtained from M. robertsii were transferable to M. anisopliae and M. pingshaense . Cross-species transferability of SSR markers isolated from M. anisopliae , M. brunneum or M. robertsii was negatively correlated with phylogenetic distance based on sequence analysis of EF1alpha (Spearman ρ = -0.902 to -0.951, N = 10, p < 0.001). Decreasing cross-species transferability with increasing taxonomic distance has also been observed for other fungal genera such as Lobaria and Phytophthora (Devkota et al., 2014; Schoebel et al., 2013). Cluster analyses performed on SSR marker data did not correspond to the multilocus sequence phylogeny of Metarhizium (Bischoff et al., 2009) and no species-specific clustering was obtained (data not shown). The complexity of evolution of SSR and their flanking region, which may lead to convergence in allele sizes among species, render SSR markers as inappropriate for reconstructing phylogenetic relationships (Barthe et al., 2012; Colson and Goldstein, 1999; Goldstein and Pollock, 1997; Orti et al., 1997). Therefore, the use of SSR markers for species identification is limited and species affiliation should be based on DNA sequences, i.e. EF1alpha sequence comparison (Steinwender et al., 2014).

Table 2.1: The number of isolates examined per species, SSR markers revealing PCR products for ≥ 75 % of all strains of a species, polymorphic markers and percentage of cross-species transferable markers. # of markers % of cross-species transferability a Species # of strains amplified polymorphic M. anisopliae M. robertsii M. brunneum M. pingshaense 4 35 28 100 100 77 M. anisopliae 4 34 28 100 b 100 74 M. robertsii 7 33 22 86 100 b 74 M. brunneum 13 37 34 100 83 89 b M. majus 9 25 22 63 50 63 M. guizhouense 9 22 17 50 67 52 M. lepidiotae 4 21 15 50 33 56 M. acridum 9 9 1 25 17 22 M. globosum 1 12 -- c 25 33 30 M. flavoviride 4 2 0 0 0 7 M. frigidum 1 3 -- c 0 17 7 a Percentage of SSR markers isolated from M. anisopliae (8 markers), M. robertsii (6 markers) and M. brunneum (27 markers) that are transferable to a different species. b Percentage of SSR markers isolated form M. anisopliae (8 markers), M. robertsii (6 markers) and M. brunneum revealing PCR products for ≥ 75 % of the strains of the respective species. c Not applicable because only one strain was tested.

Nei’s unbiased genetic diversity (He) ranged from 0.21 to 1 and varied substantially among species and SSR loci tested (Table Appendix A 4). A significant correlation was observed between He and other indices of diversity such as Shannon index (Spearman ρ = 0.98, N = 328, P < 0.001) and evenness (Spearman ρ = 0.99, N = 328, P < 0.001).

31

MULTIPLEXED MICROSATELLITE MARKERS FOR SEVEN METARHIZIUM SPECIES

PCR amplifications with all 41 SSR primer sets revealed single alleles for all strains, except for all M. majus strains, among which two alleles were obtained at one to eleven SSR loci per isolate (Table Appendix A 3 and Table Appendix A 4). Polymorphism in M. majus isolates depended on the locus and the particular strain. These results suggest duplication of the corresponding regions or possibly a diploid genome. Similar observations have been reported in previous studies of M. majus isolates using isozymes (St. Leger et al., 1992) or genome sequence analyses (Hu et al., 2014). To provide robust and generally useful SSR genotyping tools for Metarhizium , polymorphic SSR markers that amplify reliably from as many Metarhizium spp. as possible, multilocus multiplex PCR methods were identified. For this purpose fifteen SSR markers were selected and grouped into five sets including three markers each. Selection was based on four criteria: 1) applicable to M. anisopliae, M. brunneum, M. guizhouense, M. lepidiotae, M. majus, M. pingshaense , and M. robertsii , 2) high within-species diversity, 3) amplification success and 4) easily scorable SSR peak patterns (Table 2.2). The markers were grouped according to matching PCR conditions and different allele size ranges to simplify marker scorability. Additional markers with high-resolution power for individual species can be selected from Table Appendix A 3 and Table Appendix A 4 (Supplementary information).

Table 2.2: Multiplex touchdown PCR conditions (final annealing temperature (Ta, °C), concentration of MgCl 2 (mM) and number of cycles (n) for five SSR sets comprising the most powerful of the 41 SSR markers to discriminate genotypes of seven Metarhizium species ( M. anisopliae, M. brunneum, M. guizhouense, M. lepidiotae, M. majus, M. pingshaense , and M. robertsii ).

a b Set Markers Ta MgCl 2 n I Ma2049 58 4 12/22 I Ma2054 58 4 12/22 I Ma2063 58 4 12/22 II Ma2089 58 4 12/22 II Ma2103 58 4 12/22 II Ma2296 58 4 12/22 III Ma142 56 3 12/22 III Ma2097 56 3 12/22 III Ma2108 56 3 12/22 IV Ma164 56 3 12/22 IV Ma307 56 3 12/22 IV Ma2099 56 3 12/22 V Ma195 50 3 12/30 V Ma327 50 3 12/30 V Ma2287 50 3 12/30 a Markers labeled with 3 or 4 digits were published by Enkerli et al. (2005) or Oulevey et al. (2009), respectively. b Number of cycles used for touchdown PCR / number of cycles used for subsequent PCR at T a.

The currently most efficient approach to genotype Metarhizium isolate collections is to first perform SSR marker analyses using the five marker sets, then to determine species affiliation (based on EF1alpha sequence comparison) of individual multilocus microsatellite genotypes and finally, if further resolution is required, to use additional SSR markers appropriate for the identified species. This approach will be applicable and useful for strain characterization tracking introduced BCA strains in the environment and analyses of population genetic structures of M. anisopliae, M. brunneum, M. guizhouense, M. lepidiotae, M. majus, M. pingshaense , and M. robertsii .

32

MULTIPLEXED MICROSATELLITE MARKERS FOR SEVEN METARHIZIUM SPECIES

2.5 Acknowledgements

The authors are grateful to Hermann Strasser, Bernhard Steinwender and Cezary Tkaczuk for providing Metarhizium strains. This project was conducted in frame of the EU-FP7-project INBIOSOIL (Grant Agreement No. 282767)

33

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Assessing effects of the entomopathogenic fungus Metarhizium

brunneum on soil microbial communities in Agriotes spp. biological

pest control

A modified version of this chapter was published as: Mayerhofer, J., Eckard, S., Hartmann, M., Grabenweger, G., Widmer, F., Leuchtmann, A., Enkerli, J. (2017) Assessing effects of the entomopathogenic fungus Metarhizium brunneum on soil microbial communities in Agriotes spp. biological pest control. FEMS Microbiology Ecology, 93, doi: 10.1093/femsec/fix117.

3.1 Abstract

The release of large quantities of microorganisms to soil for purposes such as pest control or plant growth promotion may affect the indigenous soil microbial communities. In our study, we investigated potential effects of Metarhizium brunneum ART2825 on soil fungi and prokaryota in bulk soil using high-throughput sequencing of ribosomal markers. Different formulations of this strain, and combinations of the fungus with garlic as efficacy- enhancing agent, were tested over 4 months in a pot and a field experiment carried out for biological control of Agriotes spp. in potatoes. A biocontrol effect was observed only in the pot experiment, i.e. the application of FCBK resulted in 77% efficacy. Colony counts combined with genotyping and marker sequence abundance confirmed the successful establishment of the applied strain. Only the formulated applied strain caused small shifts in fungal communities in the pot experiment. Treatment-effects were in the same range as the effects caused by barley kernels, the carrier of the FCBK formulation, and temporal effects. Garlic treatments and time affected prokaryotic communities. In the field experiment, only spatial differences affected fungal and prokaryotic communities. Our findings suggest that M. brunneum may not adversely affect soil microbial communities.

3.2 Introduction

Soil is a complex and dynamic environment providing habitats for a tremendous number and diversity of soil microorganisms (Nannipieri et al., 2003). It has been estimated that one gram of soil may harbor up to 10 10 bacterial and 10 6 fungal cells and thousands of bacterial and fungal species (Bridge and Spooner, 2001; Roesch et al., 2007; Torsvik et al., 1990; Trevors, 2009). Soil microorganisms provide a wealth of functions. They play a central role in nutrient cycling and the formation and maintenance of soil structure, they contribute to plant health and they are involved in the natural regulation of insects, pathogens and weeds (Kennedy, 1999). All together these functions are vital for maintaining productivity in agriculture and it is important to understand which abiotic and biotic factors, including agricultural practices, may adversely affect microbial communities. The potential impacts of a number of factors including time, space and climate on microbial communities have been investigated in various systems (Lauber et al., 2013; O'Brien et al., 2016; Tedersoo et al., 2014). Likewise, effects of edaphic factors or anthropogenic activities, such as land-use, soil compaction and pesticide applications have been studied (Hartmann et al., 2014; Jacobsen and Hjelmsø, 2014; Lauber et al., 2013).

35

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

The ability of microorganisms to regulate insects, pathogens and weeds has been recognized as an important function with potential use in agriculture more than a century ago (Krassilstschik, 1888; Prior, 1996; Zimmermann, 2007). Since then a variety of microorganisms has been identified and commercialized as microbial pesticides also known as biological control agents (BCA; Faria and Wraight, 2007; Lugtenberg, 2015b). Microbial control usually implies application of large amounts of infective propagules of a BCA to soils under treatment. For instance, about 10 12 -10 14 propagules of entomopathogenic fungi are applied per hectare translating into 10 5 conidia per cm 2 of soil (Jaronski, 2010). Such high loads of propagules may have unintended side effects leading to changes in soil microbial community structures. The European Union therefore has included an assessment of potential effects on indigenous soil microorganisms in the registration process of biological pesticides (Commission regulation No. 544/2011). Most studies assessing effects of applied microorganisms on soil microbial communities have revealed only small or transient effects (Kröber et al., 2014; Trabelsi and Mhamdi, 2013; Zimmermann et al., 2016), but little is known about potential effects of the application of entomopathogenic fungi (Hirsch et al., 2013; Hu and St Leger, 2002; Kirchmair et al., 2008; Rai and Singh, 2002; Schwarzenbach et al., 2009). Agriotes spp. Eschscholtz (Elateridae) are major soil dwelling pests in the Holarctic (Kudryavtsev et al., 1993; Vernon et al., 2001) in various crops, such as cereals, different vegetables and potatoes (e.g. Blot and Brunel, 1999; Miles, 1942; Parker, 1994). Control methods have included repeated tillage, crop rotation, pesticide application and biological control with varying degrees of success (reviewed in Ritter and Richter, 2013; Traugott et al., 2015). The progressive banning of chemical insecticides has resulted in an increased focus on biological alternatives for pest control such as the application of entomopathogenic fungi or nematodes (Ritter and Richter, 2013). Studies with the entomopathogenic fungus Metarhizium brunneum ART2825 Petch (Hypocreales: Clavicipitaceae) have shown promising results in laboratory experiments in controlling A. obscurus L. , A. lineatus L. and A. sputator L. (Eckard et al., 2014; Kölliker et al., 2011). Improvements in formulation technologies and application strategies or co-applications with botanicals, chemicals or other BCAs have been shown to increase the efficacy of entomopathogenic fungi to control pest insects (Ansari et al., 2010; Behle et al., 2013; Kabaluk et al., 2015; Paula et al., 2011). Formulations have been developed for entomopathogenic fungi in order to protect their spores during storage and distribution of the products, to enhance persistence in the field, and/or to facilitate the application process (Glare and Moran-Diez, 2016). Metarhizium (Metschn.) Sorokin has been formulated based on grains, e.g. sterile barley kernels (Aregger, 1992), or was produced in form of microsclerotia (Jaronski and Jackson, 2008).

Application strategies including pheromone traps or CO 2 lures have been used to enhance the efficacy of Metarhizium spp. against Agriotes spp. (Brandl et al., 2017; Kabaluk et al., 2015). Also, several natural substances have been tested for controlling Agriotes spp. (Ritter and Richter, 2013). Among those, garlic was shown to repel and reduce movement of A. obscurus larvae, which potentially may enhance the efficacy of M. brunneum by weakening the larvae and making them more susceptible to a fungal infection (Eckard et al., 2017). In this study we investigated whether applications of the fungus M. brunneum ART2825 for controlling A. obscurus in potato production affect soil fungal and prokaryotic communities. The study relies on both an experiment in the greenhouse (pots) and a field experiment using different formulations of the fungus and garlic extract as potential efficacy-enhancing agent. Isolation and cultivation on selective medium, simple sequence repeat (SSR) genotyping, and high-throughput amplicon sequencing of ribosomal markers were used to monitor the applied fungus and observe changes in fungal and prokaryotic community structures over a period of four months.

36

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

3.3 Material and methods

Rearing of Agriotes obscurus larvae

Lab-reared A. obscurus larvae were used for artificial infestation of substrates in the pot experiment. They were reared in a laboratory livestock established by the method of Kölliker et al. (2009). Briefly, A. obscurus adults were collected from the field and placed into pots (ø 30 cm) containing 10–15 L soil rich in humus and were covered with a mesh bag until oviposition. Grass was repeatedly sown into the soil of the pots to guarantee food for the hatched larvae and the pots were kept moist. Five months after establishment, larvae were transferred into a pot containing fresh peat soil with sliced carrots as food source and stored at 10 °C in the dark. Four weeks prior to experiments each larva was placed individually into a cup with moist peat substrate and carrot slices and maintained at 22 °C. Only healthy larvae were selected for subsequent infestation of pots.

Treatments

Nine and five different treatments were applied in six replicates in the pot and the field experiment, respectively. Different treatments and applied doses are listed in Table 3.1. The entomopathogenic fungus Metarhizium brunneum strain ART2825 was either applied as unformulated fungal spore powder (F powd ) or it was formulated as fungus colonized barley kernels (FCBK), as fungal capsules (F cap ) and as fungal granules (F gran ). The FCBK were produced in the laboratory as described by Aregger (1992). 1.3 kg batches of peeled barley kernels were autoclaved twice in plastic culture bags. Subsequently, the barley kernels were inoculated with culture broth of M. brunneum ART2825 in cornsteep medium (diluted to 10 7 spores / ml with water), which had been incubated at 22–24 °C for five days. Following inoculation, the barley kernels were incubated for four to five weeks at 22–24

°C. The Fpowd was produced by FYTOVITA spol s r. o. (Ostrožská Lhota, Czech Republic) using solid-state fermentation and it was also used for the F cap which were formulated by M. Przyklenk (University of Applied

Sciences, Bielefeld, Germany) according to a modified protocol by Humbert et al. (2017). The F cap included 8 x 10 7 spores / g capsules, autoclaved baker’s yeast and calcium alginate. They were formed by dripping M. brunneum spore-alginate solution into a crosslinking solution which induced polymerization and formation of beads. The

Fgran , a prototype produced by e-nema GmbH (Schwentinental, Germany), included the same components as the

Fcap , however, an extruder and a fluid-bed dryer were used to form granules. Garlic capsules were produced by S. Gerike (University of Applied Sciences, Bielefeld, Germany) and consisted of 6 % garlic oil (Neem Biotech Ltd., Abertillery, United Kingdom), calcium alginate, acetic acid, and a chitosan coating. Garlic capsules were applied alone but also in combination with FCBK and F cap in order to study potential synergistic effects of Metarhizium and garlic. The insecticide clothianidin (Insec; Cheyenne®, Philagro, Saint-Didier-au-Mont-d'Or, France) and sterile barley kernels (BK), which represent the carrier material in the FCBK formulation, were used as positive and negative controls, respectively. The pot experiment included the following nine treatments: FCBK, F cap , F powd ,

Gcap , the combinations FCBK+G cap and F cap +G cap , Insec, BK and untreated pots. In the field experiment five treatments were applied: FCBK, F cap , F gran , Insec and untreated plots (Table 3.1).

37

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Table 3.1: Applied doses of the nine and five treatments applied in the pot and field experiment, respectively (n = 6). The amount of fungal spores in the pot and in the field experiment were 1 x 10 14 and 5 x 10 13 spores / ha and clothianidin was applied at a rate of 11 kg / ha. All pots and plots included potato plants and the pest insect. Amount applied (g / pot or field plot) Treatment Pot Field Fungus colonized barley kernels (FCBK) 5.6 270

Fungal capsules (F cap ) 7 240

Fungal granules (F gran ) NI 240

Fungal spore powder (F powd ) 0.11 NI

Garlic capsules (G cap ) 14.4 NI

Gcap & FCBK 5.6 + 14.4 NI

Gcap & F cap 7 + 14.4 NI Barley kernels (BK) 5.6 NI Clothianidin (Insec) 0.06 14 Untreated x x NI … treatment not included

Set-up of the pot experiment

The pot experiment was conducted in a greenhouse at 20–25 °C from April until September 2014. Each of the nine treatments (Table 3.1) was replicated six times resulting in 54 pots which were randomly arranged and kept at the same position during the experiment. Pots had a dimension of 22.5 x 25 x 26 cm and two mesh sealed holes (ø 2.5 cm) at the bottom for water drainage and for preventing the escape of A. obscurus larvae. Soil (3 % humus, 22 % clay, 38 % silt) with a pH of 7.9 was collected from a field at Agroscope research station Reckenholz, Zürich (Switzerland). The field soil was homogenized with a cement mixer and filled into pots four weeks prior to application of the treatments. The pots were kept moist (16 ± 3 % water content, no significant difference among treatments) and weeds were removed by hand prior to application of the treatments. Treatments were applied manually onto the soil surface, then mixed into the upper 15 cm of the soil using a small gardening rake. Subsequently, two pre-sprouted seed potato tubers (Solanum tuberosum L.) of the cultivar ‘Celtiane’ were placed in each pot at a depth of 10 cm followed by the release of ten late instar A. obscurus larvae into each pot. Bulk soil samples were collected from pots before application of treatments and potato tubers on 14 May 2014 (week 0) and post application on 1 July 2014 (week 7) and 26 August 2014 (week 15). Each soil sample consisted of four soil cores (15 cm depth and 1.5 cm width) that were collected cross-wise per pot and then mixed. The above-ground potato tissue was cut 5 cm above the soil surface after the third sampling at week 15, at the time when plants became senescent. Two weeks later, the pots were disassembled. Potatoes were harvested and washed, and the damage caused by A. obscurus , i.e. the number of holes per tuber, assessed and categorized according to standards provided by the European and Mediterranean Plant Protection Organization (EPPO; Anonymous, 2005). Released A. obscurus larvae were re-captured, counted and incubated individually in cups filled with peat substrate and carrot slices as food source at 21 °C for eight weeks to check for infection with Metarhizium spp.

Set-up of the field experiment

The field experiment was performed in an agricultural field located in Mellingen, Switzerland (47°24’24’’ N 8°16’12’’ E). The soil contained 2 % humus, 21 % clay and 32 % silt at soil pH 7.3. The site is naturally infected 38

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL with different wireworm species, predominantly of the genus Agriotes , and was planted with grass during three seasons preceding the experiment. All cultivation and farming steps were performed by the farmer owning the field except soil sampling, potato planting and potato harvesting. The experimental area was rectangular including ten blocks with three plots per block (Figure Appendix B 1). Each plot was approximately three meters wide (four rows of potato plants) and 8.3 m long. The three plots forming a block were connected at the long side and blocks were separated by a 70 cm path. The entire experimental field, including a three meter wide untreated belt surrounding the plots, measured approximately 1600 m 2. Bulk soil samples were collected before application of treatments and potato tubers on 21 April 2015 (week 0) and post application on 24 June 2015 (week 9) and 11 August 2015 (week 16). Soil samples were obtained by collecting and combining ten soil cores (15 cm depth and 2.5 cm diameter) from the inner two rows (five cores from each row) of each plot. One meter buffer zones at both ends of each plot were not sampled to prevent potential carryover from neighboring plots. The field was ploughed in March 2015 and harrowed once after the first soil sampling in April 2015. Then, treatments were applied manually and integrated into the soil by harrowing for a second time. Subsequently, potato tubers of the cultivar “Celtiane” were planted in rows which were piled immediately after application in order to prevent UV exposure of the products. Fifty-five kg per ha of the fertilizer MgS Ammonsalpeter 25 (Agroline, Roggwil, Switzerland, 25 % nitrogen, 5 % magnesium, 8.5 % sulfur) was applied and the herbicide Titus (DuPont de Nemours International Sàrl, Le Grand-Saconnex, Switzerland, 25 % rimsulfuron) + Exell (Stähler Suisse, Zofingen, Switzerland, 77 % detergents, 22 % ethylenglycolmonobuthylether), the pesticide Audienz (Omya AG, Oftringen, Switzerland, 44.2 % spinosad), and the fungicide Mapro (ISK Biosciences GmbH, Bern, Switzerland, 38,8 % fluazinam) was sprayed in May and June. The leaves of potato plants were herbicide treated, after an infection with the fungus Colletotrichum Corda had been detected in July, by applying Reglone (Syngenta AG, Basel, Switzerland, 17 % diquat) for haulm destruction. After the 3 rd soil sampling in August potato tubers of the inner two rows of each plot were harvested. Fifty potato tubers per plot were randomly selected, washed and Agriotes -caused damage scored. Weather data were obtained from the closest meteorological station in Kuenten CH (6 km from the field site). During the sampling period the daily mean temperature was 18.3 °C and ranged between 7.6 and 28 °C. During this time a total of 431.8 mm precipitation was recorded. Average humidity was 73.9 % and ranged between 53.4 and 97.7 %.

Processing of soil samples, isolation of Metarhizium CFU and identification of applied strain

Soil samples were homogenized and sieved with a five mm mesh and aliquots were used for assessment of soil moisture content, for determination and isolation of Metarhizium spp. colony forming units (CFU) and for extraction of soil DNA (described below). The CFU determination of Metarhizium spp. was performed with slight modifications according to the protocol described by (Schneider et al., 2012). Three times 20 g of soil per sample were dissolved in 100 ml pyrophosphate solution and plated onto selective medium agar plates resulting in three plates per sample. Metarhizium colonies were counted after 10 to 14 days. After CFU assessment, isolates were selected from the plates for genetic identification using simple sequence repeat marker (SSR) based genotyping. From the pot experiment, five to six isolates were selected from each treatment at week 0, one to two isolates were obtained from all fungal treatments at week 7 and six isolates were

39

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL chosen from all fungal treatments at week 15. In addition, one to eight isolates per treatment were recovered from Metarhizium spp. infected A. obscurus larvae which were re-captured after the end of the pot experiment and incubated in the lab for detection of late fungal infections. One Metarhizium spp. colony per soil sample (= plot) per sampling time point was selected from the field experiment.

Fungal tissues isolated from FCBK, F cap and F gran were used as positive controls. Isolates were transferred to potato agar plates and stored at 4 °C until all isolates of the pot or the field experiment were collected. Subsequently, all isolates were plated onto filter paper, which was placed on potato agar plates. Mycelium of each isolate was scraped off the filter paper and DNA was extracted according to the protocol described by Kepler et al. (2014). SSR analysis for genotyping of Metarhizium isolates was performed using SSR markers Ma2049, Ma2054 and Ma2063 (Set I) and Ma195, Ma307 and Ma2287 (Set V) (Mayerhofer et al., 2015). DNA extracts were diluted 10 to 100 times and PCR was performed as described in Mayerhofer et al. (2015). PCR products were visualized with an ABI 3130xl (Applied Biosystems, Foster City, California, U.S.A.) using 36 cm capillaries and POP-7TM polymer. GENESCAN TM 400 HD ROX TM was used as an internal size standard. Allele sizes were determined using the software GeneMarker® (SoftGenetics®, State College, Pennsylvania, U.S.A.) and corrected relative to allele sizes of the reference strains M. anisopliae ART2062 (Metschn.) Sorokin, M. brunneum ARSEF7524 and M. robertsii ARSEF7532 J.F. Bisch.

DNA extraction from soil, PCR and Illumina sequencing

Soil genomic DNA was extracted from each replicate (pot or plot) per treatment and per sampling time point for both experiments. One half g of each sample was placed into a 2 ml Eppendorf tube containing 0.5 g of glass beads (ø 0.1 - 0.11 mm; Sartorius, Tagelswangen, Switzerland), vortexed with 1.5 ml extraction buffer and stored at -20 °C until further use. Soil DNA was extracted as described by Bürgmann et al. (2001) and modified by Hartmann et al. (2005). Soil DNA extracts were purified with the NucleoSpin ® gDNA clean-up kit (Machery-Nagel, Düren, Germany) and stored at -20 °C. DNA concentrations were measured using PicoGreen (Invitrogen, Carlsbad, California, U.S.A.) with a Cary Eclipse fluorescence spectrophotometer (Varian, Inc., Palo Alto, California,

U.S.A.) and DNA extracts were diluted to 2 ng / µl with autoclaved dd H 20. PCR was adopted from Frey et al. (2016) with small modifications. Fungal internal transcribed spacer region two (ITS2) was amplified using the primer pair ITS3 (5’ CAHCGATGAAGAACGYRG 3’) / ITS4 (5’ TCCTSCGCTTATTGATATGC 3’) (Tedersoo et al., 2014). The prokaryotic variable region (V3-V4) of the small subunit of the ribosomal RNA (16s rRNA), targeting bacterial and archaeal sequences, was amplified with the modified version of primer pair 341F (5’ CCTAYGGGDBGCWSCAG 3’) / 806R (5’ GGACTACNVGGGTHTCTAAT 3’) (Frey et al., 2016). Forward and reverse primers for amplification of ITS2 and V3-V4 included adapter sequences CS1 (forward) and CS2 (reverse) at the 5’ end of each primer to allow multiplexing with the Fluidigm Access Array System (Fluidigm, South San Francisco, California, U.S.A.). Prior to PCR 20 ng of soil genomic DNA was incubated with 45 µg BSA in 15 µl for 5 min at 90 °C. PCR was performed in a volume of 50 µl containing the pre-incubated DNA, 1x

PCR buffer containing 15 mM MgCl 2 (Qiagen, Venlo, Netherlands), 0.4 µM of the forward and the reverse primer,

0.2 mM of each dNTP (Promega, Madison, Wisconsin, U.S.A.), 1 mM MgCl 2 (Qiagen, Venlo, Netherlands), additional 1.8 mg/ml BSA and 2 U of HotStartTaq ® Plus DNA polymerase (Qiagen, Venlo, Netherlands). PCR cycling conditions included one initial denaturation step at 95 °C for 5 min, followed by 30 or 35 cycles (for prokaryota or fungi) of denaturation at 94 °C for 40 s, annealing at 58 °C for 40 s (both primer pairs) and elongation at 72 °C for 1 min. PCR was finalized with elongation at 72 °C for 10 min. The integrity and quality of the PCR 40

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL products were checked on an agarose gel. PCR was repeated four times per sample, replicates were pooled and sent for sequencing on a Illumina MiSeq platform at the Génome Québec Innovation Center at the McGill University (Montréal, Canada). There, barcodes were added to the PCR products using Fluidigm Access Array technology to allow multiplex sequencing. Subsequently, PCR products were purified with AMPure XP beads (Beckman Coulter, Brea, California, U.S.A.) and pair-end sequencing was performed using Illumina Miseq v3 (Illumina Inc., San Diego, California, U.S.A.). Raw sequences were deposited in the GenBank database with the accession number PRJNA386024.

ITS2 sequence of the applied strain

The sequence of the ITS2 region of Metarhizium brunneum ART2825 was determined with Sanger sequencing using the primer pair ITS3 / ITS4 lacking adapter sequences CS1 and CS2 and the BigDye® Terminator v3.1 cycle sequencing kit (Applied Biosystems, Foster City, California, U.S.A.). Sequences were visualized using a capillary electrophoresis device (ABI 3130xl Genetic Analyzer, Applied Biosystems, Foster City, California, U.S.A.) and assembled using DNA Baser 3.4.5 (Heracle BioSoft, Pitesti, Romania).

Sequence processing and taxonomic classification

Sequences were processed and classified using a customized pipeline (Frey et al., 2016) mostly based on UPARSE within USEARCH v8 (Edgar, 2010; Edgar, 2013). Overlapping paired-end reads were merged using fastq_mergepairs (Edgar and Flyvbjerg, 2015) with a minimal overlap of 50 bp and a minimal merge length of 150 bp for fungal and 300 bp for prokaryotic sequences. Substitution errors were removed using the BayesHammer algorithm implemented in SPAdes 3.5 (Nikolenko et al., 2013b; Nurk et al., 2013) and primers were removed with Cutadapt 1.8.1 allowing one mismatch (Martin, 2011). Quality control was performed using fastq_filter in USEARCH discarding reads with expected total error greater than one (Edgar and Flyvbjerg, 2015). Dereplication and clustering into OTUs with 97 % identity was performed using derep_fulllength and cluster_otus within USEARCH with concurrent removal of singletons and chimera (Edgar, 2013). The eukaryotic or prokaryotic centroids were searched for ribosomal signatures with ITSx (Bengtsson-Palme et al., 2013) or Metaxa2 (Bengtsson-Palme et al., 2015), respectively, and only sequences which included these signatures were kept in the dataset. The algorithm usearch_global was used to map sequences to the centroids (maxdiffs 0, maxaccepts 0, top_hit_only). Eukaryotic sequences were compared to a custom-made NCBI Genbank database (Benson et al., 2015) and the UNITE database (Abarenkov et al., 2010) for taxonomic classification using the naïve Bayesian classifier implemented in MOTHUR v.1.35.1 (Schloss et al., 2009). Sequences that were assigned as metazoa , viridiplantae , protista and unclassified were removed from the dataset. The GREENGENES database (DeSantis et al., 2006; McDonald et al., 2012) was used for taxonomic classification of prokaryota. Subsequently, only archaeal and bacterial sequences were kept in the dataset.

Statistical analyses

The abundance of Metarhizium spp. was assessed by counting CFU and calculating CFU g -1 soil dry weight. Metarhizium CFU g -1 soil dry weight in three replicates per soil sample were averaged using the median per sample. Significance of differences was assessed with a Kruskal Wallis rank sum test (Hollander and Wolfe, 1973) followed by Dunn’s Kruskal Wallis multiple comparison test in the FSA package (Dunn, 1964; Ogle, 2016) with Benjamini-Hochberg (BH) p-value adjustment (Benjamini and Hochberg, 1995) implemented in R version 3.3.0 41

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL used with Rstudio version 0.98.994 (R-Development-Core-Team, 2008; RStudio-Team, 2015). Correlations were calculated using the Pearson correlation coefficient in R. Efficacy of the treatments in the pot experiment was determined by calculating the percent control based on percentage of undamaged potato tubers compared to the control (Abbott, 1987). Saturation of sequencing was checked using intra-sample rarefaction curve analysis (rarefaction.single in MOTHUR) with a re-sampling without replacement approach and plotted in R. Observed OTU richness and the inverse Simpson index representing effective number of species of soil fungal and prokaryotic communities were calculated with “summary.single” in MOTHUR (Jost, 2006; Simpson, 1949). This includes an iterative subsampling procedure (9999 times) to the sampling depth of the sample with the fewest sequences (pot experiment: 7425 fungal and 9088 prokaryotic sequences, field experiment: 2101 fungal and 10 896 prokaryotic sequences). Dissimilarities in the fungal or prokaryotic communities between pairs of samples was assessed using Bray Curtis (BC) dissimilarity matrices with iterative subsampling (9999) which were calculated with dist.shared in MOTHUR. Significance of differences of the fungal and prokaryotic communities among treatments and sampling time points was assessed with overall and pairwise ANOSIM (Spearman rank correlation and 9999 iterations) based on BC dissimilarities implemented in PRIMER v7 (Clarke, 1993; Clarke and Gorley, 2015) and with overall PERMANOVA based on BC dissimilarities using the function adonis within the R package vegan (Oksanen et al., 2016) followed by assessment of pairwise differences using the function pairwise.perm.manova within the R package RVAideMemoire (Hervé, 2017). Unconstrained ordinations were determined in R using nonmetric multidimensional scaling (NMDS) based on BC dissimilarities with the function metaMDS within the R package vegan (Faith et al., 1987; Minchin, 1987; Oksanen et al., 2016). Significant differences of relative sequence abundance of each OTU across sampling time points, treatments and interactions among sampling time points and treatments were assessed with PERMANOVA based on Euclidian distance using the function adonis followed by BH p-value correction. Pairwise differences were calculated for each OTU with a significant overall PERMANOVA pseudo F-statistic per sampling time point using the R function pairwise.perm.manova. In addition, the contribution of single OTUs to BC dissimilarities was calculated using the SIMPER (similarity percentage) routine in Primer 7v (Clarke, 1993) with the 100 most abundant OTUs (relative abundance and square root transformation) per sampling time point. Only significant pairwise comparisons of a treatment and the control assessed with pairwise ANOSIM and pairwise PERMANOVA were selected for the SIMPER analyses.

3.4 Results

Abundance of the applied Metarhizium strain and efficacy of biocontrol treatments in pots

The abundance of Metarhizium spp. increased significantly in all fungus treated pots from a median of 56–144 CFU g -1 soil dry weight before application to 5 569–17 596 CFU g -1 soil dry weight in BCA treated pots at week 7 and remained high until the end of the experiment (Figure 3.1 A). In contrast, the abundance of Metarhizium spp. in pots not treated with the fungus remained low with a median of 0–153 CFU during the entire experiment.

42

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Figure 3.1: Abundance of Metarhizium spp. CFU g -1 soil dry weight per treatment and sampling time in the pot (A) and the field (B) experiment and abundance of the OTU including the sequence of the applied strain per treatment and sampling time in the pot (C) and field experiment (D), * indicates significant differences to untreated control at the corresponding sampling time point (n = 6; p ≤ 0.05).

SSR marker based genotyping revealed that 92 % of the isolates (n = 39) selected from soil of fungus treated pots after applications had the genotype of the applied strain (Table Appendix B 1). The applied strain was already detected in pot substrates before treatment (week 0), but only one out of five isolates from control pots and two out of six isolates from pots treated with F cap +G cap revealed the genotype of the applied strain. The abundance of -1 Metarhizium spp. in these pots was 57 and 141 CFU g soil dry weight. Except for F powd treated pots, significantly fewer A. obscurus larvae were retrieved from fungus treated pots, as compared to the untreated controls at the end of the experiment (Figure 3.2). A median re-capture rate of six A. obscurus larvae out of ten released ones in the untreated pots was within the range of what was observed in previous experiments (unpublished data). The lowest number of A. obscurus larvae was found in the FCBK treated pots with a median of one larva per pot. From 18 mycosed A. obscurus cadavers obtained from pots treated with FCBK, F cap , F powd , FCBK+G cap and F cap +G cap 82.4 % were infected by the applied strain (Table Appendix B 1, for information on treatments see Table 3.1). One A. obscurus larva originating from an Insec treated pot was also infected with the applied strain. Treatments with

Gcap , BK or Insec did not result in decreased numbers of A. obscurus larvae and the combinations of FCBK or F cap with G cap did not enhance efficacy of treatments (no further decrease in the number of A. obscurus larvae). The number of A. obscurus larvae was moderately and significantly correlated with the percentage of damaged potatoes (r = 0.46, p < 0.001). The mean percentage of undamaged potato tubers ranged from 10 to 81 % (Figure 3.2 B). FCBK was the only treatment resulting in a significantly higher number of undamaged potato tubers as compared to the control (Figure 3.2 B) yielding an efficacy (undamaged potatoes compared to the control) of 77 %. There were no significant differences in percentages of low, medium or highly damaged potato tubers among treatments.

The combined treatments of fungus and garlic, G cap , BK and Insec did not exhibit an effect on potato tuber damage.

43

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Figure 3.2: Number of A. obscurus larvae per treatment retrieved from pots initially receiving 10 larvae (A) and levels of potato tubers damage in percentage of total number of tubers harvested per treatment (B): no (0 holes per tuber; SE = 7 – 13 %), low (1 - 2 holes per tuber; SE = 4 - 10 %), medium (3 – 5 holes per tuber; SE = 0 – 10 %), high ( > 4 holes per tuber; SE = 4 – 14 %), n = 2 to 10 potato tubers per pot, * indicates significant differences to untreated control (p ≤ 0.05). Abundance of the applied Metarhizium strain and efficacy of biocontrol treatments in the field

Abundance of Metarhizium spp. increased after the application of FCBK from a median of 1304 to 6969 CFU g -1 soil dry weight nine weeks after application and slightly decreased to 4261 CFU g -1 soil dry weight at week 16

(Figure 3.1 B). The application of F gran or F cap did not yield significantly increased Metarhizium abundances. The applied strain was not detected in any of the treatments at week 0 as shown by SSR based genotyping (Table Appendix B 1). In all fungus treated field plots 36 % of the isolates (n = 36) had the genotype of the applied strain, however, after the application of FCBK 91.7 % of the isolates (n = 12) were identified as the applied strain, whereas, after the application of F cap and F gran only 0 and 16.7 % of the isolates (n = 12) had the genotype of the

44

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL applied strain. Of the 50 potato tubers analyzed per plot, a median of 42 to 51 % was damaged and the number of damaged potato tubers did not differ among treatments (Figure Appendix B 2).

Soil microbial communities of the pots

After quality filtering a mean of 22 406 ± 8505 fungal and 19 706 ± 3418 prokaryotic sequences per sample was obtained in the pot experiment and clustered into a mean of 433 ± 51 fungal and 2795 ± 245 prokaryotic OTUs per sample, respectively. Rarefaction curve analysis revealed that fungi were sampled more exhaustively but with a higher variation among samples than prokaryota (Figure Appendix B 3 A, B). The fungal community across all pots was dominated by Ascomycota (84 %) followed by Zygomycota (7 %), Basidiomycota (6 %), Chytridiomycota (1 %), Glomeromycota (0.4 %), and Blastocladiomycota (0.004 %), besides unclassified fungi (0.01 %; Figure Appendix B 3 C). A total of 45 bacterial and 3 archaeal phyla were detected across all pots. The most abundant phyla (> 10 %) were Proteobacteria (22 %), Actinobacteria (20 %), Chloroflexi (14 %) and Acidobacteria (13 %; Figure Appendix B 3 D).

Abundance of the applied strain and effects of treatments on soil microorganisms in pots

Within the fungal sequence dataset of the pot experiment, two OTUs were assigned to the genus Metarhizium . OTU 3 and OTU 1703 were classified as Metarhizium brunneum and Metarhizium flavoviride var . flavoviride W. Gams & Rozsypal with a sequence abundance of 87 207 and 14, respectively. OTU 3 occurred in 158 of the 162 samples and included 2641 unique sequences. The most abundant unique sequence (33 693 sequences) exactly matched the ITS2 sequence of the applied strain (Genbank Acc. N. KY786031). The abundance of OTU 3 was significantly increased in all M. brunneum treated pots at week 7 and 15 and this increase correlated with the increase of Metarhizium spp. CFU g -1 soil dry weight (r = 0.65, n = 162, p < 0.001; Figure 3.1 A, C). OTU 3 was removed from the fungal dataset in order to avoid analytical bias of the abundance of OTU 3 on statistical tests used to assess treatment-effects on the community structure of soil fungi. OTU richness of the fungal communities did not differ among treatments at week 0 and at week 15. However, at week 7 OTU richness was significantly lower in BK and the FCBK treated pots as compared to the untreated pots at week 7 (Figure Appendix B 4 A). No significant differences in OTU richness were observed among prokaryotic communities of the different treatments compared to untreated pots at the respective sampling time points (Figure Appendix B 4 B). Similar results were obtained using the inverse Simpson index (data not shown). Overall ANOSIM analyses based on BC dissimilarities of the fungal communities across all pots revealed no differences among treatments at week 0, however, small but significant differences were detected at weeks 7 and 15 (Table 3.2). Pairwise ANOSIM tests of treatments compared to untreated pots and the NMDS analyses revealed that the fungal communities were moderately affected (R > 0.4) by the addition of BK, FCBK, F cap at week 7 (Table Appendix B 2, Figure 3.3 A). Also, the fungal communities in these three treatment groups differed among each other (mean pairwise ANOSIM R-value of 0.62 ± 0.09) at week 7 (Table Appendix B 2). At week 15 the fungal communities in pots treated with BK, FCBK, FCBK+G cap , F cap and Fcap +G cap differed significantly from the untreated pots and among each other (mean pairwise ANOSIM R-value of 0.41 ± 0.09; Table Appendix B 2). While prokaryotic communities in the pots did not differ among the treatments at week 0, small changes were detected at week 7 and week 15 (Table 3.2). Pairwise comparisons of the prokaryotic communities of different treatments compared to untreated pots at the respective sampling time point and the NMDS plot revealed that all treatments including garlic (G cap , FCBK+G cap , F cap +G cap ) affected the prokaryotic communities at week 7 and 15 45

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

(Table Appendix B 2, Figure 3.3 B). However, there were no differences in pairwise ANOSIM comparisons among the three garlic treatments (Table Appendix B 2). Corresponding results were obtained for overall and pairwise analyses of fungal and prokaryotic communities using PERMANOVA (Table Appendix B 3).

Figure 3.3: Unconstrained ordination of soil samples based on Bray Curtis dissimilarities of fungal communities per treatment at week 7 in the pot experiment (A, stress = 0.14), of prokaryotic communities for treatments with and without garlic at week 7 and 15 in the pot experiment (B, stress = 0.2), of fungal (C, stress = 0.11) and prokaryotic (D, stress = 0.13) communities in the untreated pots at different sampling time points and of fungal (E, stress = 0.2) and prokaryotic (F, stress = 0.09) communities along a spatial gradient across the long side of the field.

Assessing differences in relative sequence abundance of each fungal OTU among treatments revealed only 0.2 % (7) of the fungal OTUs with a significant overall PERMANOVA pseudo F-value for the factor treatment. The relative abundance of five of these seven fungal OTUs changed significantly between untreated and either BK,

FCBK, F cap , FCBK+G cap or F cap +G cap assessed with pairwise PERMANOVA (Figure 3.4). Similarity percentage analyses (SIMPER) based on BC dissimilarities were performed to identify fungal and prokaryotic OTUs 46

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL contributing to differences of microbial community structures of single treatments and untreated control pots only for significant pairwise comparisons of the microbial communities assessed with ANOSIM and PERMANOVA (Table Appendix B 4, Table Appendix B 5). Data from PERMANOVA and SIMPER analyses revealed that fungal OTU 1, which was classified as member of the family Bionectriaceae , increased significantly in the FCBK treated pots at week 7 and in the FCBK and FCBK+G cap treated pots at week 15 (Figure 3.4) and contributed 12.3 % and 10.6 % to the differences between FCBK treated pots and untreated pots at week 7 and 15, and 2 % to the differences between FCBK+G cap and untreated pots at week 15 (Table Appendix B 4). Fungal OTU 11, classified as Rhizopus oryzae , increased significantly in the BK treated pots at week 7 and in the BK and F cap +G cap treated pots at week 15 (Figure 3.4) and accounted for 12 % and 9.6 % of the differences between BK treated and untreated pots at week 7 and week 15 and 0.8 % of the differences between F cap +G cap treated and untreated pots at week 15.

Fungal OTU 13, which was identified as member of the family Nectriaceae increased significantly in FCBK+G cap treated pots at week 15 (Figure 3.4) and contributed 2.8 % to the differences between FCBK+Gcap and untreated pots at week 15. Fungal OTU 45, identified as Mortierella spp., increased significantly in F cap treated pots at week

7 (Figure 3.4) and contributed 1.5 % to the differences between F cap treated and untreated pots at week 7 (Table

Appendix B 4). The unclassified fungal OTU 291 increased significantly in FCBK, FCBK+G cap and G cap treated pots at week 7 (Figure 3.4), however it was not among the 100 most abundant OTUs which were used for the SIMPER analyses.

Table 3.2: Differences in the fungal and the prokaryotic community structures among treatments at different sampling time points (n = 6), treatments with (n = 18), of untreated pots or plots (n = 6) over time and among blocks across the long side of the field (n = 9) in the pot and / or field experiment assessed with overall ANOSIM (analysis of similarity) based on Bray Curtis dissimilarity. Fungi Prokaryota Experiment Overall test ANOSIM R ANOSIM R Pot Among treatments at week 0 0.03 0.08** Pot Among treatments at week 7 0.31*** 0.38*** Pot Among treatments at week 15 0.26*** 0.29*** Pot Untreated over time 0.22** 0.63*** Field Among treatments at week 0 -0.1 -0.09 Field Among treatments at week 9 0.04 -0.05 Field Among treatments at week 16 0.02 -0.08 Field Untreated over time 0.12* -0.04 Field Among 10 blocks along the field 0.49*** 0.61*** * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001 NA … not assessed

Overall PERMANOVA of relative sequence abundance per OTU revealed that 0.46 % (44) of the prokaryotic OTUs were significantly affected by treatments (data not shown) and of these 36 were significantly different between any treatment and untreated pots. None of these 36 OTUs were significantly different between untreated and FCBK or F powd treated pots, two changed significantly in pots treated with F cap , one changed significantly after

47

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL the addition of Insec and 33 were significantly different between untreated and any treatment including garlic (Figure Appendix B 5). Only ten of the OTUs detected with PERMANOVA were among the 100 most abundant OTUs investigated with SIMPER analyses and they contributed between 0.4 % and 5.41 % to the respective differences (Table Appendix B 5).

Changes of the microbial communities over time in pots

OTU richness of fungal communities in the untreated pots did not change over time. In contrast, OTU richness of soil prokaryotic communities in the untreated pots increased significantly and continuously from week 0 to week 15 (Figure Appendix B 4 A, B). Similarly, a significant increase of prokaryotic OTU richness was observed in

BK, F cap +G cap , FCBK+G cap , F powd and Insec treated pots and it tended to increase also in all other treatments (F cap ,

FCBK and G cap ). Overall ANOSIM values, overall PERMANOVA and NMDS revealed that fungal and prokaryotic community structures of the untreated pots differed among the sampling time points (Table 3.2, Figure 3.3 C, D, Table Appendix B 3). Fungal community structures in the untreated pots changed slightly but significantly between week 0 and 15 (pairwise ANOSIM R = 0.45, p = 0.006) and similar changes over time were observed in FCBK, BK, F cap and F cap +G cap treated pots (mean pairwise ANOSIM R-value of 0.42 ± 0.13; Table Appendix B 2). The prokaryotic communities of the untreated pots underwent a continuous significant shift across the three sampling time points which was shown by NMDS (Figure 3.3 D) and pairwise ANOSIM comparisons of week 0 to 7 and 0 to 15 resulting in R-values of 0.65 and 0.43, respectively (Table Appendix B 2). Corresponding significant changes over time of the prokaryotic community structures were observed in all treated pots (mean pairwise ANOSIM R-values of 0.63 ± 0.16 week 0 to 7 and 0.4 ± 0.14 week 7 to 15; Table Appendix B 2). Assessing differences in relative sequence abundance of fungal and prokaryotic OTUs showed that 99 fungal and 776 prokaryotic OTUs were significantly affected by time (data not shown).

Soil microbial communities in the field

A mean of 19 610 ± 12 252 fungal sequences per sample were obtained for 89 field samples (excluding one sample with a sequence abundance of only 360) and clustered into a mean of 435 ± 98 OTUs per sample. The 90 field samples included a mean of 17 322 ± 2437 prokaryotic sequences which were clustered into 1767 ± 132 OTUs per sample. Rarefaction analyses revealed that sampling the fungal diversity was closer to saturation than the prokaryotic sampling, however, variation was lower among prokaryotic samples (Figure Appendix B 6 A, B). The following six fungal phyla were detected in descending abundance in the soil of the field experiment: Ascomycota (79 %), Basidiomycota (11 %), Zygomycota (4 %), Chytridiomycota (1 %), Glomeromycota (0.7 %) and Blastocladiomycota (0.03 %) with 0.2 % unclassified fungal sequences (Figure Appendix B 6 C). Forty-five bacterial phyla were detected across the field samples. Bacterial phyla with an abundance of at least 10 % comprised Proteobacteria (23 %), Actinobacteria (17 %), Chloroflexi (11 %), Verrucomicrobia (11 %) and Planctomycetes (11 %; Figure Appendix B 6 D). The archaeal phylum Crenarchaeota (3 %) was the only one of three archaeal phyla representing more than 1 % prokaryotic sequences.

48

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Abundance of the applied strain and effects of treatments on microbial communities in the field

Three OTUs were classified as Metarhizium within the fungal sequence dataset of the field samples. OTU 1 (including 73 521 sequences), OTU 2930 (including 3 sequences) and OTU 2732 (including 4 sequences) were assigned to M. brunneum, M. anisopliae and Metarhizium spp., respectively. OTU 1 included 4871 unique sequences and the unique sequence which exactly matched the ITS2 region of the applied strain was detected 6735 times (data not shown). The relative abundance of OTU 1 was significantly higher in FCBK treated field plots 9 and 16 weeks after the treatment (Figure 3.1 D). None of the other treatments resulted in increased OTU 1 abundance. There was a positive correlation between the relative abundance of OTU 1 and the number of Metarhizium spp. CFU g -1 soil dry weight (r = 0.66, n = 90, p < 0.001). OTU 1 was deleted from the fungal dataset in order to avoid analytical bias on statistical tests when assessing changes in fungal communities. There were no significant differences in OTU richness or the inverse Simpson index of the fungal and prokaryotic communities among treatments at different sampling time point (data not shown). Overall ANOSIM and pairwise PERMANOVA showed that neither fungal nor prokaryotic communities in the field were affected by the treatments compared to the untreated plots at the respective sampling time points (Table 3.2,Table Appendix B 6).

Changes of the microbial communities over time and space in the field

Fungal as well as prokaryotic communities in the untreated plots did not differ in their OTU richness and inverse Simpson index (data not shown) and in their community structures based on BC dissimilarities over time assessed with ANOSIM (Table 3.2). However, community structure analyses based on BC dissimilarities assessed with overall PERMANOVA, revealed a significant time-effect on fungal and prokaryotic communities (Table Appendix B 6). ANOSIM as well as PERMANOVA analyses revealed significant spatial effects (Table 3.2, Table Appendix B 6). Fungal and prokaryotic communities both changed gradually from one end to the middle of the field (about 45 m) and became similar again towards the other end of the field (Figure 3.3 E, F). Fungal OTU richness differed significantly between the middle (36 and 45 m) and the end of the field (72 and 81 m, Figure Appendix B 4 C). The community structure (based on BC dissimilarities and visualized by NMDS) of fungal communities differed among blocks (including 3 plots each) along the long side of the field, i.e. among blocks from the middle section of the field compared to blocks from both ends (Table 3.2, Figure 3.3 E, Table Appendix B 2). Corresponding spatial changes were also detected for the prokaryotic community structures (Table 3.2, Figure 3.3 F, Figure Appendix B 4 D, Table Appendix B 2).

49

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Figure 3.4: Relative sequence abundance of fungal OTUs among different treatments and time points (n = 6). * indicates a significant difference between treatments and untreated pots at the respective sampling time point (p < 0.05). OTU 1, OTU 11, OTU 13 and OTU 45 were classified as Bionectriaceae, Rhizopus oryzae , Nectriaceae and Mortierella spp., respectively. OTU 291 was an unclassified fungal OTU.

50

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

3.5 Discussion

Risk assessment of any environmental hazard, i.e. an agent or activity causing a hazard, includes the assessment of exposure to the hazard and effects on the population or individual exposed to the hazard (Brown 1985; U.S. Interagency Staff Group on Carcinogenesis 1986). In this study, exposure was defined as a significant increase of M. brunneum ART2825 abundance. Exposure analysis was performed with a cultivation-dependent approach (i.e. determination of Metarhizium spp. CFU followed by identification of the genotype of the applied strain) and with a cultivation-independent approach (i.e. assessment of the OTU of the applied strain within the amplicon sequences). With both approaches, significant exposure to the applied fungal strain was demonstrated, both in the pot experiment and in FCBK treated field plots. Isolates of the genotype of M. brunneum ART2825 were detected at low frequency (6 %) in pots before application (Untreated and F cap +G cap ) and isolated from a larvae from an Insec treated pot and very likely represent natural occurrence of the strain, since the soil used in the pot experiment originated from a field at Agroscope Reckenholz where M. brunneum ART2825 has originally been isolated from an A. obscurus larva (Eckard et al., 2014; Kölliker et al., 2011). Although the applied strain established in all fungal treated pots and in FCBK treated plots the biocontrol effect was limited. Only the application of FCBK lead to a 77 % efficacy (increase of undamaged potato tubers compared to the control) and a significant reduction of A. obscurus larvae in the pot experiment, which corroborated previous laboratory experiments (Eckard et al., 2014;

Kölliker et al., 2011). The number of A. obscurus larvae was significantly reduced in F cap , F+G cap and FCBK+G cap treated pots compared to untreated pots, however, this did not result in reduced potato tuber damage. The inconsistent results of potato tuber damage and number of Agriotes larvae might result from feeding interruptions prior and post molting, which may be uncoordinated within a population (Furlan, 1998; Furlan, 2004; Sufyan et al., 2014) and differences in foraging behavior of A. obscurus larvae possibly due to different volatile organic compounds (reviewed in Barsics et al., 2014) emitted from treatments. In contrast to the pot experiment, no biocontrol success was achieved in the field in any of the treatments within one season of fungal applications. This might be explained by unfavorable conditions for the fungus possibly created by non-optimal soil moisture, soil texture, soil temperature or antagonistic microbes (Jaronski, 2007). In addition, the applied strain may not be able to provide sufficient protection against all Agriotes species present in the field, that have been shown to be difficult to control (Blackshaw and Vernon, 2008; Sufyan et al., 2013; Sufyan et al., 2014). In other field studies using Metarhizium spp. to control Agriotes larvae varying degrees of success have been reported (Kabaluk et al., 2005; Ritter et al., 2011). For instance, M. brunneum ART2825 formulated as FCBK was applied to protect lettuce from A. sputator and A. ustulatus and showed 21 and 65 % reduction of the two pest insects, respectively (Ritter et al., 2011). However, insignificant reduction in potato tuber damage was detected after the application of M. anisopliae granules (Kabaluk et al., 2005). In the pot trial of the present study the combined treatments of M. brunneum

ART2825 and garlic capsules (FCBK+G cap , F cap +G cap ) did not enhance efficacy, an observation which was also made in a laboratory experiment using two-dimensional terraria (Eckard et al., 2017). One reason for insufficient control of Agriotes larvae may be a repelling effect of Metarhizium spp. on Agriotes spp. (Kabaluk et al., 2005).

The use of attractants such as CO 2 emitting capsules or pheromone pitfalls, as tested in other studies, may help to overcome possible repelling effects (Brandl et al., 2017; Kabaluk et al., 2015). The application of the insecticide clothianidin was neither successful in the pot experiment nor in the field experiment. This is in accordance with results obtained from bioassay experiments, where A. obscurus larvae have been exposed to clothianidin treated wheat seedlings (van Herk et al., 2008a). In this bioassay over 70 % of the larvae were moribund following a 51

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL similar insecticide treatment but most recovered 14 days after application. However, even though efficacy of the treatments was limited in our study, criteria for exposure where nevertheless achieved and allowed an assessment of effects of M. brunneum ART2825 on soil microorganisms in the pot and in the field experiment.

Application of M. brunneum ART2825 formulated as FCBK and F cap (but not the application of fungal spore powder alone) resulted in slight changes of the fungal communities in the pot trial, suggesting that the observed effects on microbial communities were caused by compounds of the formulations rather than by the fungus itself. These small changes between untreated pots and FCBK treated pots were reflected in a significant increase of only two OTUs which were classified as a member of Nectriaceae and an unclassified fungus. The taxonomic classification of these OUTs allowed very limited assumptions of their functions and possible interactions with the applied strain. The changes between untreated pots and pots treated with F cap were reflected in a significant increase of only one OTU classified as Mortierella spp. which are known for their saprophytic life style. This fungus may profit from the alginate carrier, however, it increased with a strong variation as observed in most treatments and sampling time points. In a similar pot experiment, aimed at controlling Diabrotica v. virgifera LeConte, the application of FCBK and similar fungal capsules did not affect the fungal communities (J. Mayerhofer unpublished). This may suggest that the impact of FCBK and F cap application on fungal communities is context dependent involving also soil specific or environmental factors. Application of FCBK had also no effect on fungal and prokaryotic communities in the field and all fungal treatments had no effect on prokaryotic communities in the pot experiment. Our results are in agreement with other studies involving entomopathogenic fungi, e.g. Metarhizium anisopliae or Beauveria bassiana (Bals.-Criv.) Vuill., which detected no, small or only transient effects on soil microorganisms (Hirsch et al., 2013; Hu and St Leger, 2002; Kirchmair et al., 2008; Rai and Singh, 2002; Schwarzenbach et al., 2009). Likewise, the release of other microorganisms for control of phytopathogens, weeds or nematodes resulted in small or transient effects on soil microbial communities (Grosch et al., 2006; Rousidou et al., 2013; Zimmermann et al., 2016). Furthermore, microorganisms released as biofertilizers, phytostimulators or plant growth promotors had no effects on bacterial communities in the rhizosphere (García de Salamone et al., 2010; Kröber et al., 2014; Lerner et al., 2006) or only moderate effects on bacteria and fungi in the rhizosphere or the bulk soil (Schmidt et al., 2012; Trabelsi et al., 2011; van Dillewijn et al., 2002). The untreated control pots and field plots allowed to assess changes of the microbial communities over time including seasonal and environmental changes, and effects caused by the plants on the resident microbial soil communities. In our study, time-related effects on the fungal and the prokaryotic soil communities in the pot experiment were similar or greater than the treatment-effects. This is in agreement with other studies showing that seasonal changes of soil microbial composition in relation to developmental stage of the plant exceed treatment- effects of applied fungi and bacteria in bulk soil (Savazzini et al., 2009) as well as in the rhizosphere in the field (Grosch et al., 2006; van Dillewijn et al., 2002; Zimmermann et al., 2016). In the field experiment of our study the assessment of temporal changes using ANOSIM and overall PERMANOVA showed contradicting results which may have resulted from different sensitivities of the tests. However, the fungal and prokaryotic community structures varied spatially across the field. We suspect that this variation may be related to differences in edaphic factors across the field. Humus, clay and silt content as well as soil pH were assessed, but results did not yield sufficient resolution to support this hypothesis. In other studies soil edaphic factors including pH, organic carbon, texture, soil moisture and land management have been shown to influence soil microorganism at the agricultural plot scale (Chen et al., 2007; Naveed et al., 2016; Philippot et al., 2009; Rousk et al., 2010).

52

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

Entomopathogenic fungi are formulated for applications in order to increase persistence, efficacy or shelf-life of the fungi (Burges, 1998). M. brunneum ART2825 was applied in form of FCBK, F cap and F powd in the pot experiment. The addition of BK, the non-fungal component of the FCBK, also affected fungal communities. These effects were mainly due to an increased abundance of Rhizopus oryzae, a well-known degrader of organic matter. The increase of the relative sequence abundance of Rhizopus oryzae varied among replicates as shown by a large dispersion which may indicate that the response of soil microbial communities was pot specific over time and may indicate the introduction of responsive microorganism by the addition of potato tubers. Surprisingly, Rhizopus oryzae was not enhanced in the FCBK treated pots. Possibly because the niche “BK” was already occupied by the applied strain preventing R. oryzae to colonize this nutrient source. The results of the present study suggest that the formulations may have been responsible for these effects. Similarly, the effects caused by a biological nematicide containing the fungus Paecilomyces lilacinus (Thom) Samson formulated with glucose and skimmed milk were triggered by the formulation only (Rousidou et al., 2013). The prokaryotic communities reacted to the application of G cap and the combinations of FCBK+G cap and F cap +G cap in a very similar way but not after application of fungal products only, suggesting that the observed effects were due to the application of G cap . The effects of

Gcap on soil prokaryotes resulted either from garlic oil, parts of the formulation (not studied separately) or the combination of both. Garlic has been used traditionally as an antimicrobial agent in medicine and for human consumption, but more recently also to protect plants against soil-borne fungal and bacterial diseases (Curtis et al., 2004; Lawson, 1998; Sealy et al., 2007). A possible mechanism explaining the effect of garlic may be interference with quorum sensing, a common regulatory process between bacterial cells coupling gene expression to cell density as suggested by others (Bodini et al., 2009; Dessaux et al., 2011; Gonzalez and Keshavan, 2006). The systemic neonicotinoid clothianidin used in our study did not affect the fungal and the prokaryotic soil community structures, both in the pot and in the field experiment. However, the concentration of clothianidin was not monitored and therefore exposure to the compound was not confirmed. It is possible that the insecticide has been partially or completely degraded before effects became manifest although half-life of this chemical in soil is supposed to range between 20 and 1000 weeks (Simon-Delso et al., 2015). Pesticide degrading microorganisms can reduce the clothianidin concentration as it is degradable aerobically and anaerobically by microbes (Mulligan et al., 2016). Moreover, studies with other systemic neonicotinoids have documented effects on soil fungal and bacterial communities, confirming that clothianidin can potentially have adverse effects on microbial communities (Cai et al., 2016; Cai et al., 2015; Singh and Singh, 2005; Zaller et al., 2016). The release of microorganisms to soil for pest control offers great potential and benefits for agriculture. Particularly, entomopathogenic fungi provide an alternative to chemical pesticides or may allow to reduce application of such chemicals and their release to the environment. Registration of entomopathogenic fungi for pest control requires knowledge on possible effects on soil microbial communities. The present study showed that

Metarhizium brunneum ART2825 formulated as FCBK and F cap , in contrast to the application of fungal spores only, can cause small changes in fungal communities. However, changes were in the same range or even smaller than changes caused by BK (the non-fungal compound of the formulation FCBK), or natural fluctuations in community structures. Amplicon sequencing proved to be a powerful tool for simultaneously assessing exposure to the released strain and effects on the community structure of soil microorganisms. Future investigation should focus on specific functional groups (such as Rhizobia, or mycorrhizal fungi) or use meta-proteomics or transcriptomics approaches to assess possible effects at the functional level. This may provide complementary knowledge on effects of BCAs on microbial communities. 53

ASSESSING EFFECTS OF THE ENTOMOPATHOGENIC FUNGUS METARHIZIUM BRUNNEUM ON SOIL MICROBIAL COMMUNITIES IN AGRIOTES SPP . BIOLOGICAL PEST CONTROL

3.6 Funding

This study was funded by the means of the 7 th Framework Programme of the European Union in frame of the project “Innovative Biological products for Soil pest control” [INBIOSOIL, Grant Agreement No. 282767].

3.7 Acknowledgments

We would like to thank Christian Schweizer and Christian Gees for support in sampling and sample preparation. The authors declare no conflict of interest.

54

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

Response of soil microbial communities to the application of a

formulated Metarhizium brunneum biocontrol strain

A manuscript based on this chapter is submitted for publication as: Mayerhofer, J., Rauch, H., Hartmann, M., Widmer, F., Gschwend, F., Strasser, H., Leuchtmann, A., Enkerli, J. (2017) Response of soil microbial communities to the application of a formulated Metarhizium brunneum biocontrol strain.Plos ONE.

4.1 Abstract

Entomopathogenic fungi are used for biological control of insect pests. Metarhizium brunneum Petch (Hypocreales) has potential to control Diabrotica virgifera virgifera LeConte (Chrysomelidae), which is major pest of maize in North America and has recently invaded Europe. The inundative application of an entomopathogenic fungal strain in biological control results in high densities of fungal propagules in the soil which can potentially affect soil microbial communities and their multiple functions in soil. The objective of the present study was to assess potential effects of M. brunneum on soil fungal and prokaryotic communities in a pot experiment over a time course of 4 months using high-throughput sequencing (HTS) of ribosomal markers. The application of M. brunneum formulated as fungus colonized barley kernels (FCBK) led to a significant increase of the applied strain in soil, as assessed by cultivation-dependent (plating on selective medium followed by genotyping of Metarhizium isolates) and cultivation-independent (HTS of ribosomal markers) approaches. Data revealed that soil fungal and prokaryotic community structures did not change after the application of M. brunneum . Temporal changes of the fungal and prokaryotic communities were observed and the prokaryotic communities showed minor changes to barley kernels (BK), the matrix of the formulation. Results of this study are in accordance with other investigations lacking any evidence for adverse effects on microbial communities caused by applied entomopathogenic fungi.

4.2 Introduction

Diabrotica virgifera virgifera LeCont (Chrysomelidae) is a devastating univoltine pest in maize ( Zea mays L.) in North America. Since the late 1990s it has also established in south-eastern Europe and northern Italy with potential to invade parts of Asia (Bača, 1994; Ciosi et al., 2008; Krysan and Miller, 1986), which is expected to cause large damage costs (Wesseler and Fall, 2010). The larvae damage maize plants by feeding on the roots, which leads to plant lodging, difficulties during plant harvest and decrease in yield. Adults feed on pollen, silk, kernels, and foliage and as a result maize ears lack kernels and are more prone to fungal infection (Chiang, 1973). Available control options comprise crop rotation, chemical and biological insecticides, and resistant plants including trans-genetically modified varieties. These measures have been implemented with different success (Toepfer et al., 2009; van Rozen and Ester, 2010). The most frequently used methods are crop rotations and insecticide applications, however, approaches suffer from the development of adapted or resistant D. v. virgifera populations (Ciosi et al., 2009; Levine et al., 2002). Maize breeding has yielded tolerant varieties, which are

55

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN characterized by massive and regenerating root systems tolerating larval feeding (Branson et al., 1982). Trans- genetically modified maize plants expressing toxins of Bacillus thuringiensis against D. v. virgifera are available but resistance to most available toxins have been observed (Jakka et al., 2016). Biological control agents such as entomopathogenic nematodes, e.g., Heterorhabditis bacteriophora , and fungi, e.g., Metarhizium spp. and Beauveria spp., have shown promising results for control of D. v. virgifera (Toepfer et al., 2010) . They bear potential for application in integrated pest management as recommended by European authorities (Directive 2009/128/EC, L309/71). The entomopathogenic fungus Metarhizium brunneum, formerly known as M. anisopliae , reduced 31 % of emerging D. v. virgifera adults in a field study (Pilz et al., 2009). In another study it has not been successful, probably because of the high pest pressure in the field (Rauch et al., 2017). The development of a fungal biological control agent (BCA) not only requires identification of an effective and suitable strain, it also includes establishment of large-scale production, efficacy-enhancing formulation, and sustainable application, as well as assessment of quality control and monitoring tools for potential effects on non- target macro- and microorganisms. The latter issue has attracted increasing attention particularly as it has become a major requirement in the product registration process (e.g. Commission Regulation (EU) No 544/2011, L155/66). Investigation of possible effects on non-target macroorganisms is frequently performed in parallel with studies on host-specificity, which are typically addressed early in the development of a BCA. In contrast, assessment of effects on microorganisms naturally present in soil, such as changes in soil microbial community structures and the functions they fulfil, has received much less attention. However, considering the importance of microorganisms and the their functions in ecosystems, knowledge on possible adverse effects on the native soil microbial community is important for safety but also for economic reasons, as it may have implications on benefit-cost analyses (Benjamin and Wesseler, 2016). For a long time, investigation of microorganisms, particularly in soil, were limited to cultivation based approaches like plating of samples on nutrient media or community profiling based on nutrient utilization. These techniques suffer from a cultivation bias by favouring certain microbial taxa and neglecting the viable but not culturable soil microorganisms (Hugenholtz, 2002). With the development of cultivation-independent biochemical and molecular approaches, it has become possible to expand analyses to the unculturable microbiota and assess communities in much greater detail (Rastogi and Sani, 2011). Amplification of a DNA marker region for instance located on the ribosomal operon (ITS2 for fungi and 16S rRNA V3-V4 for prokaryota) and subsequent sequencing using high- throughput sequencing (HTS) technology is one of the cultivation-independent approaches that has successfully been applied to study the diversity of microbial communities and the effects certain factors may have on them (e.g. Hartmann et al., 2015). This approach allows the parallel assessment of many samples and results in tens of thousands of sequences per sample that can be compared to existing databases (Rastogi and Sani, 2011). Cultivation-independent methods have been used to assess potential effects of applied soil fungi used for control of insects, weeds, or nematodes and these studies have revealed no or only transient effects of the applied strains (Hirsch et al., 2013; Mayerhofer et al., 2017; Rousidou et al., 2013; Zimmermann et al., 2016). The present study was conducted to assess potential effects of application of M. brunneum formulated as fungus colonized barley kernels (FCBK) to control D. v. virgifera in maize production on soil fungal and prokaryotic community structure determined by HTS of ribosomal markers.

56

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

4.3 Material and methods

Set-up of the pot experiment

The pot experiment consisted of six treatments with six pots per treatment. Pots with a diameter of 32 cm and a height of 33 cm were filled with mixed sandy loam soil with a pH of 6.6 and 3.5 % humus content. Soil was obtained from an agricultural field close to the research center Laimburg (Auer, Italy). Pots were placed in an open greenhouse with a glass roof at the research center and left undisturbed for 13 days in order to allow equilibration of the soil microbial communities. During the entire experiment 300 ml of water was applied per day to each pot by a drip irrigation system. Each pot was fertilized with 5 g of Nitramoncal (13.5% ammonium and 13.5% nitrate; Borealis L.A.T, Austria) and 0.66 g of monoammonium phosphate (NP 12-61; Arpa speciali, Mantova, Italy) once at a depth of 10 cm before placing one kernel of Zea maize L. cultivar Mas 47.P (Maisadour, Austria) in the centre of each pot at a depth of 5 cm. Soil temperature was monitored hourly at a depth of 5 and 20 cm using an Escort Junior data logger (Escort Messtechnik, Zurich, Switzerland). The daily mean soil temperature at 5 cm depth was 23.2 °C on average (ranging between 12.2 and 41.5 °C), and at 20 cm depth it was 22.7 °C (ranging between 13 and 37.1 °C).

Treatments

The entomopathogenic fungus M. brunneum strain EAMa01/58-Su (CECT 20764 Spanish Type Culture Collection; E. Quesada Moraga, Universidad de Córdoba, Spain), whose virulence against D. v. virgifera has been demonstrated (M. Schumann, personal communication), was applied in form of fungus colonized barley kernels (FCBK). Four g of FCBKs, which correspond to an application dose of about 1 x 10 7 colony forming units (CFU) per g were applied. The same number of sterile barley kernels (BK), which resulted in 2.9 g of BK were used as control for possible effects of non-fungal compounds of the formulations. The insecticides Poncho TM 600 FS (clothianidin, Bayer CropScience AG, Germany) and Mesurol TM (methiocarb, Bayer CropScience AG, Germany) were applied in form of dressed maize seeds (Insec). Six untreated pots served as controls (Untreated). All treatments were mixed into the upper 5 cm layer of the soil of each pot by hand immediately after the addition of the maize seed.

Application and assessment of D. v. virgifera and root damage of maize

Each pot was inoculated with 150 (± 47) eggs of D. v. virgifera 38 days after application of the treatments. The eggs were obtained as an egg / sand mixture from a rearing facility (Austrian Agency for Health and Food Safety, Vienna, Austria). The Number of eggs in the egg / sand matrix was assessed by extracting and counting the eggs in 5 g of the egg / sand mixture (10 replicates). Successful hatching was tested by incubating hundred eggs in the laboratory at 25 °C for three weeks. In order to assess the number of emerging D. v. virgifera adults individual pots were covered with traps (Rauch et al., 2016). Traps were installed 26 days after the addition of the eggs and checked three times a week. D. v. virgifera beetles were collected with an aspirator and counted. Root damage caused by larval feeding was evaluated according to the node-injury scale (Oleson et al., 2005) at the end of the experiment.

57

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

Soil sampling

Soil samples were collected right before maize sowing and application of treatment (week 0) on 19 May 2014, and 9 and 18 weeks later on 22 July and 24 September 2014. Samples consisted of four soil cores (diameter of 1 cm and depth of 15 cm) per pot, which were pooled and thoroughly mixed. At each sampling time point 24 soil samples were collected (4 treatments and 6 replicates), sieved through a 5 mm mesh and split into three subsamples, which were used for: (1) the determination of soil dry weight, (2) for the isolation of Metarhizium spp. and subsequent SSR genotyping and (3) for the assessment of effects on indigenous soil microorganisms.

Monitoring of the applied strain

Metarhizium spp. colonies were isolated from soil samples using plating on selective medium according to the protocol described by Laengle et al. (2005). Soil suspensions (0.25 g ml -1) of each sample were plated on four selective medium agar plates, Metarhizium spp. colony forming units (CFU) were counted and median CFU per sample were calculated. One single colony was randomly selected per soil sample for genotyping. DNA extraction of fungal mycelium was performed according to the protocol described by Kepler et al. (2014). Isolates were grown on sterile filter paper placed on PDA agar plates until a firm layer of mycelium was observed. Mycelium of 2 - 4 mm 2 was scraped off the filter paper and transferred to a 2 ml Eppendorf tube containing a mix of 1 mm (Vitaris, Braun Biotech, Germany) and 3 mm glass beads (Merck KGsA, Germany) and 350 μl of Prepman extraction buffer (Applied Biosystems, USA). Fungal mycelium was homogenized with a ball-mill (MM301, Retsch, Germany) at maximum speed for 15 - 30 s. Following homogenization, tubes were incubated in a heating block at 99 °C for 5 min, mixed by inverting, and incubated again for 5 min. After incubation the tubes were centrifuged at 14,000 g for 5 min, rotated by 180°, and centrifuged again for 5 min. The supernatant (175 μl) was transferred to a new tube and stored at -20 °C. DNA extracts were diluted by 1:10 or 1:100 for PCR analyses. Genotyping of isolates was performed with simple sequence repeat (SSR) marker analysis according to the protocol of Mayerhofer et al. (2015). SSR marker set I (Ma2049, Ma2054, Ma2063) and set V (Ma195, Ma327, Ma2287) were amplified in multiplex reactions and amplicon sizes were determined with an ABI 3130XL (Applied Biosystems, USA) using 36 cm capillaries, POP-7TM and GENESCAN TM 400HD ROX TM as an internal size standard. M. brunneum ARSEF 7524 served as reference for allele sizes. PCR fragment sizes were calculated with GeneMarker ® v. 2.4.0 software (SoftGenetics ®, USA).

Determination of the ITS2 sequence of the applied strain

The sequence of internal transcribed spacer region 2 (ITS2) of strain M. brunneum EAMa01/58-Su was determined as described previously (Mayerhofer et al., 2017) using Sanger sequencing with primer pair ITS3 (5’ CAHCGATGAAGAACGYRG 3’) / ITS4 (5’ TCCTSCGCTTATTGATATGC 3’) (Tedersoo et al., 2014).

Analyses of microbial community composition

Soil DNA extraction was performed as described by Mayerhofer et al. (2017) which included three consecutive DNA extractions from 0.5 g soil of each sample. Soil DNA was dissolved in TE buffer (at 1 ml per 1 g soil dry weight) and subsequently subjected to a clean-up procedure using NucleoSpin ® gDNA clean-up kit (Machery- Nagel, Germany). Soil DNA concentration was determined with a Cary Eclipse fluorescence spectrophotometer (Varian, Inc., Australia) and the extract subsequently diluted to a concentration of 2 ng / µl. The fungal ITS2 was 58

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN amplified using primer pair ITS3 / ITS4 (Tedersoo et al., 2014) and the prokaryotic (bacterial and archaeal) variable region (V3-V4) of the small subunit of the ribosomal RNA (16S rRNA) was amplified using a modified version of primer pair 341F (5’ CCTAYGGGDBGCWSCAG 3’) / 806R (5’ GGACTACNVGGGTHTCTAAT 3’) as published by Frey et al. (2016). The 5’ end of the forward and reverse primers were tagged with CS1 and CS2 adapters, respectively, in order to allow multiplexing with the Fluidigm Access Array System (Fluidigm, USA). PCR amplifications were performed as described by Mayerhofer et al. (2017). PCRs were conducted in four replicates for each soil sample. Replicates were pooled and sent to the Génome Québec Innovation Center at the McGill University (Montréal, Canada) for barcoding (Fluidigm Access Array technology), purification with AMPure XP beads (Beckman Coulter, USA) and paired-end sequencing using the Illumina MiSeq v3 platform (Illumina Inc., USA).

Sequence processing and taxonomic classification

The sequences were processed as described by Mayerhofer et al. (2017) and according to the customized pipeline reported by Frey et al. (2016) using UPARSE implemented in USEARCH v8.0.1623 (Edgar, 2010; Edgar, 2013) if not stated differently. In short, quality control included merging of overlapping paired-end reads using fastq_mergepairs (Edgar and Flyvbjerg, 2015), correction of substitution errors using BayesHammer algorithm implemented in SPAdes 3.5 (Nikolenko et al., 2013b; Nurk et al., 2013), removing PCR primer sequences with Cutadapt 1.8.1 (Martin, 2011) and filtering reads with a maximum expected total error of 1 using fastq_filter in USEARCH (Edgar and Flyvbjerg, 2015). After exact dereplication using derep_fulllength in USEARCH, singletons were removed and reads identified as being chimeric were discarded during clustering at 97% sequence identity using cluster_otus in USEARCH (Edgar, 2013). The presence of ribosomal signatures was verified with ITSx (Bengtsson-Palme et al., 2013) or Metaxa2 (Bengtsson-Palme et al., 2015) for eukaryotic or prokaryotic centroids, respectively, and all centroids lacking ribosomal signatures were discarded. Taxonomic classification was performed with the naïve Bayesian classifier implemented in MOTHUR v.1.35.1 (Schloss et al., 2009) using a custom-made database extracted from NCBI GenBank (Benson et al., 2015) and the UNITE (Abarenkov et al., 2010) database for eukaryotes. Prokaryotic V3-V4 sequences were queried against the GREENGENES database (DeSantis et al., 2006; McDonald et al., 2012). Non-fungal sequences were discarded from the eukaryotic dataset and in the prokaryotic dataset only archaeal and bacterial sequences were retained and OTUs assigned as organelles, i.e., chloroplasts and mitochondria were excluded. In order to avoid influences of the abundance of the OTU of the applied strain (OTU 3) on statistical tests it was excluded from the dataset prior to comparison of microbial diversities among treatments.

Statistical analyses

Statistical analyses were performed as outlined in Mayerhofer et al. (2017). Procedures using iterative subsampling such as the assessment of OTU richness, rarefaction curve analyses, and calculation of Bray Curtis (BC) dissimilarities were performed at the lowest sampling depth of a sample in the dataset, i.e., 6180 and 8942 fungal and prokaryotic sequences, respectively. Differences in microbial community structures across treatments and sampling time points were assessed with overall and pairwise PERMANOVA based on BC dissimilarities using the functions adonis within vegan and pairwise.pairm.manova within RVAideMemoire (Hervé, 2017; Oksanen et al., 2016) and Benjamini-Hochberg p-value correction in R (R-Core-Team, 2016). Multivariate homogeneity of groups’ dispersions (variances) of fungal and prokaryotic community structures were assessed among treatments 59

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN and sampling time points using the function betadisper within the package vegan in R and were based on BC distances. Non-metric multidimensional scaling (NMDS) was calculated in R with the function metaMDS included in the package vegan (Oksanen et al., 2016; Shepard, 1962). Overall and pairwise PERMANOVA tests and tests for multivariate homogeneity of groups’ dispersions of relative sequence abundance at OTU level among treatments and sampling time points were based on Euclidean distances and were performed as described above. Fungal and prokaryotic community structures of untreated pots were compared between the present dataset and another dataset derived from a study with a very similar set up comprising a biocontrol experiment against Agriotes spp. in potato (Mayerhofer et al., 2017; SRA accession number PRNJA386024) using OTU richness, overall PERMANOVA, and Multivariate homogeneity of groups’ dispersions as described above. Principal coordinates analyses (PCO) based on BC distances were calculated using cmdscale included in the R core package (Gower, 1966; R-Core-Team, 2016). Venn diagrams displaying shared OTUs were drawn using the function draw.pairwise.venn within the package VennDiagram in R (Chen and Boutros, 2011)

4.4 Results

Abundance of the applied strain, number of adult beetles and damage assessment

The abundance of Metarhizium spp. increased significantly from a median of 2’451 CFU g -1 soil dry weight over all pots at week 0 to a median of 64’521 and 23’993 CFU g -1 soil dry weight in pots treated with FCBK at week 9 and 18, respectively (Figure 4.1 A). Five and six out of six isolates from FCBK-treated pots had the genotype (SSR marker based) of M. brunneum EAMa01/58-Su at week 9 and 18, respectively (Table Appendix C 1). M. brunneum EAMa01/58-Su was not detected at week 0 but one out of six isolates from the treatment Insec and the control at week 9, and two isolates from treatment Insec had the genotype of the applied strain (Table Appendix C 1). The mean number of emerging D. v. virgifera adults ranged from 4.7 to 7.8 per treatment with standard deviations ranging from 3 to 5.3. The number of D. v. virgifera adults and damage of maize plants across all treatments correlated weakly but significantly (Spearman r = 0.53, p < 0.01), and both measures did not differ significantly when comparing individual treatments with untreated pots (data not shown).

60

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

Figure 4.1: Abundance of Metarhizium spp. in CFU g -1 soil dry weight (A) and relative abundance of the OTU (B) including the sequence of the applied strain to total sequence abundance per treatment (fungus colonized barley kernels [FCBK], barley kernels [BK] and clothianidin [Insec]) at week 0, 9 and 18. * indicates significant differences to untreated control at the respective sampling time points (n = 6; p < 0.05).

Soil microbial diversity

After quality filtering the 24 samples contained in total 1’601’688 fungal (22’246 +/- 8’686 sequences per sample) and 1’233’720 prokaryotic sequences (17’135 +/- 2’958 sequences per sample), including 21’160 archaeal sequences. The sequences were clustered into 2’022 fungal (388 +/- 65 OTUs per sample) and 9’428 prokaryotic OTUs (including 39 archaeal OTUs; 2’670 +/- 242 OTUs per sample). Rarefaction analyses, indicated that the sampling effort for fungal sequences of most samples was close to saturation (Figure Appendix C 1 C, D). The variation among samples, however, was higher for fungal as compared to prokaryotic samples (Figure Appendix C 1 C, D). Fungal OTUs were assigned to six phyla (Figure Appendix C 2 A). The total prokaryotic community comprised 45 phyla (including 3 archaeal phyla; Figure Appendix C 2 B). OTU 3 with an abundance of 57’664 sequences was assigned to M. brunneum. Within OTU 3 14’889 sequences exactly matched the ITS2 sequence of M. brunneum EAMa01/58-Su. The relative abundance of OTU 3 increased significantly only in the FCBK-treated pots from a median of 1 % of all sequences per sample prior to the application to 15 % and 10 % at 9 and 18 weeks after the treatment (Figure 4.1 B). CFU counts and relative OTU 3 abundance strongly correlated (r = 0.7, n = 72, p < 0.001). After the exclusion of OTU 3 (to avoid interference with statistical analyses) the total number of fungal sequences decreased to 1’544’024 (21’445 +/- 8’700 per sample).

Effects of treatments on microbial diversity

There were no significant differences in fungal and prokaryotic OTU richness among the treatments and the untreated control at any sampling time point (Figure Appendix C 1 A, B). Prior to application (week 0) no significant differences in the fungal and soil prokaryotic communities were detected based on BC dissimilarities and assessed with overall PERMANOVA (fungi: pseudo F= 1.14, p = 0.13; prokaryota: pseudo F = 1.09 p = 0.13). Overall PERMANOVA including all time points revealed that soil fungal communities were affected slightly by

61

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN treatments, however, pairwise tests did not reveal differences among treatments and untreated pots (Table 4.1, Table 4.2).

Table 4.1: Differences of the community structures of all fungi and prokaryota between treatments at sampling time points assessed with overall PERMANOVA. Degrees of Sums of Mean of sums of Pseudo F- Organism Factor R2 P-value freedom squares squares statistic Treatment 3 0.5199 0.1733 1.6123 0.0543 0.0131 Time 2 1.9782 0.9891 9.2026 0.2068 0.0001 Fungi Treatment x time 6 0.6195 0.1033 0.9607 0.0648 0.5758 Residuals 60 6.4487 0.1075 NA 0.6741 NA Total 71 9.5663 NA NA 1.0000 NA Treatment 3 0.2834 0.0945 1.7386 0.0605 0.0002 Time 2 0.8168 0.4084 7.5162 0.1744 0.0001 Prokaryota Treatment x time 6 0.3231 0.0539 0.9911 0.0690 0.4928 Residuals 60 3.2602 0.0543 NA 0.6961 NA Total 71 4.6835 NA NA 1.0000 NA * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001 NA … not assessed

Multivariate homogeneity of groups’ dispersion tests revealed no differences in dispersion of fungal community structures among treatments. Also, overall PERMANOVA of the relative sequence abundance per fungal OTU revealed no significant changes after the application of any of the treatments. Small overall differences in the soil prokaryotic communities among the treatments were detected (Table 4.1, Figure 4.2 A, B) and pairwise comparisons among treated and untreated pots revealed significant changes of the soil prokaryotic communities in BK-treated pots (Table 4.2).

62

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

Figure 4.2: NMDS of prokaryotic samples showing treatment-effects at week 9 (A, stress = 0.13) and week 18 (B, stress = 0.15) and NMDS of fungal (C, stress = 0.12) and prokaryotic communities (D, stress = 0.09) in untreated control samples over time (n = 6).

Multivariate homogeneity of groups’ dispersion tests revealed no differences in dispersion of prokaryotic community structures among treatments. Also, relative sequence abundances of none of the prokaryotic OTUs was changed significantly after the application of any treatments.

Table 4.2: Differences of the community structures of all fungi and prokaryota between treatments at sampling time points assessed with pairwise PERMANOVA.

Pairwise comparison of each treatment with untreated (p-value) Organism BK FCBK Insec Fungi 0.3515 0.5930 0.5694 Prokaryota 0.0300 0.1962 0.1962

Pairwise comparison of sampling time points (p-value) Organism Week 0 and 9 Week 0 and 18 Week 9 and 18 Fungi 0.0001 0.0001 0.0001 Prokaryota 0.0001 0.0001 0.0001

Temporal changes of the microbial diversity

Soil fungal and prokaryotic community structures were assessed at three sampling time points over a period of 18 weeks. OTU richness of fungal and prokaryotic communities in untreated pots did not change significantly during this time period (Figure Appendix C 1 A, B). However, a tendency of decreased OTU richness of fungal communities and an increased OTU richness of prokaryotic communities was observed in untreated pots and in all treatments over time (Figure Appendix C 1 A, B). Overall PERMANOVA showed that fungal and prokaryotic communities changed over time and pairwise tests revealed that fungal and prokaryotic community structures 63

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN differed at all sampling time points (Table 4.1, Table 4.2). Dispersion of fungal and prokaryotic community structures also changed significantly over time (fungi: pseudo F = 8.49, p = 0.0008; prokaryota: pseudo F = 10.47 p = 0.0002). NMDS plots showed continuous changes of the fungal and the prokaryotic communities in untreated pots from week 0 to 9 and 18 (Figure 4.2 C, D). Overall PERMANOVA of relative sequence abundance of each OTU among sampling time points revealed that 9.7 % of the fungal OTUs (197) were significantly affected by time and 64 % of these fungal OTUs varied also in dispersion. Most of these significantly affected OTUs belonged to the phylum Ascomycota (108), followed by Glomeromycota (34), Basidiomycota (28), Zygomycota (12), unclassified fungi (12), and Chytridiomycota (3). Of these five fungal phyla, Glomeromycota (pseudo F = 54.6, p<0.001) and Chytridiomycota (pseudo F = 3.9, p<0.05) were the only phyla for which relative sequence abundances changed at phylum level. Means and dispersions of relative sequence abundances of Chytridiomycota (in total 102 OTUs) and Glomeromycota (in total 126 OTUs) changed significantly over time. Chytridiomycota increased slightly at week 9 followed by a decrease at week 18. Glomeromycota were significantly more abundant in soil of all treated and untreated pots at week 9 and 18 than at week 0 (Figure 4.3 A) and the community structure of Glomeromycota based on BC distances changed significantly over time (pseudo F = 21; p = 0.0001; Figure 4.3 B). The 34 OTUs belonging to the Glomeromycota which changed significantly over time belonged to the genera Ambispora , Claroideoglomus , Glomus , Paraglomus, Rhizophagus, and Septoglomus (or unknown). Overall PERMANOVA analyses at OTU level showed that 13.2 % of the prokaryotic OTUs (1249) were affected by time and thereof 78 % also changed significantly in dispersion. These OTUs belonged to 31 different phyla and the five most prevalent phyla were Proteobacteria (348), Acidobacteria (137), Verrucomicrobia (136), Bacteroidetes (133), and Planctomycetes (108).

Figure 4.3: Relative sequence abundance of the phylum Glomeromycota per treatment and sampling time point (A). Letters indicate significant difference (p < 0.05) among sampling time points for each treatment individually (n = 6). Ordination (NMDS) of community structures of Glomeromycota (B, stress = 0.11) per treatment and sampling time point (n = 6).

64

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

Comparison of effects on soil microbial communities to a similar pot experiment

Finally, fungal and prokaryotic OTU richnesses and community structures in untreated pots were compared to a similar pot experiment performed by Mayerhofer et al. (2017). The comparison revealed a significantly lower fungal OTU richness in the present study, but no differences in prokaryotic richness (Figure 4.4 A, B). Venn diagrams displayed that 30.8% and 46.2% of fungal and prokaryotic OTUs were shared between the untreated pots of both experiments (Figure 4.4). Furthermore, PCO and overall PERMANOVA analyses showed major and significant differences in fungal and prokaryotic communities between the two experiments (fungi: pseudo F = 39.04, p = 0.0001, prokaryota: pseudo F = 53.23, p = 0.0001; Figure 4.4 C, D). Also, dispersions of fungal and prokaryotic community structures in untreated pots were significantly greater in the present study compared to the Agriotes-pot experiment (Mayerhofer et al., 2017; p < 0.01, Figure 4.4 E, F).

Figure 4.4: Comparative analysis of soil fungal communities in untreated pots of the biocontrol experiment to control Agriotes obscurus (A, Mayerhofer et al. 2017) and D. v. virgifera (D, present study): OTU richness of fungal (A) and prokaryotic communities (B) in untreated pots at different sampling time points (n = 6). Letters indicate significant differences (p < 0.05). Venn diagrams showing the shared number of fungal (C) and prokaryotic (D) between the untreated pots of both experiments. Ordination revealing fungal (E) and prokaryotic (F) community structures in untreated pots before application (pre) and after application (post I and post II) for each study (n = 6).

65

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN

4.5 Discussion

Assessment of presence and abundance of an applied BCA strain in soil is important for evaluating efficacy and possible effects on native soil microbial communities. In our study this was achieved by using a cultivation- dependent as well as a cultivation-independent approach, i.e., assessment of Metarhizium spp. CFU g -1 soil dry weight followed by SSR marker based genotyping of selected isolates as well as assessment of the OTU including the marker sequence of the applied strain within the HTS dataset. Results of both methods showed significant increases in the pots treated with FCBKs, but not in the other treatments. M. brunneum EAMa01/58-Su was also detected in some pots treated with the insecticide and untreated control pots, however, at low prevalence (one to two out of six strains analyzed). In these treatments the applied strain was possibly transferred from treated to untreated pots by insects as has been reported also in other studies (Baverstock et al., 2008). Although data indicate that the abundance of the applied strain increased to a median of about 6.5 x 10 4 CFU g -1 soil dry weight in the FCBK-treated pots, no acute biocontrol effects were observed. The variation of the mean number of D. v. virgifera beetles per treatment was high (standard deviation ranging from 3 to 5.3), which was possibly due to variability in inoculum quantities (150 ± 47 eggs per pot) leading to an unequal infestation rate with D. v. virgifera. This may have obscured detection of small effects of the different treatments. Varying efficacies of M. brunneum (formerly M. anisopliae ) to infect D. v. virgifera have also been found in field experiments, i.e., no biocontrol effect (Rauch et al., 2017) or 31 % reduction of D. v. virgifera beetles (Pilz et al., 2009). Despite the fact that no biocontrol effects were observed in this study, the 19-fold increase of the density of M. brunneum EAMa01/58-Su in the FCBK treatments observed in our study demonstrated that native soil was sufficiently exposed to the applied strain, which allowed assessment of possible effects on microbial communities. Prior to registration of plant protection products in Europe the assessment of potential effects of applied microorganism on native soil microbial populations is required (Commission regulation No 544/2011, L155/66). The present study revealed that neither changes in soil fungal nor prokaryotic community structures as well as relative sequence abundance of individual OTUs could be detected upon the application of M. brunneum formulated as FCBK. This is in accordance with previous studies that detected no or only transient effects on soil microorganisms involving Beauveria bassiana (Hirsch et al., 2013; Rai and Singh, 2002) , B. brongniartii (Rai and Singh, 2002; Schwarzenbach et al., 2009), Metarhizium anisopliae sensu lato (Hu and St Leger, 2002; Kirchmair et al., 2008), and M. brunneum (Mayerhofer et al., 2017). In the latter study a different M. brunneum strain formulated as FCBK and BK alone have been used in a pot experiment and small changes in fungal community structures were observed (Mayerhofer et al., 2017). However, because fungal spores alone did not affect fungal communities authors suggested that effects resulted mainly from the compounds of the formulation. Moreover, effects of FCBK were only observed in the pot experiment and not in the field. Regarding effects of BK treatment, only in the present study changes on soil prokaryotic communities were observed, but not in the previous study by Mayerhofer et al. (2017). The experimental set-up of the two studies was similar but differed in biotic factors including insect pest species ( Agriotes spp. vs D. v. virgifera ), plant species (potatoes vs . maize), and applied M. brunneum strain (ART2835 vs EAMa01/58-Su). Moreover, the soil microbial community structures appeared to be different as reflected in significant differences in fungal OTU richness, low percentages of shared OTUs in untreated pots (30.8% of the fungal and 46.2% of the prokaryotic OTUs), and major differences in community structures (Table 4.1, Table 4.2, Figure 4.4). These differences in microbial communities suggest that the degree of resistance of a microbial community, also referred to as 66

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN insensitivity to disturbance (e.g. by Shade et al., 2012), to applications of FCBK and BK may depend on soil specific microbial communities. To our knowledge, differences in resistance to massive application of microorganisms in soil between different microbial community structures have not been assessed. However, in a review on the effects of various disturbances, such as heat, addition of heavy metals, organic matter, or pesticides, freeze-thaw or dry-wet cycles, compression, and tillage on soil microorganisms, Griffiths and Philippot (2013) concluded that resistance depends on the type of disturbance, soil physical-chemical parameters, as well as the indigenous soil microbial community composition and soil history. A prerequisite to assess the resistance of a microbial community to disturbance is the assessment of the intrinsic variability of the system, which comprises fluctuations in community composition without perturbation (Shade et al., 2012). Thus, the lacking effects of FCBK on fungal communities in the present study may be explained by the variability of the system, as shown in untreated control pots (Figure 4.4 E). A variety of carrier materials are used to formulate entomopathogenic fungi as BCAs, including kernels (rice or barley), liquids (water or oils), or granules (alginate; Glare and Moran-Diez, 2016). Carrier materials provide a structure for the product and they can protect the fungus from biotic and abiotic stresses and/or serve as nutrient source to support growth. BK had a small effect on prokaryotic community structures, whereas the corresponding product FCBK did not affect the prokaryotic community structures and none of them had an effect on individual OTUs. Decomposition of the carrier materials may be the reason for the prokaryotic community changes in the BK treatment, however, changes were subtle as no single affected OTU could be identified. Barley kernels, which include compounds such as starch, lipids, proteins, dietary fibre (particularly beta-glucans), vitamins, and minerals (Šterna et al., 2017), were shown to be degradable by microorganisms such as Rhizopus oligosporus , various Saccharomyces, and Lactobacillus species in the fermentation process of the food product barley tempeh (Feng et al., 2007) or by microorganisms such as Lactobacillus plantarum , Aspergillus niger , Trichoderma reesei , Rhizopus oligosporus, and Geotrichum candidum during malting in the beer production process (Hattingh et al., 2014). In FCBK the niche (BK) may have been occupied and/or the respective compounds may have been consumed by M. brunneum and thereby turned it inaccessible to the soil microbial communities. The fungal and prokaryotic communities changed significantly over time. This was reflected in differences in community structures and in differences in relative sequence abundance of 10 % of the fungal and 13 % of the prokaryotic OTUs. Temporal variations were also observed in the pot experiment conducted by Mayerhofer et al. (2017). Changes in biotic factors represented by the addition of plants and insects as well as abiotic factors such as temperature or moisture (Dematheis et al., 2012; Guo et al., 2017; Poll et al., 2013) may have caused these temporal effects. The addition of maize was probably responsible for the major part of the temporal changes in fungal communities. These changes were strongly associated with six genera belonging to Glomeromycota, also known as arbuscular mycorrhiza forming symbiotic interactions with maize roots (Schüβler et al., 2001). In a study comparing diversity of arbuscular mycorrhiza in fields including maize crop rotations or grassland only, one species was exclusively found in the maize field (Oehl et al., 2003), however, the specificity of these interactions is still under debate (Sanders, 2003). Furthermore, temporal effects might have been caused by root exudates released by the maize plant, which differ in quantitative and qualitative composition among growth stages (reviewed in Badri and Vivanco, 2009). The maize plants at the three sampling time points were at different growth stages (week 0 no plants, week 9 approximately 10 leaves unfolded, week 18 mature plant) and thus may have differently affected microbial communities in the rhizosphere as shown in other studies (e.g. Cavaglieri et al., 2009). Another biotic factor contributing to temporal changes might have been D. v. virgifera larvae feeding on 67

RESPONSE OF SOIL MICROBIAL COMMUNITIES TO THE APPLICATION OF A FORMULATED METARHIZIUM BRUNNEUM BIOCONTROL STRAIN roots, which has been shown to alter soil microbial community structures (Dematheis et al., 2012), and may have affected community structures in our study. Furthermore, effects of changing temperatures on microbial communities are well documented (e.g. Riah-Anglet et al., 2015). As the daily median temperature in our experiment ranged from 16 to 29 °C with minimal temperature of 12 °C and with maximal temperature of 42 °C, this factor most likely also contributed to observed variabilities. Soil microbial communities appeared to be resistant to massive application of M. brunneum in our study. This is consistent with previous studies that have reported no or only minor and transient changes of microbial community structures upon the application of entomopathogenic fungi ( Beauveria and Metarhizium ). The growing number of studies revealing resistance of soil microorganisms to Metarhizium and Beauveria applications suggests that compositional diversity of soil microbial communities is not affected by these applications. Whether stability of soil microbial communities at the taxonomic level transfers into stability of functions provided by soil microorganism awaits further investigation.

4.6 Acknowledgements

We would like to thank Enrique Quesada Moraga for providing the fungal biocontrol strain. We are grateful to Christopher Oliver Pabst for supporting soil sampling and detection of the applied strain and to Roland Zelger for supporting the experimental set-up of the pot experiment. This study was part of the EU-project INBIOSOIL (Grant Agreement No. 282767).

68

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

Evaluating PCR biases of one dominant target on the assessment of

community structures of a soil microbiome

A manuscript based on this chapter will be submitted for publication. Mayerhofer, J., Enkerli, J. & Widmer Franco (2018) Evaluating PCR biases of one dominant target on the assessment of community structures of a soil microbiome.

5.1 Introduction

Amplicon sequencing of genetic markers is a widely used method in microbial ecology to assess microbial diversity and community structures (Shokralla et al., 2012). For this, DNA is extracted from an environmental sample and PCR is performed with primer pairs specific for a genetic marker, which allows the assessment at a taxonomic level, e.g., ribosomal marker region, or at a functional level, e.g., gene of an ammonium monoxygenase (Herbold et al., 2015). PCR products are then sequenced using a next generation sequencing (NGS) platform, such as Illumina MiSeq (Boughner and Singh, 2016). Both steps, the PCR and NGS may introduce biases, such as preferential amplification of marker regions and sequencing errors (Hartmann and Widmer, 2008; Polz and Cavanaugh, 1998; Schirmer et al., 2015). Amplicon sequencing of ribosomal marker regions has been used to assess various questions regarding soil microorganisms in agricultural soils and one of them is the assessment of potential effects of applied microorganisms on soil microbial community structures (Hirsch et al., 2013; Kröber et al., 2014). In such studies it is important to assess exposure of native soil microbial communities to the applied strains prior to investigating changes in community structures of native soil microorganisms. Amplicon sequencing has been shown to be useful for estimating exposure and simultaneously assessing native soil microbial community structures (Chapters 3 and 4; Hirsch et al., 2013; Kröber et al., 2014). However, it remains to be investigated, if preferential amplification of ribosomal marker regions occurred and impeded the assessment of soil microbial community structures, due to the presence of one highly abundant sequence, and if sequencing errors influenced the confirmation of exposure to the applied strain. Chapter 3 describes a study where the biological control strain M. brunneum ART2825 was applied to soil microbial communities, which resulted in an increase of the operational taxonomic unit (OTU) containing the sequence of the applied strain up to 62 % relative sequences abundance. The remaining sequences might have not accurately represented the native soil microbial community due to preferential amplification during PCR of the genetic markers (16S V3-V4 for prokaryotic and ITS2 for fungal sequences) (Polz and Cavanaugh, 1998). Also, the exposure of soil microbial communities to M. brunneum ART-2825, i.e., presence and abundance of the applied strain may have been hampered by sequencing errors in the target sequence because the identification of the applied strain was based on an exact match of the ITS2 of the applied strain. In order to simulate the addition of one highly abundant strain without any potential in vivo interactions of the soil microbial community and the applied strain, a plasmid containing the ITS2 of the applied strain M. brunneum ART2825 (pITS2-ART2825) was added at different quantities to the DNA extract of one soil sample, in which 69

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME previously a very low abundance of native Metarhizium was detected (Chapter 3). Then the OTU containing the added sequence was removed from the dataset, and apparent structures of the soil microbial community structures was assessed among replicates to which different levels of the plasmid pITS2-ART2825 were added. In order to assess sequencing errors in the sequence of the applied strains, the amount of unique sequences within the OTU containing the sequence of the applied strain were compared among the replicates containing different levels of the plasmid pITS2-ART2825 and the control.

5.2 Material and methods

Soil sample

All analyses of the ribosomal marker regions for NGS were performed with the same DNA extract, which was obtained from the soil of an untreated pot before the application of any treatments (week 0) from the pot experiment described in Chapter 3 (sample: AGR-5-1-1). The background Metarhizium abundance in this sample was 115 CFU g -1 soil dry weight and was assessed by plating on selective medium and the abundance of the OTU containing the sequence of the applied strain was 16 (resulting in a relative abundance of 0.0004) Sampling procedure, sample processing, soil DNA extraction and subsequent clean-up of the DNA extract was also reported in Chapter 3.

Plasmid construction and addition

The ITS2 of M. brunneum ART2825 (Genbank Acc. N. KY786031) was amplified using primer pair ITS3 (5’ CAHCGATGAAGAACGYRG 3’) / ITS4 (5’ TCCTSCGCTTATTGATATGC 3’) (Tedersoo et al., 2014) and cloned into the pCR ®4Blunt-TOPO ® vector following the instructions of the Zero Blunt ® TOPO ® PCR Cloning Kit (Invitrogen, Carlsbad, California, U.S.A.). Escherichia coli JM109 competent cells (Promega, Madison, Wisconsin, U.S.A.) were transformed with the plasmid pITS2-ART2825 (total length of 4’537 bp including PCR product). Positive clones were identified using colony PCR as described in Schneider et al. (2011). The maximum abundance of Metarhizium spp. observed in Chapter 3 and 4 was 5.7 x 10 5 CFU g-1 soil dry weight, which is a commonly observed level (Scheepmaker and Butt, 2010). This translates into 114 Metarhizium CFU per PCR containing 20 ng of DNA (using a soil DNA extract with 100 µg DNA g -1 soil dry weight). It has been estimated that the Metarhizium genome includes approximately 100 copies of ITS2 (Schneider et al., 2011). Therefore, 1.2 x 10 4 plasmid copies of pITS2-ART2825 were added in the medium level treatment used to spike the soil DNA extract (Table 5.1). The high and low level treatments included 100 fold more or 100 fold less plasmid copies, respectively.

Table 5.1: Concentration of the plasmid pITS2-ART2825 added to 20 ng of soil DNA extract at different levels (n = 6). Concentration of pITS2-ART2825 Level Plasmid DNA [ng / PCR] Plasmid copy number per PCR Control 0 0 Low 6 x 10 -7 1.2 x 10 2 Medium 6 x 10 -5 1.2 x 10 4 High 6 x 10 -3 1.2 x 10 6

70

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

Amplicon sequencing

Six replicates of each concentrations of the plasmid were prepared containing 20 ng of the soil DNA extract (Table 5.1) resulting in a total of 24 replicates. PCR reactions amplifying the ITS2 and 16S v3-v4 to assess fungal and prokaryotic soil communities were performed as described in Chapter 3 using primer pairs ITS3 / ITS4 and 341F (5’ CCTAYGGGDBGCWSCAG 3’) / 806R (5’ GGACTACNVGGGTHTCTAAT 3’) (Frey et al., 2016) which included adapter sequences CS1 (forward) and CS2 (reverse) at the 5’ end of the respective primer to allow multiplexing with the Fluidigm Access Array System (Fluidigm, South San Francisco, California, U.S.A.). PCR was repeated four times, and amplification products were pooled. Sequencing on the Illumina MiSeq v3 platform was performed at the Génome Québec Innovation Center at the McGill University (Montréal, Canada).

Sequence processing, taxonomic classification and statistical analyses

Quality filtering of the sequences was performed according to a pipeline (Frey et al., 2016) largely based on UPARSE (Edgar, 2013) and described in detail in Chapter 3. Sequences with 97 % sequence identity were clustered into operational taxonomic units (OTU) using USEARCH (Edgar, 2013). The OTU which contained the sequence of M. brunneum ART2825 was additionally clustered based on 100 % identity to obtain abundance of unique sequences within this OTU. Taxonomic classification was performed using the naïve Bayesian classifier implemented in MOTHUR v.1.35.1 (Schloss et al., 2009) and a custom-made Genbank (Benson et al., 2015) as well as the UNITE (Abarenkov et al., 2010) databases for eukaryotic sequences and the GREENGENES database (DeSantis et al., 2006; McDonald et al., 2012) for prokaryotic sequences. Statistical analyses were performed as described in Chapter 3. Differences in relative sequence abundance of the OTU containing the sequence of the plasmid among replicates spiked with different levels of the plasmid was assessed by ANOVA and TukeyHSD multiple comparison in R (R-Core-Team, 2016). Venn diagrams were visualized using the package VennDiagram in R (Chen, 2016). The comparison of fungal and prokaryotic OTU richnesses and community structures, using nonmetric multidimensional scaling (NMDS) and analysis of similarity (ANOSIM) was performed as described in Chapter 3 (Clarke, 1993; Clarke and Gorley, 2015; Oksanen et al., 2016; Shepard, 1962).

5.3 Results

Assessment of the sequence data obtained

A total number of 880’183 fungal sequences (36’674 ± 18’189 sequences per replicate) and of 441’132 prokaryotic sequences (18’381 ± 1’772 sequences per replicate) were obtained after quality filtering. Clustering at 97 % identity level revealed 1’332 fungal OTUs (530 ± 126 OTUs per replicate) and 6’625 prokaryotic OTUs (2’612 ± 117 OTUs per replicate).

The OTU containing the sequence of pITS2-ART2825

OTU-1 was assigned to Metarhizium brunneum and confirmed the ITS2 sequence of plasmid pITS2-ART2825. In total (all replicates combined), OTU-1 contained 282’323 sequences. Spiking the soil DNA extract with low level of pITS2-ART2825 (1.2 x 10 2 plasmid copies per PCR) did not result in significantly increased relative abundances of OTU-1 compared to the control, whereas spiking with medium (1.2 x 10 4 plasmid copies per PCR) and high levels (1.2 x 10 6 plasmid copies per PCR) resulted in a significant increase of the relative abundances of the OTU-

71

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

1 of 0.08 ± 0.005 and 0.7 ± 0.024, respectively (Figure 5.1A). OTU-clustering based on 100 % sequence identity revealed 9’987 unique sequences and the most abundant unique sequence with 107’948 sequences exactly matched the sequence of the applied strain. The abundance of the second most abundant unique sequence was only 2’845 sequences and of the remaining unique sequences 1008 were singletons. In the unspiked DNA extract (control) between 3 and 28 unique sequences occurred in OTU-1.

Figure 5.1: Relative sequence abundance of OTU-1 which contains the sequence of the applied plasmid (A) and relative sequence abundance of total sequences without OTU-1 (B) among replicates to which different levels of the plasmid pITS2-ART2825 (low: 1.2 x 10 2, medium: 1.2 x 10 4 and high: 1.2 x 10 6 copies of the plasmid per PCR) were added (n = 6).

Potential effects of spiking with plasmid pITS2-ART2825

After removal of OTU-1, which contained the sequence of pITS2-ART2825, 68 % of the fungal sequences (a total of 597’860 sequences and 24’911 ± 18’398 sequences per replicate) were retained Figure 5.1B. Only robustly detectable OTUs, which occurred in all six replicates in at least one group (plasmid levels or the control), were used for further statistical analysis, i.e., for Figure 5.2, Figure 5.3 and ANOSIM. The dataset included 341 fungal and 1’550 prokaryotic robustly detectable OTUs. Venn diagrams showed that 53 % of the fungal 48 % of the prokaryotic robustly detectable OTUs were found in all 24 replicates (Figure 5.2). 0.6 to 13 % of the fungal OTUs and 5 – 7 % of the prokaryotic OTUs were specifically found in all replicates treated with the same level of pITS2- ART2825 or the control (Figure 5.2). Replicates containing low and medium levels of pITS2-ART2825 each shared 67 and 66 % of the fungal OTUs with the control, whereas the replicates with high level of pITS2-ART2825 and the control shared only 60 % of the OTUs (Figure 5.2). The control and replicates, which contained low or medium or high level of pITS2-ART2825 shared 57 - 59 % of the prokaryotic OTUs.

72

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

Figure 5.2: Venn diagram showing the shared robustly detected OTUs (occurring in all replicates of at least one group) in fungal (A) and prokaryotic (B) communities among groups treated with low (1.2 x 10 2), medium (1.2 x 10 4), and high (1.2 x 10 6 copies of the plasmid pITS2-ART2825 per PCR) and the control (n = 6). A total number of 341 fungal (A) and 1550 prokaryotic (B) OTUs were robustly detected.

OTU richness of fungal communities did not differ among different levels of added plasmid pITS2-ART2825 and the control (Figure 5.3A), whereas prokaryotic OTU richnesses of replicates with high level of the plasmid were significantly lower than with low level of added pITS2-ART2825. Furthermore, overall ANOSIM revealed no differences among plasmid levels and the control for fungal (R = 0.07, p = 0.03) and prokaryotic community structures (R = 0.002, p = 0.5). Also, NMDS indicated that fungal and prokaryotic community structures did not differ among levels of added plasmid and the control (Figure 5.3C, D).

73

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

Figure 5.3: OTU richness of fungal (A) and prokaryotic (B) soil communities among replicates which were treated with increasing levels of the plasmid pITS2-ART2825. NMDS revealing community structure of fungi (C) and prokaryotes (D) in soil DNA extracts treated with different levels of the plasmid (n = 6).

5.4 Discussion

Investigations of potential effects of applied microbial strains on soil microbial community structures may include biases, which occur during PCR such as preferential amplification of target regions, or during NGS including sequencing errors. To evaluate, if the fungal and prokaryotic communities after removal of the OTU containing the ITS2 sequence of the applied strain remains the same, a spiking experiment, in which increasing concentration of a plasmid containing the sequence of the applied strain were added to the same soil DNA extract, was designed. The addition of the medium and high level of the plasmid pITS2-ART2825 lead to a significant increase in relative abundance of OTU-1, which contained the sequence of the plasmid. However this increase was not proportional to the increasing amount of pITS2-ART2825 added to the DNA extract, most likely because amplicon sequencing is not a method for quantification. The high level addition of pITS2-ART2825 yielded on average 70 % relative abundance of the OTU of the applied strain which was 8 % more than the maximum relative sequence abundance 74

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME of the applied strain observed and described in Chapter 3. The confirmation of exposure to the applied strain requires assessing the abundance of the exactly matching sequence and this may be hampered by random sequencing errors. Depending on the amount of erroneous positions in the sequence, erroneous sequences may cluster within the clustering threshold of 97 % sequence identity to the applied strain or form a new OTU. The latter probably did not occur in our dataset, because no other Metarhizium OTUs besides the OTU containing the sequence of pITS2-ART2825 were obtained. However, the OTU which contained the sequence of the plasmid pITS2-ART2825 also contained 9’986 other unique sequences, which differed in sequence identity of up to 3 % to the sequence of pITS2-ART2825. As in the untreated replicates only a very low number of unique sequences (3 – 28) with a very low relative sequence abundance (Figure 5.1) was detected, which may represent native Metarhizium strains, the majority of the unique sequences in the spiked replicates most likely represent sequencing or PCR errors. These most likely erroneous sequences remained in the dataset after thorough quality control. Erroneous sequences may result from substitution type errors which have been found to be the most prevalent errors in Illumina sequencing (Schirmer et al., 2015). Due to the high abundance of probably erroneous unique sequences within the OTU of pITS2-ART2825 (62 %), we used the abundance of the OTU containing the sequence of pITS2-ART2825 rather than the abundance of the exact sequence of pITS2-ART2825 in order to confirm the presence and abundance of pITS2-ART2825. The significantly increased relative sequence abundance of OTU-1 in replicates to which medium and high levels of pITS2-ART2825 was added allowed the assessment of potential biases in amplicon sequencing of soil microbial communities in the presence of one highly abundant sequence. After the removal of the OTU containing the sequence of pITS2-ART2825 and selecting OTUs which were robustly detectable (present in all replicates of at least one plasmid addition level or the control) only 26 % of the total fungal and 23 % of the total prokaryotic OTUs were retained. The lost OTUs may either represent errors or their abundance may have been just at the detection limit of PCR and amplicon sequencing and therefore they were not detected in all replicates of one plasmid addition level or the control. Comparing the number of prokaryotic OTUs which each level of plasmid addition individually shared with the control revealed that similar percentages of shared prokaryotic OTUs (57 – 59 %) were found regardless which amount of plasmid was added. Whereas when comparing shared fungal OTUs, the replicates containing high levels of pITS2-ART2825 shared fewer OTUs (60 %) with the control as compared to replicates containing low (67 %) and medium levels (66 %). As there were no differences in shared OTUs among different levels of added plasmid, except for the high level on fungal OTUs, probably no preferential amplification due to the presence of one highly abundance sequence occurred. Furthermore, there were no significant differences in OTU richness of fungal communities and community structures of soil fungi and prokaryota assessed with ANOSIM and NMDS among replicates spiked with different levels of the plasmid. OTU richnesses of prokaryotic sequences differed significantly among replicates containing low and high level of pITS2-ART2825, but OTU richnesses of both did not differ compared to the control, therefore these differences are unlikely to be caused by the addition of pITS2-ART2825. As the primer pair for the 16S V3-V4, which was used for the prokaryotic community analysis, does not bind to pITS2-ART2825 in silico, biases due to preferential amplification were unlikely to occur. In contrast, primer pairs used for the assessment of fungal community structures amplified the ITS2 sequence contained in pITS2-ART2825, and the only effects observed were a lower number of shared OTUs between replicates containing the highest level of pITS2-ART2825 and the control, and slightly lower OTU richnesses in the replicates which contained the highest level of pITS2-ART2825.

75

EVALUATING PCR BIASES OF ONE DOMINANT TARGET ON THE ASSESSMENT OF COMMUNITY STRUCTURES OF A SOIL MICROBIOME

In conclusion, OTU clustering is most likely an important step to group erroneous sequences to the original seed sequence, and even at a high relative sequence abundance of the OTU containing the sequence of pITS2-ART2825 of 0.7 the soil microbial community structures were accurately assessed, suggesting that preferential amplification did not occur.

76

GENERAL DISCUSSION

General discussion

6.1 Quantifying exposure of soil microorganisms to applied biocontrol strains

Abundance of the applied strains using a cultivation-dependent approach

Exposure of soil microorganisms to the applied M. brunneum strains in both pot experiments and the field experiment was confirmed by quantifying the abundance of Metarhizium spp. on selective medium agar plates followed by identification of the applied strains using genotyping with SSR markers. For Metarhizium there are 41 SSR markers available (Enkerli et al., 2005; Oulevey et al., 2009). All or some of these markers have been used in several studies to discriminate applied strains from those of native Metarhizium spp. (Castro et al., 2016b; Rauch et al., 2017; Rogge et al., 2017), as well as to study genetic diversity of Metarhizium populations (Enkerli et al., 2016; Freed et al., 2011; Oulevey et al., 2009; Steinwender et al., 2015; Steinwender et al., 2014). Originally these SSR markers were developed for M. anisopliae but this was before the most recent reclassification of the former M. anisopliae species complex into 10 new species (Bischoff et al., 2009; Kepler et al., 2014). Because of the recent taxonomic revision, it was required to assess transferability of these 41 SSR markers among Metarhizium spp. as well as the degree of polymorphism of each marker in different Metarhizium spp. Transferability of SSR markers was reported in Chapter 2 and allowed the selection of two sets of three markers each (Set I and V), which were polymorphic in M. brunneum strains and therefore suitable to discriminate the applied M. brunneum strains from those of native Metarhizium spp. Both applied strains, M. brunneum ART2825 and EAMa01/8-Su, differed in allele sizes at all six loci (markers; Table Appendix D 1). In the Diabrotica -experiment (Chapter 4) data from control and treated pots revealed that all six SSR markers had distinct allele sizes for the applied and native strains of Metarhizium spp. These were grouped into five native genotypes (Table Appendix D 1). Therefore, theoretically one SSR marker (anyone of the six markers) would have been sufficient to identify the applied strain. In contrast, in the Agriotes -pot experiment (Chapter 3) six of a total of 17 native genotypes displayed one to five SSR markers (of a total of six) with identical allele sizes to the applied strain (ART2825, Table Appendix D 1). One genotype differed only in one SSR marker (Ma2049, Set I) from the biocontrol strain and the difference in allele size was only one repeat (2 nucleotides) therefore this strain was possibly a mutant of the biocontrol strain. For the Agriotes - pot experiment it was necessary to include all six SSR markers, and especially marker Ma2049 was important, to discriminate ART2825 from native Metarhizium genotypes. Within the 41 available SSR markers there are additional markers (to the six markers used in Chapter 3 and 4) which revealed high allele diversity in M. brunneum strains (Chapter 2,Table Appendix A 4), and which might be useful to confirm the difference between the ART2825 and native genotypes. Strain M. brunneum ART2825 was originally isolated from a field at the research station Reckenholz from where soil was also obtained for the pot experiment, i.e., all genotypes originate from soil of the same location. This may explain why close relatives to ART2825 were detected in this experiment. A similar situation was observed in the Agriotes -field experiment, where six of 12 native genotypes shared one to five SSR marker alleles (of a total of six) with the applied strain ART2825. The field experiment was located about 15 km from the original location of isolation of ART2825, which may again explain the occurrence of close relatives of ART2825. Besides ART2825, three native genotypes were found in the soil of both, the pot and the

77

GENERAL DISCUSSION field experiment, suggesting that both soils may harbour related Metarhizium populations. Similarly, one Metarhizium genotype was shared between isolates from two locations in Switzerland (approximately 140 km apart) using 41 SSR markers (Oulevey et al., 2009). In our studies one genotype was found in all three experiments, the Diabrotica -pot experiment in Italy as well as in the pot and the field Agriotes -experiments in Switzerland, and another genotype was found in the Diabrotica -experiment and the Agriotes -field experiment. This revealed that these genotypes occurred over a distance of approximately 240 km. Assessment of large scale distribution of Metarhizium genotypes among different studies has been hampered by the use of different techniques to assess genetic diversity, different genetic markers and different sampling schemes (Freed et al., 2011; Inglis et al., 2008; Oulevey et al., 2009). However, using the same sampling strategy and effort two M. brunneum genotypes (out of 29) were detected in Switzerland as well as in Germany (Enkerli et al., 2016) suggesting occurrence of the same Metarhizium strain over great distances. Selective cultivation followed by SSR marker analyses proved to be an efficient and reliable tool for the detection of the applied biocontrol strains in our experiments.

Detection of the applied strains using amplicon sequencing

Amplicon sequencing of ribosomal marker genes was primarily chosen because of its applicability to assess soil microbial communities (Boughner and Singh, 2016). However, it was also possible to estimate exposure of soil microorganisms to the applied strains. In all three experiments the OTU that contained the exact sequence of the applied strains was detected, and its relative sequence abundance correlated with CFU counts assessed on selective medium (Chapter 3 and 4). The sequence which exactly matched the sequence of the applied strains was the most abundant sequence within this OTU in all fungal treated samples, i.e., 49 % in the Agriotes -pot experiment, 48 % in the Diabrotica -pot experiment and 13 % in the Agriotes-field experiment, and was significantly more abundant only in fungal treated samples. The remaining sequences within the OTU containing the sequence of the applied strain may have represented either native Metarhizium strains or erroneous sequences. The presence of native Metarhizium strains in the pot and field experiments was also detected with the cultivation-dependent approach and genotyping using SSR marker analyses. In Chapter 5 the plasmid pITS2-ART2825, which contained the ITS2 sequence of the applied strain M. brunneum ART2825, was added to a soil sample with low abundance of native Metarhizium strains. Therefore the great number of unique sequences, which did not match the sequence of the plasmid but still clustered within the OTU that contained the sequence of the plasmid were most probable erroneous sequences (Chapter 5). This also demonstrates the importance of OTU clustering to group erroneous sequences to the original sequence. Erroneous sequences are known to occur during NGS, e.g., Illumina MiSeq has an error rate of 0.8 % in raw reads (Quail et al., 2012). Especially if erroneous sequences have a high quality score, it is very hard to detect them (Schirmer et al., 2015) although some filtering tool are available which are rather based on signatures or frequency of occurrence than on quality scores (Bengtsson-Palme et al., 2015; Nikolenko et al., 2013a; Nurk et al., 2013). However, because a simultaneous assessment of soil microbial communities and estimation of the abundance of the applied strain was possible with amplicon sequencing, it is a valuable tool for studies assessing potential effects on soil microbial communities. Amplicon sequencing is less time consuming than the cultivation-dependent approach for assessing the abundance of the applied strain and far more efficient than a cultivation-dependent approach in assessing the soil microbial community since it allows the assessment of the unculturable fraction of soil microbial communities (Singh et al., 2006). Also, amplicon sequencing will become cheaper as the sequencing costs have dropped constantly during the last 20 years (Wetterstrand, 2016) and will probably continue to decrease. 78

GENERAL DISCUSSION

However, amplicon sequencing is not a method for strict quantification because of PCR biases, and for solely monitoring the applied strain the cultivation-dependent method is probably preferable.

ITS2 for phylogenetic resolution of Metarhizium

The ribosomal gene cluster and especially the ITS2 is widely used for taxonomic classification of fungi (Schoch et al., 2012). The ITS2 sequence of the applied strains M. brunneum ART2825 and EAMa01/58-Su were identical. Therefore, the ITS2 was not useful for discrimination of Metarhizium strains as opposed to SSR genotyping which discriminated the two applied strains at all six SSR loci. A multiple sequence alignment of the ITS2 region (310 positions) of 21 sequences including isolates of M. anisopliae , M. brunneum , M. guizhouense , M. majus , M. pingshaense and M. robertsii contained only 10 variable positions (Appendix D 2), and only 2 positions, which were species-specific for M. brunneum , but no species-specific positions for the other five species were detected. The resulting maximum likelihood phylogenetic tree showed a monophyletic clade only for M. brunneum (Figure 6.1). This shows that the ITS2 sequence provides poor discrimination of Metarhizium species. Even more, as the greatest difference between the sequence of the applied strain and any of the other 20 sequences was five positions, which translates into 98.4 % sequence identity, all 21 sequences would have clustered into one single OTU. Therefore, 97 % sequence identity of the ITS2 region did not resolve the phylogenetic structure of these six Metarhizium spp. Because it is known that ITS2 is limited for identification of Metarhizium at the species level and even less so at below species level (Bischoff et al., 2009; Driver et al., 2000; Huang et al., 2005a; Huang et al., 2005b; Schneider et al., 2011), we did not rely on amplicon sequencing for identification of the applied strain, but used isolation on selective medium followed by SSR marker based genotyping. We have chosen the ITS2 because it has been widely used in microbial ecology (Boughner and Singh, 2016) and a large database is available for taxonomic identification of OTUs (Schoch et al., 2012). It is questionable, whether there is a marker region with sufficient resolution for identification of a broad spectrum of soil fungi and simultaneously discriminating indigenous Metarhizium strains from the applied strain. A possible solution would be to perform PCR with ITS2 (for assessment of fungal community structures) and with a target sequence which discriminates Metarhizium strains better than ITS2 and subsequently pool both PCR products for sequencing. Possibly, the seven nuclear intergenic spacer regions, which have been suggested as suitable for discriminating Metarhizium spp. (Kepler and Rehner, 2013), may also include strain specific signatures.

79

GENERAL DISCUSSION

M. majus ARSEF1946 64 M. majus ARSEF1914 M. majus ARSEF1015 55 M. majus ARSEF4566 M. guizhouense ARSEF4303 M. guizhouense ARSEF4321 M. guizhouense ARSEF6238 70 M. guizhouense CBS258.90 M. pingshaense ARSEF3210 M. anisopliae ARSEF7487 64 M. anisopliae ARSEF7450 M. pingshaense CBS257.90 M. pingshaense ARSEF4342 M. robertsii ARSEF2527 M. robertsii ARSEF727 M. brunneum ARSEF2825* 64 M. brunneum ARSEF4179 M. brunneum ARSEF4152 66 M. brunneum ARSEF7524 M. brunneum ARSEF1066 66 M. brunneum BIPESCO5

0.001

Figure 6.1: Unrooted phylogenetic tree based on ITS2 sequences of 21 strains belonging to six different Metarhizium species using the Maximum likelihood method based on Jukes-Cantor model and 1000 bootstraps in MEGA6. * the applied strain M. brunneum ARSEF2825 was used in Chapter 3 and the ITS2 sequence matched exactly with the ITS2 of the applied strain M. brunneum EAMa01/58-Su.

Alternative approaches for detection of the applied strain

Instead of using cultivation on selective medium followed by genotyping with SSR markers to track the applied strain in soil, other approaches, such as restriction-associated DNA sequencing (RADSeq), whole genome sequencing, targeted capture and tracking GFP or barcode labelled strains may be useful (Andrews et al., 2016; Hu and St Leger, 2002; Mamanova et al., 2010). They are based on either using either SSR or single nucleotide polymorphisms (SNPs) for the identification and discrimination of genotypes or directly evaluating a labelled biological control strain. Recently, RADSeq has been developed to simultaneously assess thousands of genetic markers, i.e., SNPs, using high-throughput NGS techniques (Andrews et al., 2016). RADSeq is based on cutting genomic DNA of pure cultures with one or more restriction enzymes, adding adapters to the DNA fragments and using NGS for analysing DNA fragments of appropriate size. Genotypes are based on identified SNPs. The major advantage of RADSeq over SSR marker based genotyping is the availability of enormous number of genetic markers to define and identify genotypes and possibly also species. The latter was not possible using the 41 SSR markers as revealed by cluster analyses using SSR data which did not result in species-specific clades (Chapter 2). Alternatively to RADSeq, the whole genome of single isolates may be sequenced, however, this is more expensive than RADSeq (Andrews et al., 2016). Library preparation and sequencing of one isolate using RADSeq has been

80

GENERAL DISCUSSION estimated to cost 10 – 20 $ with an optimized method (Peterson et al., 2012). In contrast, whole genome sequencing of one isolate may require a whole sequencing run, depending on the size of the genome and the required sequencing coverage, which has been estimated to cost 6’500 $ (Sboner et al., 2011). For instance, for whole genome sequencing of one Metarhizium isolate with a genome size of 40 Mbp (Gao et al., 2011) and using Illumina MiSeq sequencing of yielding 5 to 7.5 x 10 6 high quality sequences with 300 bp length each (as were obtained in our experiment), one run would probably yield a sufficient sequencing coverage. SSR marker analysis, RADSeq as well as whole genome sequencing rely on DNA from single strains obtained by cultivation approaches. Cultivation-dependent approaches are known to be biased and to overcome these biases cultivation-independent direct assessment of the applied strain may be implemented. A cultivation-independent genotyping tool using NGS is targeted (probe-based) capture in which several pre-selected genomic regions of interest are enriched and sequenced (Jones and Good, 2016; Mamanova et al., 2010). The greatest challenge in this approach is the design of appropriate capture probes. Suitable sequences for the design of species-specific capture probes for Metarhizium may be revealed by sequencing genetically variable regions, such as seven nuclear intergenic regions (Kepler and Rehner, 2013). Strain-specific probes may be found in SSR marker regions (Chapter 2) or may be designed based on SNPs detected using, e.g., RADSeq. Another possible approach would be to assess the presence and/or abundance of the applied biological control strain directly using GFP labelled (Hu and St Leger, 2002) or barcode- tagged biological control strains. However, both techniques include genetically modified organisms and therefore applications are restricted. Nevertheless, genotyping based on SSR markers will continue to be a valuable tool for discriminating the applied strain from native Metarhizium strains, because of lower costs than NGS and of wider availability of equipment.

6.2 Potential effects of M. brunneum strains on soil microorganisms

Do inundative applications of M. brunneum affect soil microbial communities?

In the non-target effect studies performed in pots soil fungal and prokaryotic communities were exposed to M. brunneum ART2825 applied in different formulations (FCBK, Fcap ) and as unformulated fungal spores or to M. brunneum EAMa01/58-Su in form of FCBK (Chapter 3 and 4). In the field experiment M. brunneum ART2825 was applied as FCBK, F cap and Fgran (Chapter 3). All fungal treatments in both pot experiments and the FCBK treatment in the field experiment resulted in an increased abundance of the applied strain and therefore exposure of soil microbial communities was confirmed. Pairwise ANOSIM R-values revealing the comparisons of soil microbial community structures between treated and untreated pots at different sampling time points in the Agriotes - and the Diabrotica -pot experiment are compared and summarized in Table 6.1. In the Agriotes -pot experiment fungal communities changed slightly after the application of FCBK and F cap but not after the application of fungal spores alone. In the Diabrotica -pot experiment as well as in the field study none of the fungal products caused any changes in the fungal community structures. In addition to the small changes caused by FCBK and F cap in the Agriotes -pot experiment, fungal communities responded to the addition of barley kernels (the carrier material of FCBK), which suggest that the effects may have rather resulted from the barley kernels than from the fungal strain. This reinforces the conclusion that applications of M. brunneum did not cause effects on soil fungal communities in our experiments. The prokaryotic community structures were not affected by any fungal treatments in any of the three experiments, except for marginal effects by fungal spores. However, prokaryotic soil microorganisms reacted on the application of the treatments including garlic capsules ( Agriotes pot experiment)

81

GENERAL DISCUSSION and the addition of BK, which is the non-fungal component of FCBK. Our experiments suggest that applications of M. brunneum did not cause effects on soil prokaryotic communities.

Table 6.1: Pairwise ANOSIM comparing the soil microbial communities between treated and untreated pots at different sampling time points (postapplication 1 & postapplication 2) and in untreated pots between preapplication (week 0) and postapplication 1 or 2 in the Agriotes and the Diabrotica pot experiment. Mean of Metarhizium Fungi Prokaryota CFU g -1 soil dry weight Factor 1 Sampling time point Agriotes Diabrotica Agriotes Diabrotica Agriotes Diabrotica Postapplication 1 17‘596 64‘521 0.51* -0.09 0.14* 0.23 FCBK 2 Postapplication 2 7‘159 23‘993 0.51* 0.09 0.02 0.11 Postapplication 1 60 2‘917 0.55* 0.14 -0.09 0.2 BK 3 Postapplication 2 76 2‘457 0.46* -0.11 -0.01 0.42* Postapplication 1 11‘201 NA 0.42* NA 0.2 NA 4 Fcap Postapplication 2 9‘132 NA 0.28* NA 0.06 NA Postapplication 1 5‘569 NA -0.03 NA 0.25* NA 5 Fpowd Postapplication 2 5‘944 NA 0.02 NA 0.03 NA

FCBK 2 & Postapplication 1 6‘449 NA 0.09 NA 0.82* NA 6 Gcap Postapplication 2 5‘220 NA 0.34* NA 0.6** NA Postapplication 1 7‘856 NA -0.01 NA 0.56* NA 4 6 Fcap & G cap Postapplication 2 8‘033 NA 0.46* NA 0.57* NA Postapplication 1 60 NA 0.03 NA 0.9* NA 6 Gcap Postapplication 2 0 NA 0.03 NA 0.59* NA Postapplication 1 60 1‘680 0.109 0.13* 0.65* 0.57* Untreated 7 Postapplication 2 120 923 0.45* 0.51* 0.89* 0.87*

*… p-value < 0.05 NA … not assessed 1… treatments compared with untreated pots at respective sampling time point and Untreated compared between preapplication and sampling timepoint 2… fungal colonized barley kernels 3… barley kernels 4… fungal capsules 5… fungal spores 6… garlic capsules

There are several possible explanations for why we did not detect any effects on soil microbial community structures. I) M. brunneum did not interact with other soil microorganisms or was less competitive, II) natural fluctuations in the soil microbial community structures exceeded effects of interactions between M. brunneum and soil microorganisms and III) effects may have been below the resolution of the methods used. I) Only a few cultivation based studies are available on direct interaction of Metarhizium and fungi or bacteria by assessing growth inhibition on nutrient media plates (reviewed in Jaronski, 2007). These studies revealed 82

GENERAL DISCUSSION

varying results depending on the isolate studied and nutrient medium used. Nutrient media provide an artificial environment for microorganisms and therefore results may not reflect conditions in soil. However, a more general observation about the behaviour of Metarhizium in soil is that it may experience growth inhibition in non-sterile soils, in which CFU abundance of Metarhizium declined or remained constant, as compared to sterile soils, in which CFU abundance of Metarhizium increased, while infectivity of the spores on insects was unaffected (Jaronski, 2007; Walstad et al., 1970). This fungistatic effect may be due to direct inhibition by soil microorganisms or due to a weak competitiveness of Metarhizium (for nutrient) compared to soil indigenous and saprotrophic microorganisms (Jaronski, 2007; Meyling and Eilenberg, 2007). In our experiments we observed that after the mass application of M. brunneum its density either remained constantly high or declined within 4 months after application, however, densities of M. brunneum in soil may have not been monitored in sufficiently close time intervals. Nevertheless, the observed decrease in Metarhizium CFU g -1 soil dry weight over time supports the idea that M. brunneum experienced fungistatic effects in soil. Either these fungistatic effects or the lack of interactions between the applied strains and native soil microorganisms at all may have been the reason why no effects on soil microorganisms in the pot and field experiments were observed. II) Another reason why no effects of M. brunneum on soil microorganisms were observed could be that natural fluctuation of the soil microbial communities exceeded effects of the applied strain, and therefore other factors were more important in shaping the soil microbial communities than the addition of high densities of M. brunneum propagules. Natural fluctuation of soil microbial communities probably develop due to a combination of different abiotic factors, such as temperature and humidity, and biotic factors, such as plants and animals (Lauber et al., 2013; Wieland et al., 2001). The natural fluctuations of soil microbial communities in our experiments may be revealed by the variation among samples of untreated pots / plots and by their variation over the time course of the experiments (~ 4 months). A statistical method to analyse homogeneity of dispersion (PERMDISP; Anderson, 2006) revealed that dispersions of samples reflecting the microbial community structures in untreated pots differed significantly between the Agriotes - and the Diabrotica -pot experiment. This was also revealed in the PCO (Figure 4.4), and may be caused by the greater variability of climatic conditions in the semi-open greenhouse of the Diabrotica -experiment than the closed greenhouse of the Agriotes -pot experiment. The greater variability in the structure of soil microbial communities in the Diabrotica -pot experiment compared to the Agriotes -pot experiment may provide an explanation for why effects of FCBK and BK on fungal community structures were detected in the Agriotes-pot experiment, but not in the Diabrotica -pot experiment (Table 6.1). Continuous changes of the soil fungal and prokaryotic communities in untreated samples over time were observed in both the Agriotes - and the Diabrotica -pot experiment but not in the field experiment. In both pot experiments insects (Agriotes larvae and Diabrotica eggs) were added to the soil, and they might have influenced soil microbial communities in multiple ways, such as feeding on plants (Dematheis et al., 2012) and depositing excrements containing their gut microflora or leaving remains of eggs, larval cuticles or pupal skins during changes of life stages. The potato and maize plants grown in the soil could have also influenced soil microbial communities for instance by releasing root exudates (Cavaglieri et al., 2009). In the Diabrotica -pot experiment the growth of maize may have induced an increase in abundance of the fungal phylum Glomeromycota and changes of its community composition (Chapter 4), because Glomeromycota form arbuscular mycorrhiza with maize (Schüβler et al., 2001). Abiotic factors such as temperature and humidity 83

GENERAL DISCUSSION

may have also contributed to the changes of soil microbial communities over time (Auffret et al., 2016; Chen et al., 2007). Data on temperature during the course of the experiment was only available for the Diabrotica -pot experiment. Temperature changed in the Diabrotica -pot experiment and daily mean soil temperature was 23.2 °C on average and temperature ranged between 12.2 and 41.5 °C. Unfortunately, we cannot draw conclusions on which factors were most dominant in our experiments. Additional treatments without plants and / or insects would have been necessary to control for these factors, however, these treatments would not represent realistic conditions for biocontrol. The greenhouse provided some protection to climatic changes, however, constant temperature and humidity were not achieved. In the field spatial changes were observed in fungal and prokaryotic communities. These changes may be caused by differences in soil factors, e.g., pH, organic matter content, texture or humidity, which have shown to affect soil microbial communities (Cao et al., 2016b; Naveed et al., 2016; Rousk et al., 2010). Edaphic factors, which were assessed in the present experiments (pH, soil texture), were not assessed in high enough resolution and soil water content did not reveal significant spatial differences. Therefore, we cannot draw any conclusions whether these factors caused the observed spatial differences in soil microbial community structures. The aim of the field-experiment was to study potential effects of M. brunneum in systems which are close to agricultural practice and we found that spatial differences were more important in shaping soil microbial community structures than any applications in the field. III) Possibly no effects on soil microbial communities were detected due to methodological constraints such as, effects were below the threshold of detection of amplicon sequencing, or the presence of one highly abundant sequence impeded the assessment of changes in native soil microbial communities. Determination of the threshold of amplicon sequencing has been performed for single strains (de Boer et al., 2015), however, the limit of detection of changes in soil microbial community structures has not been investigated thoroughly. Estimating the threshold of detecting changes in soil microbial communities is not possible if no effects are detected. If there were effects, the threshold of detection could be assessed by adding increasing amounts of the applied strains to soil until effects on soil microbial communities are reliably detected. Although assessing the limit of detection of changes of soil microbial communities may be a challenging task because it probably differs for the different fungal products and soils. Alternatively as an approximation, artificial soil microbial communities could be assembled in different combinations and abundances, and this may allow the assessment of how much change in a community is needed to obtain detectable effects using amplicon sequencing. A similar approach was performed to assess the detection limit of changes of soil microbial communities using TRFLP (Hartmann and Widmer, 2008). Another methodological constraint in the assessment of changes in soil microbial communities may arise due to the preferentially amplification of the target sequence in the PCR step in the presence of one highly abundant sequence (Polz and Cavanaugh, 1998; Schirmer et al., 2015). We attempted to answer this question by designing an experiment in which different amounts of plasmid pITS2-ART2825, that contained the ITS2 region of the applied strain, were added to the same soil extract (Chapter 5). Subsequently, soil microbial community structures were analysed in exactly the same way as for the soil samples of the pot and field experiments (Chapter 3 and 4). The statistical analyses revealed that the addition of pITS2-ART2825 did not impede the assessment of the soil microbial communities. Therefore, it is unlikely that the assessment of changes in soil microbial communities after the application of M. brunneum was hampered by the presence of the applied strain in Chapter 3 and 4. However, adding even 84

GENERAL DISCUSSION

higher concentrations of pITS2-ART2825 may have impeded the analysis, as the highest concentration of added plasmid pITS2-ART2825 showed already slight but insignificant effects on the assessment of fungal communities (i.e., decreased OTU richness, Chapter 5).

Sequencing errors and limitations of amplicon sequencing

It is known that NGS techniques include errors, and the error rate for Illumina MiSeq platform, which was used in our experiments, was 0.8 % in a study which compared different NGS platforms (Quail et al., 2012). In order to obtain high quality data, quality control is crucial in bioinformatical analyses of NGS data. Quality control included multiples steps, i.e., removal of sequences which yielded a too short overlap after merging the complementary reads (Edgar and Flyvbjerg, 2015), removal of sequences in which substitution errors were detected (Nikolenko et al., 2013a; Nurk et al., 2013), removal of sequences with erroneous primer sequences (Martin, 2011), removal of low quality sequences based on maximum expected error below 1 (Edgar and Flyvbjerg, 2015), removal of sequences which did not include signatures specific for the marker region (Bengtsson-Palme et al., 2015; Bengtsson-Palme et al., 2013), removal of sequences which represented chimera and removal of sequences which occurred only once (singletons) (Edgar, 2013). These steps were performed using different programs and functions consecutively connected (pipeline). A very similar version of this pipeline was also used in other studies (Frey et al., 2016; Moll et al., 2017; Rime et al., 2016). After quality control 43 % of the eukaryotic (including non-fungal sequences) and 49 % of the prokaryotic sequences were removed from the dataset of the Agriotes and the Diabrotica -pot experiment (Chapter 3 and 4). In the sequencing run including the Agriotes -field experiment and the plasmid addition experiment (Chapter 5) 35 % of the fungal and 66 % of the prokaryotic sequences were removed due to low quality. These percentages are quite high but still comparable to the commonly found 20 – 40 % loss of sequences after quality control (Lindahl et al., 2013). The step of our pipeline in which most of the fungal (19 - 22 %) and prokaryotic (34 - 52 %) sequences were lost was filtration based on expected error. Dominant errors of Illumina MiSeq have been found to be substitution type miscalls (Schirmer et al., 2015). Quality filtration removes erroneous sequences with low quality scores, but erroneous sequences with high quality scores remain in the dataset. For instance, an enormous amount of most likely erroneous sequences which passed quality filtration were detected within the OTU containing the sequence of the applied strain by comparison to the control samples (Chapter 5). Because quality scores predict only sequencing errors but not PCR errors, these errors were most likely introduced in the PCR. After quality control sequences were clustered into OTUs and compared to curated databases, i.e., a custom-made GENBANK and the UNITE database for fungal sequences and the GREENGENES database for prokaryotic sequences (Abarenkov et al., 2010; Benson et al., 2015; DeSantis et al., 2006; McDonald et al., 2012). OTUs not assigned to Fungi, Bacteria or Archaea were removed. On average across all experiments 90 % of the fungal and 96 % prokaryotic OTUs were classified at phylum level, but only 29 % of the fungal and 1 % of the prokaryotic sequences were assigned at species level. These low percentages of OTUs assigned down to species level suggest that databases are not nearly complete. In order to assess the severity and extent of potential effects of M. brunneum on soil microbial communities, it would be necessary to assess soil microbial communities at species level, because this would better allow to assess their ecological role at least for some taxa. However, in our experiments most OTUs which were significantly associated to factors were not classified at species level, such as OTU 1, OTU 13, OTU 45 and OTU 291 which were classified at genus to kingdom level (Chapter 3). An exception, for instance, was the phylum Glomeromycota, which includes only arbuscular mycorrhizal fungi, and therefore classification 85

GENERAL DISCUSSION at phylum level already provides information about the functional potential. Classification success at species level probably depends on the quality and completeness of the database, and because these databases are constantly growing, reassessing taxonomic classification at a later time point would probably increase classification success at species level and thereby improve the informative content of the present studies.

Alternative future approaches

An alternative method to amplicon sequencing would be soil metagenome sequencing which aims at sequencing all genomes present in a soil sample. This would produce vast amounts of information on the composition as well as on the functional potential of soil microbial communities (Rastogi and Sani, 2011). However, this approach is still hampered by the difficulty to obtain a high enough sequence coverage for all genomes present in a soil sample (including rare ones) and the costs that are related to the effort to obtain high sequence coverage (Ni et al., 2013). With the development of third generation sequencing some of the limitations of amplicon sequencing will probably be overcome (Shokralla et al., 2012). For instance, longer reads will be obtained (e.g. Pacific Biosciences), the PCR amplification will be omitted due to single cell sequencing and thereby avoiding PCR biases and quantification will be improved (Shokralla et al., 2012). Additionally the multi-omics approach, which includes the parallel assessment of the metagenome, metatranscriptome, metaproteome and metametabolome, seems to be promising for future soil microbial community assessment, because it allows to assess the function and composition of soil microbial communities based on DNA, RNA, proteins and metabolites (Franzosa et al., 2015).

Comparison to other hypocralean entomopathogenic fungi

A growing number of studies on potential effects of the application of different Metarhizium strains, including our experiments, detected no or transient effects on soil microorganisms (Hu and St Leger, 2002; Kirchmair et al., 2008). It is not possible to rule out that other Metarhizium strains do affect soil microbial communities as ecological aspects (e.g. virulence against insects, production of destruxins, endophytic growth) may vary among Metarhizium strains and species (Cao et al., 2016a; Golo et al., 2014). However, studies assessing potential effects of Beauveria strains (Hirsch et al., 2013; Rai and Singh, 2002; Schwarzenbach et al., 2009), which is an entomopathogenic fungus belonging to the same order as Metarhizium (i.e., Hypocreales) and has a similar ecology (e.g. infection process) revealed no effects on soil microbial communities. This suggests that applications of entomopathogenic fungi of the order Hypocreales may not affect soil microbial communities, although further confirmation by assessing potential effects on soil microbial community structures of other hypocralean entomopathogenic fungi is needed.

6.3 Efficacy of biological control using M. brunneum

Previous studies have shown the potential of M. brunneum to control the insect pests A. obscurus (Eckard et al., 2014; Kölliker et al., 2011) and Diabrotica v. virgifera (Pilz et al., 2009). One of the major goals of the EU-Project INBIOSOIL, which funded the present thesis, was to develop biological control of these two pests using novel formulations based on alginate capsules, exploiting synergistic effects between entomopathogenic fungi, nematodes and natural substances as well as testing strategies, such as attract and kill (Brandl et al., 2017). In our experiments, a biocontrol effect was observed only in the pot experiment to control A. obscurus , but not in the Agriotes -field and Diabrotica -pot experiment (Chapter 3 and 4). In the Agriotes -pot experiment significantly lower numbers of larvae were obtained after the application of F cap , FCBK, and both in combination with G cap . However, 86

GENERAL DISCUSSION significantly fewer damaged potatoes were obtained only after the applications of the traditional product FCBK (Figure 3.2). The difference in efficacy among the fungal products may be related to the slightly higher average abundances of M. brunneum ART2825 in the pots treated with FCBK (Figure 3.1). The strategy of combining garlic, which was shown to repel and reduce movement of A. obscurus larvae, and the fungal products did not improve the efficacy of M. brunneum ART2825. In contrast, the attract and kill strategy using CO 2 as an attractant to lure Agriotes larvae to the capsules containing M. brunneum resulted in an efficacy of 37 – 75 % in the field (Brandl et al., 2017). Using an attractant seemed to be the superior strategy, compared to a repellent, in controlling Agriotes larvae. Another reason for the differences in efficacy of M. brunneum ART2825 between our studies and the studies conducted by Brandl et al. (2017) might be a difference in the species composition of Agriotes larvae. It is known that species composition may vary among fields, and that Agriotes species differ in susceptibility to M. brunneum ART2825 (Ritter and Richter, 2013). The lack of biocontrol in the Diabrotica -pot experiment may have a methodological reason, related to infestation with D. v. virgifera , and a biological reason, related to the nature of D. v. virgifera . First, varying number of eggs were added to each pot due to the sand / egg mixture in which the eggs were obtained from the rearing facility. This may have impeded the detection of any biocontrol effects due to the varying abundance of the pest among pots. Secondly, D. v. virgifera larvae spend a part of their life inside the roots of maize and therefore the contact time between the fungus and the insect might have been too short to cause infection. Although, the biological control options for both pests need further improvement, for instance by implementing other IPM strategies, the increased abundance of the applied biological control strains M. brunneum ART2825 and M. brunneum EAMa01/58-Su in soil after the applications allowed the assessment of potential effects on soil microbial communities structures.

6.4 Overall conclusions

1) Isolation of Metarhizium CFU followed by genotyping using SSR marker analyses allowed to assess the abundances of Metarhizium spp. in soil, to identify the applied strains M. brunneum ART2825 and M. brunneum EAMa01/58-Su and to discriminate them from native Metarhizium isolates. Therefore, this approach was valuable for confirming inoculation success and exposure of soil microbial communities to the applied strains. Although estimating exposure of soil microbial communities to the applied strains was possible using amplicon sequencing, isolation of Metarhizium CFU followed by genotyping with SSR markers was the preferable method, because the ITS2 amplicon did not allow the discrimination among Metarhizium species and strains and amplicon sequencing is not a strictly quantitative method. 2) Amplicon sequencing of ribosomal markers allowed the assessment of changes of soil microbial communities after application of different treatments and of temporal and spatial changes of soil microbial communities. PCR and sequencing biases due to the presence of one highly abundant sequence were ruled out, and therefore it is unlikely that the presence of one highly abundant sequence constrained the assessment of changes in the soil microbial communities. A high number of erroneous sequences was obtained with amplicon sequencing and even after several steps of quality control a substantial number of erroneous sequences with high quality scores remained in the dataset. Additionally, most OTUs were identified at phylum level while, especially in the prokaryotic dataset, only low numbers of OTUs were identified at species level. Although, obvious constraints of amplicon sequencing were detected it is the

87

GENERAL DISCUSSION

most suitable technique at the moment, for the required sequencing depth and the large number of samples and replicates required in experiments as performed in this study. 3) Effects of edaphic factors in shaping the spatial and temporal differences of soil microbial community structures were similar or greater than any treatment-effects detected by applying M. brunneum in different formulations or as unformulated fungal spores. Therefore, it is unlikely that applications of M. brunneum ART2825 and M. brunneum EAMa01/58-Su adversely affect soil microbial communities. In the Agriotes - pot experiment treatment with FCBK affected fungal communities to the same extent as the carrier material and therefore effects probably resulted from the application of the latter. Effects of common treatments in all three experiments on soil microbial community structures were not consistently detected, possibly due to differences in edaphic factors, in types of plants and insects, in soil microbial community composition and in the degree of natural fluctuation of soil microbial community structures.

6.5 Perspectives

Identification of applied fungal strains using SSR analyses

In Chapter 4 the genotype of the most closely related native Metarhizium strain differed only by the allele size of one repeat of a single SSR marker (locus) from the applied biological control strain M. brunneum ART2825. Additional SSR markers of the 41 available ones could be used to corroborate this difference and possibly identify more native genotypes. However, as the most polymorphic SSR marker available for M. brunneum was included, it is questionable if this would change the outcome to an extent, which would be relevant for the purpose of confirming application success and exposure of soil microorganisms to the applied strains. The amount of markers could be drastically increased by using RADSeq, but at the same time biases, such as sequencing errors may be introduced, RADSeq requires more starting DNA and is still more expensive (Andrews et al., 2016). Therefore, SSR marker analysis will continue to be a valuable tool, at least in the near future.

Methodological aspects regarding effects on soil microbial communities

Amplicon sequencing of ribosomal markers allowed the assessment of soil microbial communities even in the presence of one highly abundant sequence. However, low taxonomic classification success at the species level, especially for prokaryotic OTUs, occurred, which constrained interpretation of observed effects. As the database of verified sequences of ribosomal markers is growing it would be advisable to repeat taxonomic classification of the sequencing data in the future. This might provide improved insight into functions of certain OTUs, which changed upon treatments or over time. Also, the detection limit of changes of soil microbial communities should be assessed in order to gain further understanding on the severity of effects. This could be done by using artificial or model communities containing the same species in different concentrations or combinations (Hartmann and Widmer, 2008). Sequencing and/or PCR errors were frequent and it is not in detail understood, to which extent they influenced the results. Therefore, single molecule sequencing using for instance the PacBio RS SMRT system, which omits the PCR step and therefore may reduce errors in the sequence data obtained (Rhoads and Au, 2015; Shokralla et al., 2012). Lastly, it would be interesting to study potential effects on functions provided by soil microorganisms. Changes in functional diversity of soil microorganisms would provide additional information to evaluate severity of effects found. Furthermore, this would help to determine links between taxonomic and functional diversity, which are currently not well understood. This could be performed for instance by assessing

88

GENERAL DISCUSSION metagenomes and metatranscriptomes, which ideally provide information on all transcribed gene and products of all soil organisms in a soil sample.

Experimental aspects regarding effects on soil microbial communities

In order to better understand the natural fluctuation of soil microbial communities, which were observed in both pot experiments and the field experiment (Chapter 4 and 5), abiotic soil factors, such as water content, nutrient availability, soil texture, and pH could be assessed in all soil samples. This applies particularly to the field samples, because for the pot experiments, soil was thoroughly mixed before it was added to the pots and therefore it should have similar properties in all pots. Assessing environmental factors, however, is not necessary to answer the main research question of projects assessing potential effects of M. brunneum strains on soil microbial communities. However, it may be of assistance when searching for explanations of differences detected among different experiments performed at different times or locations as well as when using different soils. In order to assess potential effects under conditions of reduced natural fluctuations in the pot experiments, additional control treatments could be used, e.g., without insects and / or plants. Even though this would differ greatly from conditions found in biological control experiments it would provide information on the magnitude of effects of other factors. All three experiments included one single application of M. brunneum . In agricultural practice, however, several applications may be performed in one year or in several consecutive years in order to increase efficacy (Kaaya, 2000; Rauch et al., 2017; Wraight and Ramos, 2015). Possibly, M. brunneum densities would have accumulated over several applications and seasons and may have resulted in detectable effects on soil microbial communities. Although two different M. brunneum strains were applied, and both did not affect soil microbial communities it is not possible to exclude that applications of other M. brunneum strains and other Metarhizium species might affect soil microbial communities. However, there is a growing body of literature showing that mass applications of hypocralean fungi have not affected soil microbial communities. But recently it has been discovered that some Metarhizium strains also show an endophytic lifestyle and therefor potential effects of applied strains on the microbiome of plants may be of interest, also with regard to potential protection of plants form insect pests, suppression of plant pathogenic microorganisms or undesired effects on plant-beneficial microorganisms.

89

APPENDIX

Appendix

A) Supporting information to Chapter 2

Table Appendix A 1: Sixty-five Metarhizium strains representing 11 Metarhizium species included in the present study and GenBank accession numbers for their 5’ elongation factor 1-alpha sequences given in parentheses.

Species Strains M. acridum ARSEF 0324 a (EU248844), ARSEF 3391 (EU248873), ARSEF 5736 (EU248878), ARSEF 5748 (EU248879), ARSEF 6421 (EU248883), ARSEF 6592 (EU248886), ARSEF 6597 (EU248887), ARSEF 6600 (EU248888), ARSEF 7486 (EU248845) M. anisopliae ARSEF 6347 (EU248881), ARSEF 7450 (EU248852), ARSEF 7487 (DQ463996), ART Ma2062 b (KR706492) M. brunneum ARSEF 0988 (EU248890), ARSEF 2107 (EU248855), ARSEF 3826 (EU248874), ARSEF 4152 (EU248853), ARSEF 4179 (EU248854), ARSEF 5198 (EU248876), ARSEF 5625 (EU248877), ARSEF 6120 (EU248880), ARSEF 6392 (EU248882), ARSEF 6477 (EU248885), ART 2825 (KR706488), BIPESCO 5 d (KR706489), ARSEF 7524 (=ART 714, KR706491) M. flavoviride Mf98SS e (KR706493), MfAP023PL f (KR706494), MfSWE7.2.3.2 e (KR706495), CBS 218.56 c (=ARSEF 2133, DQ463988) M. frigidum ARSEF 4124 (DQ463978) M. globosum ARSEF 2596 (EU248846) M. guizhouense ARSEF 4303 (EU248859), ARSEF 4321 (EU248860), ARSEF 4604 (EU248894), ARSEF 5714 (EU248856), ARSEF 6238 (EU248857), ARSEF 7420 (EU248892), ARSEF 7502 (EU248861), ARSEF 7507 (EU248858), CBS 258.90 (EU248862). M. lepidiotae ARSEF 4587 (EU248893), ARSEF 4660 (EU248895), ARSEF 7412 (EU248864), ARSEF 7488 (EU248865) M. majus ARSEF 0473 (EU248875), ARSEF 0978 (EU248889), ARSEF 1015 (EU248866), ARSEF 1858 (EU248872), ARSEF 1914 (EU248868), ARSEF 1946 (EU248867), ARSEF 2808 (EU248871), ARSEF 4566 (EU248869), ARSEF 7505 (EU248870) M. pingshaense ARSEF 3210 (DQ463995), ARSEF 4342 (EU248851), ARSEF 7929 (EU248847), CBS 257.90 (EU248850) M. robertsii ARSEF 0727 (DQ463994), ARSEF 2575 (KR706486), ARSEF 4628 (KR706487), ARSEF 4739 (EU248848), ARSEF 6472 (EU248884), ARSEF 7501 (EU248849), ARSEF 7532 (= ART 500, KR706490) a ARSEF strains obtained from USDA-ARS Collection of Entomopathogenic Fungal Cultures, Ithaca, NY, USA. b ART strains obtained from Agroscope Reckenholz-Tänikon collection, Zurich, Switzerland. c CBS strains obtained from the Centraalbureau voor Schimmelcultures collection, Utrecht, Netherlands. d BIPESCO 5 obtained from Dr. Hermann Strasser, Leopold-Franzens University Innsbruck, Austria. e isolated and provided by Bernhardt Steinwender, University of Copenhagen, Denmark. f isolated and provided by Cezary Tkaczuk, Siedlce University of Natural Sciences and Humanities, Poland.

91

APPENDIX

Table Appendix A 2: Hit-table of nucleotide BLAST searches of 41 SSR loci in the genome assemblies of M. brunneum ARSEF 3297, M. anisopliae ARSEF 549 and M. robertsii ARSEF 23.

M. brunneum ARSEF 3297 M. anisopliae ARSEF 549 M. robertsii ARSEF 23 Locu GenBank Scaff % subject subject Scaff % subject subject Scaff % subject subject s accession # Subject ID Subject ID Subject ID old identity starts stops old identity starts stops old identity starts stops gi|743641680|gb|AZNG010 gi|743637694|gb|AZNF010 gi|734840523|gb|ADNJ020 2049 EF670679 12 99.252 168309 168709 2 96.774 2527238 2526836 3 97.506 2539283 2539679 00012.1| 00002.1| 00003.1| gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 2054 EF670680 1 91.46 5481914 5482275 3 91.473 1798709 1798457 2 91.949 1838400 1838168 00001.1| 00003.1| 00002.1| gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 2054 1 98.444 4342036 4341780 3 90 2961716 2961965 2 88.519 2988968 2989211 00001.1| 00003.1| 00002.1| gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 2054 3 97.143 2961996 2962065 2 95.714 2989240 2989309 00003.1| 00002.1| gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 2055 EF670681 1 96.943 4841056 4840828 3 96.482 2322357 2322159 2 97.561 2336592 2336388 00001.1| 00003.1| 00002.1| gi|743643255|gb|AZNG010 gi|743631134|gb|AZNF010 gi|734840523|gb|ADNJ020 2056 EF670682 7 97.525 463462 463663 11 95.455 734600 734795 3 97.475 190633 190438 00007.1| 00011.1| 00003.1| gi|743643255|gb|AZNG010 gi|734838369|gb|ADNJ020 2057 EF670683 7 98.276 1568293 1568350 not found 14 95.161 441203 441018 00007.1| 00014.1| gi|743643255|gb|AZNG010 gi|734838369|gb|ADNJ020 2057 EF670683 7 95.161 1568338 1568519 14 100 441251 441205 00007.1| 00014.1| gi|743634730|gb|AZNG010 gi|743630110|gb|AZNF010 gi|734840067|gb|ADNJ020 2060 EF670684 22 98.619 191095 190734 14 96.45 682909 683245 4 95.726 370517 370172 00022.1| 00014.1| 00004.1| gi|743645682|gb|AZNG010 gi|743628612|gb|AZNF010 gi|734839658|gb|ADNJ020 2063 EF670685 3 93.443 1720445 1720214 21 96.386 311137 311385 5 97.131 2612636 2612875 00003.1| 00021.1| 00005.1| gi|743646596|gb|AZNG010 gi|743635620|gb|AZNF010 gi|734842655|gb|ADNJ020 2064 EF670686 2 94.421 2087177 2086953 4 88.936 2677496 2677716 1 92.34 3856447 3856677 00002.1| 00004.1| 00001.1| gi|743643735|gb|AZNG010 gi|743633372|gb|AZNF010 gi|734839229|gb|ADNJ020 2065 EF670687 6 98.759 550900 550177 7 95.902 520579 521305 7 96.594 1552765 1553497 00006.1| 00007.1| 00007.1| gi|743647662|gb|AZNG010 gi|743630392|gb|AZNF010 gi|734838983|gb|ADNJ020 2069 EF670688 1 98.3 6640530 6640178 13 92.758 270982 271340 8 93.696 1104074 1103733 00001.1| 00013.1| 00008.1| gi|743632811|gb|AZNF010 gi|734838848|gb|ADNJ020 2070 EF670689 not found 8 93.578 663320 663428 9 93.578 769427 769535 00008.1| 00009.1| gi|743645682|gb|AZNG010 gi|743634851|gb|AZNF010 gi|734839658|gb|ADNJ020 2077 EF670690 3 98.384 706279 705785 5 93.075 1383138 1382663 5 95.723 1435166 1434686 00003.1| 00005.1| 00005.1| gi|743641680|gb|AZNG010 gi|743637694|gb|AZNF010 gi|734840523|gb|ADNJ020 2089 EF670691 12 97.658 933945 933523 2 94.226 1762411 1762839 3 93.303 3307967 3307543 00012.1| 00002.1| 00003.1| gi|743645682|gb|AZNG010 gi|743634851|gb|AZNF010 gi|734839658|gb|ADNJ020 2097 EF670692 3 98.714 849896 850206 5 95.597 1526886 1527203 5 94.937 1579050 1579359 00003.1| 00005.1| 00005.1| gi|743644285|gb|AZNG010 gi|743631654|gb|AZNF010 gi|734842655|gb|ADNJ020 2098 EF670693 5 95.111 1416906 1416690 10 90.625 846579 846364 1 95.595 858979 859205 00005.1| 00010.1| 00001.1| gi|743633273|gb|AZNG010 gi|743631134|gb|AZNF010 gi|734838719|gb|ADNJ020 2099 EF670694 24 91.618 61817 61476 11 91.813 284723 285058 10 95.195 217535 217203 00024.1| 00011.1| 00010.1| gi|743647662|gb|AZNG010 gi|743630392|gb|AZNF010 gi|734841079|gb|ADNJ020 2103 EF670695 1 97.101 5980264 5979989 13 97.436 948544 948816 2 96.324 3487782 3487518 00001.1| 00013.1| 00002.1| gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 2108 EF670696 1 98.133 5044530 5044900 3 93.478 2525475 2525834 2 92.513 2539512 2539879 00001.1| 00003.1| 00002.1| gi|743646596|gb|AZNG010 gi|743635620|gb|AZNF010 gi|734842655|gb|ADNJ020 2109 EF670697 2 95.393 54808 55176 4 93.103 694072 694419 1 94.737 5893207 5892869 00002.1| 00004.1| 00001.1| gi|743641680|gb|AZNG010 gi|743637694|gb|AZNF010 gi|734840523|gb|ADNJ020 2224 EF670698 12 96.497 134237 134788 2 96.989 2561419 2560955 3 97.863 2505043 2505510 00012.1| 00002.1| 00003.1| gi|743637694|gb|AZNF010 gi|734840523|gb|ADNJ020 2224 2 90.426 2561489 2561401 3 92.784 2504972 2505064 00002.1| 00003.1| 92

APPENDIX

gi|743645682|gb|AZNG010 gi|743632270|gb|AZNF010 gi|734839658|gb|ADNJ020 2269 EF670699 3 98.442 2730940 2731260 9 93.603 980023 980473 5 97.241 3490943 3491232 00003.1| 00009.1| 00005.1| gi|743645682|gb|AZNG010 gi|734839658|gb|ADNJ020 2269 EF670699 3 100 2730841 2730942 5 98.101 3490797 3490954 00003.1| 00005.1| gi|743644285|gb|AZNG010 gi|743628378|gb|AZNF010 gi|734842655|gb|ADNJ020 2274 EF670700 5 98.879 412060 411619 23 89.286 172265 172674 1 95.067 1903769 1904197 00005.1| 00023.1| 00001.1| gi|743646596|gb|AZNG010 gi|743630735|gb|AZNF010 gi|734842655|gb|ADNJ020 2279 EF670701 2 95.926 2400479 2401002 12 93.148 306416 306933 1 93.554 3542470 3541952 00002.1| 00012.1| 00001.1| gi|743646596|gb|AZNG010 gi|743630735|gb|AZNF010 gi|734842655|gb|ADNJ020 2283 EF670702 2 97.566 2660135 2660578 12 97.566 566234 566680 1 93.421 3282995 3282557 00002.1| 00012.1| 00001.1| gi|743644877|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734840067|gb|ADNJ020 2287 EF670703 4 98.475 1985720 1986175 1 94.915 4793334 4793801 4 95.492 1788847 1788363 00004.1| 00001.1| 00004.1| gi|743644877|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734840067|gb|ADNJ020 2292 EF670704 4 98.491 1535695 1535233 1 96.963 4343078 4342619 4 97.831 2238098 2238557 00004.1| 00001.1| 00004.1| gi|743642584|gb|AZNG010 gi|743635620|gb|AZNF010 gi|734842655|gb|ADNJ020 2296 EF670705 9 93.048 990356 990170 4 91.892 635859 635677 1 94.565 5951534 5951715 00009.1| 00004.1| 00001.1| Ma0 gi|743643735|gb|AZNG010 gi|743628871|gb|AZNF010 gi|734839229|gb|ADNJ020 AY842937 6 93.985 1448748 1448484 19 95.865 221525 221790 7 92.424 649341 649602 97 00006.1| 00019.1| 00007.1| Ma0 gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 AY842938 1 90.764 5696369 5696674 3 93.393 3176224 3176556 2 92.628 3203250 3203553 99 00001.1| 00003.1| 00002.1| Ma1 gi|743646596|gb|AZNG010 gi|743629318|gb|AZNF010 gi|734842655|gb|ADNJ020 AY842939 2 95 3234657 3234916 17 97.61 19640 19890 1 94.253 2699852 2699592 42 00002.1| 00017.1| 00001.1| Ma1 gi|743647662|gb|AZNG010 gi|743636528|gb|AZNF010 gi|734841079|gb|ADNJ020 AY842940 1 93.023 3876611 3876397 3 98.068 1332757 1332551 2 96.651 1373224 1373016 45 00001.1| 00003.1| 00002.1| Ma1 gi|743647662|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734839420|gb|ADNJ020 AY842941 1 91.852 1126376 1126628 1 90.441 1747441 1747690 6 90.809 951926 952176 64 00001.1| 00001.1| 00006.1| Ma1 gi|743647662|gb|AZNG010 gi|743630392|gb|AZNF010 gi|734838983|gb|ADNJ020 AY842942 1 97.518 6740670 6740389 13 98.227 170494 170774 8 96.491 1204830 1204546 65 00001.1| 00013.1| 00008.1| Ma1 gi|743641422|gb|AZNG010 gi|743634118|gb|AZNF010 gi|734841079|gb|ADNJ020 AY842943 13 97.906 6629 6818 6 97.906 1043638 1043449 2 96.825 5805698 5805511 95 00013.1| 00006.1| 00002.1| Ma2 gi|743647662|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734839420|gb|ADNJ020 AY842944 1 89.231 96541 96843 1 97.945 702439 702584 6 93.827 701745 702052 10 00001.1| 00001.1| 00006.1| Ma2 gi|743639034|gb|AZNF010 1 95.775 702586 702726 10 00001.1| Ma3 gi|743634730|gb|AZNG010 gi|743630110|gb|AZNF010 gi|734840067|gb|ADNJ020 AY842945 2 92.157 208155 207754 14 94.608 665798 666202 4 99.747 387671 387276 07 00022.1| 00014.1| 00004.1| Ma3 gi|743647662|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734841079|gb|ADNJ020 AY842946 1 89.13 1104637 1104946 1 92.105 1726377 1726669 2 98.013 4877083 4877379 25 00001.1| 00001.1| 00002.1| Ma3 gi|743647662|gb|AZNG010 gi|743639034|gb|AZNF010 gi|734839420|gb|ADNJ020 AY842947 1 95.962 1483450 1483032 1 97.108 2103530 2103118 6 99.272 1308982 1308572 27 00001.1| 00001.1| 00006.1| Ma3 gi|743641014|gb|AZNG010 gi|743637694|gb|AZNF010 gi|734840523|gb|ADNJ020 AY842947 15 98.305 720620 720385 2 99.153 3798667 3798902 3 98.729 1257082 1256847 27 00015.1| 00002.1| 00003.1| Ma3 gi|743629768|gb|AZNG010 gi|743628185|gb|AZNF010 gi|734838719|gb|ADNJ020 AY842948 33 96.795 47850 48161 25 95.556 95302 95123 10 98.397 655383 655694 75 00033.1| 00025.1| 00010.1| Ma3 gi|743628185|gb|AZNF010 25 95.513 95407 95252 75 00025.1| Ma4 gi|743643735|gb|AZNG010 gi|743633372|gb|AZNF010 gi|734839229|gb|ADNJ020 AY842949 6 96.17 767442 767208 7 94.714 304998 305218 7 96.139 1336661 1336914 16 00006.1| 00007.1| 00007.1| Ma4 gi|743645682|gb|AZNG010 gi|743632270|gb|AZNF010 gi|734839658|gb|ADNJ020 AY842950 3 95.778 2601841 2602290 9 99.772 851371 851808 5 99.772 3361217 3361654 17 00003.1| 00009.1| 00005.1|

93

APPENDIX

Table Appendix A 3: Multiplex touchdown PCR conditions (annealing temperature T a [°C], MgCl 2 concentration [mM] and number of PCR cycles n), 16 primer pair combinations for multiplex PCRs, number of alleles (N all ) and allele range (range in bp) for each SSR marker. Number of strains analyzed per species is provided in parentheses.

M. M. M. M. M. M. PCR conditions M. anisopliae (4) M. brunneum (13) M. pingshaense (4) M. robertsii (7) M. majus (9) guizhouens lepidiotae acridum globosum flavoviride frigidum Multiplex primer pair Marker a,b e (9) (4) (9) (1) (4) (1) combinations c Rang Nal Rang Nal Rang Nal Rang Nal Rang Nal Rang Nal Rang Ta MgCl 2 n Nall Range Nall Range Nall Range Nall Range Nall e l e l e l e l e l e l e 168- 170- Ma097 50 3 30 9 2 3 166-176 3 3 171-176 ------177 181 161- 136- 147- Ma099 d 56 4 30 14 2 3 149-152 1 152 3 152-153 3 3 1 147 - - 1 160 - - - - 181 148 152 109- 108- 112- 108- Ma142 d 56 3 22 13 3 5 104-118 2 4 126-143 5 4 - - 1 130 ------113 111 137 137 108- 136- Ma145 56 3 22 7 1 111 7 114-130 2 2 113-117 - - - - 3 ------112 143 114- 123- 94- 105- 109- Ma164 d 56 3 22 10 3 1 119 2 1 114 2 3 2 1 126 1 110 - - - - 164 126 105 116 148 139- Ma165 56 3 22 7 1 139 1 139 2 1 142 - - 1 144 ------143 104- 106- 103- 94- Ma195 50 3 30 9 3 5 105-134 3 3 98-119 2 2 ------124 111 108 103 94- 94- 89- Ma210 d 56 3 30 11 3 82-121 3 94-98 2 71-73 3 91-105 4 2 2 ------101 108 98 147- 150- 150- Ma307 56 3 22 7 3 5 162-200 1 162 3 146-159 3 2 - - - - 1 131 - - - - 165 166 162 142- 148- 139- Ma325 50 3 30 6 3 7 139-169 3 5 154-175 - - - - 1 175 5 1 120 - - 1 348 191 164 151 200- 195- 205- 198- 212- Ma327 50 3 30 15 2 5 198-223 2 2 206-208 2 4 3 ------207 198 214 212 218 150- 150- Ma375 58 4 22 3 2 2 150-159 3 3 122-159 ------159 170 116- 116- 116- Ma416 56 3 22 10 2 1 116 2 2 127-132 2 1 117 1 119 ------127 127 120 Ma417 56 3 22 13 1 257 5 258-269 1 258 1 258 1 258 1 257 1 258 ------128- 117- 117- 115- Ma2049 58 4 22 1 1 130 5 129-139 2 2 129-133 4 4 2 ------135 139 132 119 219- 214- 216- 199- Ma2054 d 58 4 22 4 3 6 231-251 4 2 216-218 5 4 - - 1 185 1 196 - - - - 223 245 247 216 Ma2055 58 4 22 2 1 138 4 142-153 1 140 1 144 - - 1 148 ------139- 136- Ma2056 58 4 22 4 2 5 133-149 2 2 137-141 ------1 223 ------143 139 114- 138- Ma2057 56 3 22 8 1 138 6 155-231 2 3 159-179 2 1 136 ------123 173 153- 162- Ma2060 58 4 22 1 3 4 142-165 ------4 ------178 165 142- 133- 137- 140- Ma2063 d 58 4 22 2 2 6 128-156 3 1 136 2 1 137 2 1 153 1 178 - - - - 144 170 139 144 152- 162- 145- 158- 154- Ma2064 d 58 4 22 5 2 7 151-189 3 2 162-166 4 6 2 ------162 166 163 187 174 126- Ma2065 d 58 4 22 12 1 130 4 127-145 1 128 2 127-129 2 1 128 ------128 223- 219- Ma2069 d 58 4 22 1 - - 7 212-238 - - 1 148 3 2 ------235 239

94

APPENDIX

Ma2070 58 4 22 3 - - 4 107-115 ------256- 253- 256- Ma2077 d 50 3 30 9 2 4 271-289 1 258 - - 4 3 - - 1 246 ------261 265 265 196- 194- 194- Ma2089 58 4 22 2 2 4 191-203 3 2 196-202 2 1 194 1 189 1 195 1 189 - - - - 200 198 197 185- 185- 184- 183- 184- Ma2097 d 56 3 22 10 2 5 185-195 2 3 183-187 7 4 3 ------197 187 208 208 192 162- 183- 168- 164- Ma2098 58 4 22 12 2 4 173-186 4 7 183-212 3 1 167 2 ------172 195 174 168 250- 251- 251- Ma2099 56 3 22 8 2 4 251-263 4 2 251-253 1 251 3 1 251 ------252 345 307 253- 244- 233- 231- 233- Ma2103 d 58 4 22 5 3 4 248-259 3 1 243 6 7 3 1 240 1 240 1 231 1 247 259 251 250 284 244 284- 290- 251- Ma2108 d 56 3 22 8 2 4 288-304 2 2 290-292 5 1 274 1 245 ------294 292 291 Ma2109 58 4 22 5 - - 3 267-290 - - 1 281 - - 1 251 1 250 ------146- 132- Ma2224 d 56 3 22 13 - - 4 150-177 2 1 143 5 1 139 ------153 158 263- 261- 259- 259- Ma2269 58 4 22 12 2 7 239-310 2 1 259 1 259 2 2 ------272 263 261 260 Ma2274 58 4 22 4 - - 5 256-262 - - 3 243-247 ------1 227 ------259- 267- 262- 260- 287- Ma2279 56 3 30 11 2 5 265-293 2 1 271 5 3 3 1 298 1 101 1 101 1 101 269 279 276 269 312 268- 261- Ma2283 56 3 30 11 2 1 278 ------2 - - 1 274 - - - - 271 266 285- 285- 276- 283- Ma2287 50 3 30 15 1 301 7 276-302 2 2 304-313 3 5 2 ------289 297 298 287 183- 201- 198- 202- Ma2292 d 50 3 30 6 2 3 195-197 1 196 1 197 4 3 3 1 197 1 200 - - - - 196 205 202 206 140- 128- 123- 126- Ma2296 d 58 4 22 3 3 5 125-166 2 1 143 5 5 1 126 1 122 1 122 - - - - 149 137 136 134 a loci with 4 digit labels were published by Oulevey et al. (2009) and markers with 3 digit labels were published by Enkerli et al. (2005). b Marker Ma097 – Ma210, Ma307 – Ma417 and Ma2049 – Ma2296 were isolated from M. anisopliae strain ART 2062, M. robertsii strain ARSEF 7532 and M. brunneum strain ARSEF 7524, respectively. c Numbers refer to primer pairs combined in the multiplex PCR used to study cross-species transferability. d Markers revealing two alleles for some M. majus strains.

95

APPENDIX

Table Appendix A 4: Nei’s unbiased genetic diversity (H e) and percentage of strains per species from which PCR products were obtained (% Amp) for each SSR marker. Numbers of strains analyzed per species is provided in parentheses.

Marker a,b M. anisopliae (4) M. brunneum (13) M. pingshaense (4) M. robertsii (7) M. majus (9) M. guizhouense (9) M. lepidiotae (4) M. acridum (9) M. globosum (1) M. flavoviride (4) M. frigidum (1)

He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp He % Amp

Ma097 0.5 100 0.29 100 0.83 100 0.83 57 - 0 - 0 - 0 - 0 - 0 - 0 - 0 Ma099 c 0.5 100 0.43 92 0 100 0.23 100 0.57 100 0.71 89 0 75 - 0 0 100 - 0 - 0 Ma142 c 0.83 100 0.69 100 0.67 100 0.71 100 0.71 100 0.78 100 - 0 0 100 - 0 - 0 - 0 Ma145 0 100 0.79 100 0.5 100 0.48 100 - 0 - 0 0.83 100 - 0 - 0 - 0 - 0 Ma164 c 0.83 100 0 100 0.5 100 0 100 0.43 100 0.56 100 0.5 100 0 100 0 100 - 0 - 0 Ma165 0 100 0 100 0.5 100 0 100 - 0 0 44 - 0 - 0 - 0 - 0 - 0 Ma195 0.83 100 0.78 100 0.83 100 0.71 100 0.22 100 0.6 56 - 0 - 0 - 0 - 0 - 0 Ma210 c 0.83 100 0.62 100 0.5 100 0.67 100 0.77 100 0.48 78 0.5 100 - 0 - 0 - 0 - 0 Ma307 0.83 100 0.74 100 0 100 0.67 100 0.64 100 0.56 100 - 0 - 0 0 100 - 0 - 0 Ma325 0.83 100 0.85 100 0.83 100 0.86 100 - 0 - 0 0 25 0.81 100 0 100 - 0 0 100 Ma327 0.5 100 0.72 69 0.5 100 0.23 100 0.5 44 0.75 89 0.83 100 - 0 - 0 - 0 - 0 Ma375 0.5 100 0.30 92 0.83 100 0.67 100 - 0 - 0 - 0 - 0 - 0 - 0 - 0 Ma416 0.5 100 0 100 0.5 100 0.29 100 0.22 100 0 100 0 100 - 0 - 0 - 0 - 0 Ma417 0 100 0.68 100 0 100 0 100 0 100 0 100 0 25 - 0 - 0 - 0 - 0 Ma2049 0 25 0.69 100 0.50 100 0.48 100 0.64 89 0.78 100 0.50 100 - 0 - 0 - 0 - 0 Ma2054 c 0.83 100 0.82 100 1 100 0.57 100 0.64 89 0.87 67 - 0 0 11 0 100 - 0 - 0 Ma2055 0 100 0.62 100 0 100 0 100 - 0 0 11 - 0 - 0 - 0 - 0 - 0 Ma2056 0.5 100 0.71 85 0.5 100 0.5 57 - 0 - 0 - 0 0 11 - 0 - 0 - 0 Ma2057 0 25 0.85 92 0.67 75 0.73 86 0.33 67 0 11 - 0 - 0 - 0 - 0 - 0 Ma2060 0.83 100 0.53 100 - 0 - 0 - 0 1 44 - 0 - 0 - 0 - 0 - 0 Ma2063 c 0.5 100 0.78 100 0.83 100 0 100 0.51 100 0 100 0.5 100 0 100 0 100 - 0 - 0 Ma2064 c 0.5 100 0.85 100 0.83 100 0.29 100 0.74 100 0.92 100 0.67 75 - 0 - 0 - 0 - 0 Ma2065 c 0 100 0.68 100 0 100 0.48 100 0.43 100 0 100 - 0 - 0 - 0 - 0 - 0 Ma2069 c - 0 0.79 100 - 0 0 14 0.71 56 0.40 56 - 0 - 0 - 0 - 0 - 0 Ma2070 - 0 0.71 54 - 0 - 0 - 0 - 0 - 0 - 0 - 0 - 0 - 0 Ma2077 c 0.5 100 0.6 100 0 100 - 0 0.68 100 0.56 100 - 0 0 100 - 0 - 0 - 0 Ma2089 0.5 100 0.62 100 0.83 100 0.48 100 0.54 89 0 56 0 100 0 100 0 100 - 0 - 0 Ma2097 c 0.5 100 0.78 100 0.67 100 0.67 100 0.85 100 0.78 100 0.83 100 - 0 - 0 - 0 - 0 Ma2098 0.5 100 0.68 100 1 100 1 100 0.83 44 0 67 0.5 100 - 0 - 0 - 0 - 0 Ma2099 0.5 100 0.65 92 1 100 0.33 86 0 100 0.56 100 0 75 - 0 - 0 - 0 - 0 Ma2103 c 0.83 100 0.68 100 0.83 100 0 100 0.85 100 0.92 100 0.83 100 0 100 0 100 0 100 0 100

96

APPENDIX

Ma2108 c 0.5 100 0.68 100 0.5 100 0.48 100 0.81 89 0 67 0 25 - 0 - 0 - 0 - 0 Ma2109 - 0 0.56 69 - 0 0 29 - 0 0 22 0 75 - 0 - 0 - 0 - 0 Ma2224 c - 0 0.72 100 0.67 75 0 100 0.75 89 0 89 - 0 - 0 - 0 - 0 - 0 Ma2269 0.5 100 0.79 100 0.5 100 0 100 0 100 0.39 100 0.5 100 - 0 - 0 - 0 - 0 Ma2274 - 0 0.73 92 - 0 0.67 100 - 0 - 0 - 0 0 11 - 0 - 0 - 0 Ma2279 0.5 100 0.77 100 0.67 100 0 100 0.81 100 0.56 100 0.83 100 0 100 0 100 0 100 0 100 Ma2283 0.5 100 0 8 - 0 - 0 - 0 - 0 0.5 100 - 0 0 100 - 0 - 0 Ma2287 0 100 0.87 100 0.67 75 0.23 100 0.62 100 0.89 100 0.67 75 - 0 - 0 - 0 - 0 Ma2292 c 0.5 100 0.6 100 0 100 0 100 0.57 100 0.72 100 0.83 100 0 11 0 100 - 0 - 0 Ma2296 c 0.67 100 0.73 100 0.5 100 0 100 0.83 100 0.81 100 0 100 0 100 0 100 - 0 - 0

a loci with 4 digit labels were published by Oulevey et al. (2009) and markers with 3 digit labels were published by Enkerli et al. (2005). b Marker Ma097 – Ma210, Ma307 – Ma417 and Ma2049 – Ma2296 were isolated from M. anisopliae strain ART 2062, M. robertsii strain ARSEF 7532 and M. brunneum strain ARSEF 7524, respectively. c Markers revealing two alleles for some M. majus strains.

97

APPENDIX

B) Supporting information to Chapter 3

Figure Appendix B 1: Experimental design in the field experiment. Each plot measured 8.3 times 3 m (4 rows of potato plants). Three plots together formed a block and blocks were separated by a 70 cm wide path (thick line). The 90 plots were surrounded by a buffer zone. Colours indicate different treatments, each replicated six times.

Figure Appendix B 2: Percentage of damaged potato tubers of 50 potato tubers per plot from each treatment in the field experiment (n = 6).

98

APPENDIX

Figure Appendix B 3: Rarefaction curve analysis for the fungal (A) and prokaryotic (B) sequencing dataset including the OTU of the applied strain of the pot experiment (n = 6). Colours of curves correspond to sampling time points: week 0, 7 and 15 were coloured green, blue and orange. Relative abundance of all fungal phyla (C) and prokaryotic phyla with an abundance greater than 1 % (D). a indicates a taxon including all phyla with an abundance of less than 1 %.

99

APPENDIX

Figure Appendix B 4: OTU richness of fungal (A) and prokaryotic (B) communities per treatment and sampling time point in the pot experiment (n = 6) and OTU richness across the long side of the field of the fungal (C) and the prokaryotic (D) communities (n = 9). Letters (a, b) indicate significant differences among samples of the three time points in each treatment and * indicates significant differences between a treatment and the corresponding untreated control at the respective sampling time point (p ≤ 0.05).

100

APPENDIX

Figure Appendix B 5 continued on next page. 101

APPENDIX

Figure Appendix B 5 continued on next page. 102

APPENDIX

Figure Appendix B 5 continued on next page.

103

APPENDIX

Figure Appendix B 5: Relative sequence abundance of prokaryotic OTUs (classified to lowest identified rank) among different treatments and time points (n = 6). * indicates a significant difference between treatments and untreated pots at respective sampling time points (p < 0.05).

104

APPENDIX

Figure Appendix B 6: Rarefaction curve analysis for the fungal (A) and prokaryotic (B) sequencing dataset including the OTU of the applied strain of the field experiment (n = 6). Colours of curves correspond to sampling time points: week 0, 7 and 15 were coloured green, blue and orange. Relative abundance of all fungal phyla (C) and prokaryotic phyla with an abundance greater than 1 % (D). a indicates a taxon including all phyla with an abundance of less than 1 %, b indicates an archaeal phylum.

105

APPENDIX

Table Appendix B 1: Number of isolates identified as M brunneum ARSEF 2825 using six SSR markers in the pot and the field experiment at three sampling time points and isolates obtained from infected cadavers during the monitoring of A. obscurus larvae retrieved from the pot experiment. In parentheses: number of strains selected for genotyping based on SSR analyses. Number of isolates identified as applied strain Experiment Treatment Week 0 Week 7 a / 9 b Week 15 a / 16 b Infected cadaver Pot Untreated 1 (5) NA 0 (2) NC Pot BK 0 (5) NA NA NC Pot Insec 0 (6) NA NA 1 (1) Pot FCBK 0 (6) 2 (2) 6 (6) 1 (1) Pot Fcap 0 (6) 2 (2) 5 (6) 2 (4) Pot Fpowd 0 (6) 1 (1) 6 (6) 7 (8) Pot Gcap 0 (6) NA NA NC Pot FCBK+G cap 0 (6) 2 (2) 6 (6) 2 (2) Pot Fcap +G cap 2 (6) 1 (2) 5 (6) 2 (3) Field Untreated 0 (6) 0 (6) 0 (6) NA Field Fcap 0 (6) 0 (6) 0 (6) NA Field Fgran 0 (6) 2 (6) 0 (6) NA Field FCBK 0 (6) 5 (6) 6 (6) NA Field Insec 0 (6) 0 (6) 0 (6) NA NA … not assessed NC … no infected cadavers a … sampling time points pot experiment b … sampling time points field experiment

Table Appendix B 2: Pairwise ANOSIM comparisons based on Bray Curtis similarities of the fungal and prokaryotic communities in the pot and field experiment among treatments (n=6), among garlic- (n=18) and not- garlic treated pots (n=36), among sampling time points of untreated pots and plots (n=6), among blocks along the long side of the field (n=9). Experiment Organism Paired groups ANOSIM R Pot Fungi BK_7 & Untreated_7 0.55***

Pot Fungi Fcap +G cap _7 & Untreated_7 -0.006

Pot Fungi Fcap _7 & Untreated_7 0.415* Pot Fungi FCBK_7 & Untreated_7 0.506**

Pot Fungi FCBK+G cap _7 & Untreated_7 0.094

Pot Fungi Fpowd _7 & Untreated_7 -0.028

Pot Fungi Gcap _7 & Untreated_7 0.033 Pot Fungi Insec_7 & Untreated_7 0.02 Pot Fungi BK_15 & Untreated_15 0.459*

Pot Fungi Fcap +G cap _15 & Untreated_15 0.459**

Pot Fungi Fcap _15 & Untreated_15 0.278** Pot Fungi FCBK_15 & Untreated_15 0.509**

Pot Fungi FCBK+G cap _15 & Untreated_15 0.341**

Pot Fungi Fpowd _15 & Untreated_15 0.024

Pot Fungi Gcap _15 & Untreated_15 0.031 Pot Fungi Insec_15 & Untreated_15 -0.07 Pot Fungi BK_7 & FCBK_7 0.496**

Pot Fungi BK_7 & F cap _7 0.719**

Pot Fungi FCBK_7 & F cap _7 0.644** Pot Fungi BK_15 & FCBK_15 0.433**

Pot Fungi BK_15 & F cap Gcap _15 0.557** 106

APPENDIX

Pot Fungi FCBK_15 & F cap Gcap _15 0.676** Pot Fungi Untreated_0 & Untreated_7 0.109 Pot Fungi Untreated_0 & Untreated_15 0.454** Pot Fungi Untreated_7 & Untreated_15 0.194* Pot Fungi BK_0 & BK_7 0.6** Pot Fungi BK_0 & BK_15 0.42* Pot Fungi BK_7 & BK_15 -0.026

Pot Fungi Fcap Gcap _0 & Fcap Gcap _7 0.319*

Pot Fungi Fcap Gcap _0 & F cap Gcap _15 0.454**

Pot Fungi Fcap Gcap _7 & F cap Gcap _15 0.317*

Pot Fungi Fcap _0 & F cap _7 0.389*

Pot Fungi Fcap _0 & F cap _15 0.22*

Pot Fungi Fcap _7 & F cap _15 0.543* Pot Fungi FCBK_0 & FCBK_7 0.574** Pot Fungi FCBK_0 & FCBK_15 0.591** Pot Fungi FCBK_7 & FCBK_15 0.054 Pot Prokaryota BK_7 & Untreated_7 -0.089

Pot Prokaryota Fcap +G cap _7 & Untreated_7 0.556**

Pot Prokaryota Fcap _7 & Untreated_7 0.196 Pot Prokaryota FCBK_7 & Untreated_7 0.141*

Pot Prokaryota FCBK+G cap _7 & Untreated_7 0.819**

Pot Prokaryota Fpowd _7 & Untreated_7 0.25*

Pot Prokaryota Gcap _7 & Untreated_7 0.9** Pot Prokaryota Insec_7 & Untreated_7 0.078 Pot Prokaryota BK_15 & Untreated_15 -0.009

Pot Prokaryota Fcap +G cap _15 & Untreated_15 0.57**

Pot Prokaryota Fcap _15 & Untreated_15 0.063 Pot Prokaryota FCBK_15 & Untreated_15 0.019

Pot Prokaryota FCBK+G cap _15 & Untreated_15 0.6**

Pot Prokaryota Fpowd _15 & Untreated_15 0.026

Pot Prokaryota Gcap _15 & Untreated_15 0.585** Pot Prokaryota Insec_15 & Untreated_15 -0.063

Pot Prokaryota Fcap +G cap _7 & G cap _7 0.156

Pot Prokaryota Fcap +G cap _7 & FCBK+G cap _7 0.089

Pot Prokaryota Gcap _7 & FCBK+G cap _7 0.07

Pot Prokaryota Fcap +G cap _15 & G cap _15 0.078

Pot Prokaryota Fcap +G cap _15 & FCBK+G cap _15 -0.087

Pot Prokaryota Gcap _15 & FCBK+G cap _15 -0.022 Pot Prokaryota Untreated_0 & Untreated_7 0.646** Pot Prokaryota Untreated_0 & Untreated_15 0.891** Pot Prokaryota Untreated_7 & Untreated_15 0.426** Pot Prokaryota BK_0 & BK_7 0.789** Pot Prokaryota BK_0 & BK_15 0.974** Pot Prokaryota BK_7 & BK_15 0.454**

Pot Prokaryota Fcap Gcap _0 & F cap Gcap _7 0.572**

Pot Prokaryota Fcap Gcap _0 & F cap Gcap _15 0.783**

Pot Prokaryota Fcap Gcap _7 & F cap Gcap _15 0.322*

Pot Prokaryota Fcap _0 & F cap _7 0.65**

Pot Prokaryota Fcap _0 & F cap _15 0.883** 107

APPENDIX

Pot Prokaryota Fcap _7 & F cap _15 0.515** Pot Prokaryota FCBK_0 & FCBK_7 0.239** Pot Prokaryota FCBK_0 & FCBK_15 0.772** Pot Prokaryota FCBK_7 & FCBK_7 0.193**

Pot Prokaryota FCBK+G cap _0 & FCBK+G cap _7 0.73**

Pot Prokaryota FCBK+G cap _0 & FCBK+G cap _15 0.807**

Pot Prokaryota FCBK+G cap _7 & FCBK+G cap _15 0.393**

Pot Prokaryota Fpowd _0 & F powd _7 0.606**

Pot Prokaryota Fpowd _0 & F powd _15 0.889**

Pot Prokaryota Fpowd _7 & F powd _15 0.206*

Pot Prokaryota Gcap _0 & G cap _7 0.706**

Pot Prokaryota Gcap _0 & G cap _15 0.885**

Pot Prokaryota Gcap _7 & G cap _15 0.463** Pot Prokaryota Insec_0 & Insec_7 0.724** Pot Prokaryota Insec_0 & Insec_15 0.698** Pot Prokaryota Insec_7 & Insec_15 0.644** Field Fungi Untreated_0 & Untreated_9 0.112 Field Fungi Untreated_0 & Untreated_18 0.267** Field Fungi Untreated_9 & Untreated_18 -0.013 Field Fungi 0 m & 9 m -0.026 Field Fungi 0 m & 18 m 0.263*** Field Fungi 0 m & 27 m 0.641*** Field Fungi 0 m & 36 m 0.867*** Field Fungi 0 m & 45 m 0.731*** Field Fungi 0 m & 54 m 0.913*** Field Fungi 0 m & 63 m 0.592*** Field Fungi 0 m & 72 m 0.729*** Field Fungi 0 m & 81 m 0.751*** Field Fungi 9 m & 18 m 0.167* Field Fungi 9 m & 27 m 0.664*** Field Fungi 9 m & 36 m 0.827*** Field Fungi 9 m & 45 m 0.696*** Field Fungi 9 m & 54 m 0.862*** Field Fungi 9 m & 63 m 0.558*** Field Fungi 9 m & 72 m 0.669*** Field Fungi 9 m & 81 m 0.678*** Field Fungi 18 m & 27 m 0.203** Field Fungi 18 m & 36 m 0.592*** Field Fungi 18 m & 45 m 0.428*** Field Fungi 18 m & 54 m 0.664*** Field Fungi 18 m & 63 m 0.523*** Field Fungi 18 m & 72 m 0.623*** Field Fungi 18 m & 81 m 0.614*** Field Fungi 27 m & 36 m 0.24** Field Fungi 27 m & 45 m 0.184** Field Fungi 27 m & 54 m 0.452*** Field Fungi 27 m & 63 m 0.402*** Field Fungi 27 m & 72 m 0.571*** Field Fungi 27 m & 81 m 0.7*** 108

APPENDIX

Field Fungi 36 m & 45 m 0.068 Field Fungi 36 m & 54 m 0.398*** Field Fungi 36 m & 63 m 0.482*** Field Fungi 36 m & 72 m 0.703*** Field Fungi 36 m & 81 m 0.769*** Field Fungi 45 m & 54 m 0.336*** Field Fungi 45 m & 63 m 0.418*** Field Fungi 45 m & 72 m 0.572*** Field Fungi 45 m & 81 m 0.623*** Field Fungi 54 m & 63 m 0.308*** Field Fungi 54 m & 72 m 0.557*** Field Fungi 54 m & 81 m 0.641*** Field Fungi 62 m & 72 m 0.079 Field Fungi 62 m & 81 m 0.389*** Field Fungi 72 m & 81 m 0.235** Field Prokaryota 0 m & 9 m -0.011 Field Prokaryota 0 m & 18 m 0.16** Field Prokaryota 0 m & 27 m 0.795*** Field Prokaryota 0 m & 36 m 0.967*** Field Prokaryota 0 m & 45 m 0.996*** Field Prokaryota 0 m & 54 m 0.939*** Field Prokaryota 0 m & 63 m 0.854*** Field Prokaryota 0 m & 72 m 0.886*** Field Prokaryota 0 m & 81 m 0.856*** Field Prokaryota 9 m & 18 m 0.115* Field Prokaryota 9 m & 27 m 0.783*** Field Prokaryota 9 m & 36 m 0.965*** Field Prokaryota 9 m & 45 m 0.997*** Field Prokaryota 9 m & 54 m 0.924*** Field Prokaryota 9 m & 63 m 0.799*** Field Prokaryota 9 m & 72 m 0.863*** Field Prokaryota 9 m & 81 m 0.798*** Field Prokaryota 18 m & 27 m 0.334** Field Prokaryota 18 m & 36 m 0.855*** Field Prokaryota 18 m & 45 m 0.942*** Field Prokaryota 18 m & 54 m 0.671*** Field Prokaryota 18 m & 63 m 0.334*** Field Prokaryota 18 m & 72 m 0.499*** Field Prokaryota 18 m & 81 m 0.544*** Field Prokaryota 27 m & 36 m 0.47*** Field Prokaryota 27 m & 45 m 0.55*** Field Prokaryota 27 m & 54 m 0.22* Field Prokaryota 27 m & 63 m 0.487*** Field Prokaryota 27 m & 72 m 0.859*** Field Prokaryota 27 m & 81 m 0.92*** Field Prokaryota 36 m & 45 m -0.038 Field Prokaryota 36 m & 54 m 0.419*** Field Prokaryota 36 m & 63 m 0.854*** Field Prokaryota 36 m & 72 m 0.969*** 109

APPENDIX

Field Prokaryota 36 m & 81 m 0.995*** Field Prokaryota 45 m & 54 m 0.38*** Field Prokaryota 45 m & 63 m 0.907*** Field Prokaryota 45 m & 72 m 0.995*** Field Prokaryota 45 m & 81 m 1*** Field Prokaryota 54 m & 63 m 0.49*** Field Prokaryota 54 m & 72 m 0.847*** Field Prokaryota 54 m & 81 m 0.933*** Field Prokaryota 62 m & 72 m 0.354*** Field Prokaryota 62 m & 81 m 0.68*** Field Prokaryota 72 m & 81 m 0.219** * p ≤ 0.05 ** p ≤ 0.01 *** p ≤ 0.001

110

APPENDIX

Table Appendix B 3: Effects of treatments, garlic, time and interactions on fungal and prokaryotic communities in soil of the pot experiment based on BC dissimilarities assessed with overall PERMANOVA corrected p-values of pairwise PERMANOVA tests among untreated pots and treatments. Degrees of Sums of Mean of sums Pseudo F- Test Organism Factor R2 P-value freedom squares of squares statistic Treatment 8 2.2011 0.2751 2.7359 0.1196 0.0001 Time 1 0.6750 0.6750 6.7116 0.0367 0.0001 Fungi Treatmentxtime 7 1.0440 0.1305 1.2977 0.0567 0.0013 Residuals 143 14.4817 0.1006 NA 0.7870 NA Total 161 18.4017 NA NA 1.0000 NA Overall Treatment 8 0.4099 0.0512 1.3207 0.0617 0.0001 Time 1 0.2873 0.2873 7.4064 0.0433 0.0001 Prokaryota Treatmentxtime 7 0.3555 0.0444 1.1456 0.0536 0.0001 Residuals 143 5.5863 0.0388 NA 0.8414 NA Total 161 6.6391 NA NA 1.0000 NA

Pairwise comparison of each treatment with untreated (p-value) Sampling time BK F+G cap Fcap FCBK FCBK+G cap Fpowd Gcap Insec Week 0 0.5240 0.1116 0.5157 0.3128 0.1116 0.3698 0.3128 0.3128 Fungi Week 7 0.0061 0.5173 0.0221 0.0061 0.1632 0.8854 0.3857 0.0809 Week 15 0.0161 0.0060 0.0085 0.0060 0.0085 0.1330 0.2889 0.7119 Pairwise Week 0 0.0801 0.0666 0.2473 0.5454 0.1671 0.3876 0.1868 0.0801 Prokaryota Week 7 0.7175 0.0074 0.0318 0.0318 0.0074 0.0420 0.0074 0.0729 Week 15 0.6474 0.0063 0.3710 0.4478 0.0063 0.3710 0.0063 0.2771

111

APPENDIX

Table Appendix B 4: Percent contribution of the 100 most abundant fungal OTUs to differences in BC dissimilarities between a treatment and the respective control at week 7 and 15 analysed using SIMPER, and the taxonomic classification of each OTU. Treatments were selected based on a significant pairwise ANOSIM and PERMANOVA value assessing fungal community structures among treatments.

Percent contribution (SIMPER) % Week 7 Week 15

OTU Taxonomic classification BK FCBK Fcap OTU Taxonomic classification BK FCBK Fcap FCBK+G cap Fcap +G cap OTU 1* Bionectriaceae 1.49 12.31 0.7 OTU 1* Bionectriaceae 0.93 10.55 0.47 2.01 1.02 OTU 100 Mortierellales 0.9 1.14 1.25 OTU 100 Mortierellales 0.87 0.65 0.61 0.9 0.93 OTU 102 Sordariales 0.74 0.63 0.71 OTU 103 Agaricales 0.66 0.62 0.67 0.78 0.64 OTU 104 Cladorrhinum 0.72 0.87 0.76 OTU 106 Nectriaceae 0.92 0.49 0.59 0.86 1.34 OTU 106 Nectriaceae 0.38 0.55 0.69 OTU 11* Rhizopus oryzae 9.62 0.85 0.81 1.05 0.82 OTU 107 Pseudeurotium 0.44 0.68 0.64 OTU 114 Nectriaceae 1.1 1.16 0.93 0.86 1.01 OTU 11* Rhizopus oryzae 12.03 2.74 0.9 OTU 121 Ascomycota 0.35 0.37 0.89 2.45 0.66 OTU 114 Nectriaceae 0.52 2.47 0.4 OTU 123 Agaricales 0.92 0.76 0.8 0.51 1.53 OTU 123 Agaricales 0.76 1.26 0.97 OTU 125 Volutella ciliata 0.31 0.39 1.49 0.35 0.53 OTU 127 Talaromyces variabilis 0.08 0.16 0.07 OTU 1254 Colletotrichum dematium 1.73 2.01 2.24 1.75 2.05 OTU 13* Nectriaceae 0.92 0.8 1.18 OTU 13* Nectriaceae 1.48 1.13 0.75 2.81 0.5 OTU 140 Trichoderma 0.42 0.41 0.69 OTU 132 Sordariomycetes 0.72 1.08 1 0.82 0.94 OTU 142 Gaeumannomyces graminis 0.12 0.12 0.13 OTU 135 Aspergillus fumigatus 0.33 0.3 0.41 0.64 0.33 OTU 144 Pezizaceae 1.49 1.61 3.21 OTU 144 Pezizaceae 1.72 1.79 1.54 2.53 2.44 OTU 15 Sordariomycetes WF149 2.03 2.33 2.58 OTU 147 Helotiales 0.51 1.04 0.53 0.57 0.46 OTU 154 Actinomucor elegans 0.46 0.45 0.46 OTU 148 Torula 0.26 0.3 0.4 0.32 0.94 OTU 16 Davidiellaceae 1.34 1.43 1.25 OTU 15 Sordariomycetes WF149 1.72 1.62 1.83 2.25 2.36 OTU 166 Hypocreales 1.02 1.19 1.15 OTU 154 Actinomucor elegans 0.8 0.44 0.39 1.18 0.38 OTU 17 Tetracladium 0.86 0.77 1.44 OTU 16 Davidiellaceae 0.61 0.77 1.06 0.57 0.8 OTU 171 Myrothecium verrucaria 0.73 0.7 0.84 OTU 17 Tetracladium 0.82 0.81 0.65 0.99 0.84 OTU 173 Gibberella zeae 0.7 0.64 0.8 OTU 171 Myrothecium verrucaria 0.96 0.58 0.8 0.81 0.62 OTU 178 Sordariales 0.47 0.74 0.7 OTU 18 Acremonium rutilum 1.59 1.78 2.11 1.56 1.38 OTU 18 Acremonium rutilum 1.86 2.18 1.98 OTU 1889 Fungi 1.23 1.29 1.38 1.22 1.66 OTU 1811 Mortierella elongata 0.47 0.47 0.53 OTU 189 Lasiosphaeriaceae 1.81 1.94 2.44 1.87 2.13 OTU 19 Trichocladium asperum 1.47 2.02 1.51 OTU 19 Trichocladium asperum 1.35 1.64 1.81 2.4 1.55 OTU 197 Dothideomycetes WF116 0.15 0.16 1.57 OTU 197 Dothideomycetes WF116 0.12 0.16 0.18 0.19 0.15

112

APPENDIX

OTU 198 Sordariomycetes 0.3 0.45 0.95 OTU 199 Lasiosphaeriaceae 0.86 0.84 1.16 0.76 1 OTU 2 Sordariomycetes 2.86 3.82 5.98 OTU 2 Sordariomycetes 1.37 2.12 1.75 1.34 0.9 OTU 20 Mortierella elongata 0.93 1.16 1.26 OTU 20 Mortierella elongata 1.12 1.02 1.13 1.26 1.37 OTU 2046 Sordariales 0.56 1.2 0.68 OTU 200 Penicillium olsonii 0.92 0.96 1.12 1.07 1.44 OTU 21 Talaromyces 6.86 0.75 0.92 OTU 204 Ascomycota 0.47 0.94 1.12 1.17 0.82 OTU 2114 Podospora 0.21 0.19 0.2 OTU 2046 Sordariales 0.59 0.57 0.5 1.24 0.51 OTU 25 Fusarium merismoides 0.97 0.72 0.74 OTU 205 Trechisporales 0.94 0.91 1.21 1.04 1.21 OTU 260 Psathyrella stercoraria 0.28 0.25 0.22 OTU 21 Talaromyces 6.64 4.32 0.44 0.7 0.57 OTU 262 Acremonium aff curvulum NRRL 62959 0.49 0.49 0.7 OTU 215 Fungi 0.65 0.85 1.6 0.76 0.87 OTU 2683 Cadophora luteo olivacea 0.95 1 1.12 OTU 23 Ilyonectria torresensis 0.59 0.63 1.12 0.66 1.27 OTU 27 uncultured Coprinellus 0.9 0.82 1.13 OTU 2364 Podospora 0.81 0.89 0.92 0.58 1.54 OTU 272 Fungi 0.55 0.84 0.37 OTU 239 Spizellomycetales 0.35 0.34 0.34 0.79 0.74 OTU 29 Ascomycota 0.81 0.75 1.4 OTU 25 Fusarium merismoides 0.89 0.99 0.95 0.82 0.92 OTU 30 Trichoderma gamsii 1.42 1.14 1.23 OTU 27 uncultured Coprinellus 0.8 0.7 0.6 0.61 0.95 OTU 305 Lasiosphaeriaceae 0.85 0.29 0.24 OTU 272 Fungi 0.99 0.63 0.51 1.38 1.71 OTU 31 uncultured Tetracladium 0.98 0.89 1.18 OTU 286 Thanatephorus cucumeris 1.05 1.07 1.48 1.24 1.35 OTU 313 Chaetothyriales 0.41 0.27 0.44 OTU 29 Ascomycota 1.15 0.9 0.98 0.96 1.63 OTU 32 Ascomycota 1.26 1.6 2.32 OTU 295 Piloderma 0 0 0 0 0 OTU 320 uncultured Calyptella 0 1.18 0.09 OTU 30 Trichoderma gamsii 0.47 0.61 0.61 3.48 0.53 OTU 327 Lasiosphaeriaceae 0.54 0.34 0.28 OTU 31 uncultured Tetracladium 0.66 0.81 1 0.8 0.98 OTU 358 Fungi 0 0.33 0 OTU 32 Ascomycota 0.47 0.55 1 0.95 0.8 OTU 36 Pseudeurotium 1.26 1.06 0.97 OTU 338 Saccharomycete P20 0.28 0.26 0.21 0.35 0.55 OTU 365 Tirispora 0.84 0.32 0.54 OTU 343 Cordyceps 0.35 0.36 1.02 0.32 0.47 OTU 37 Myrothecium cinctum 1.24 1.98 2.41 OTU 3464 Sordariomycetes 0.51 0.6 0.58 0.49 0.47 OTU 38 Chaetomiaceae 1.12 1.16 1.69 OTU 36 Pseudeurotium 0.92 0.73 1.09 0.71 1.17 OTU 3918 Beauveria 0 0.14 0 OTU 37 Myrothecium cinctum 0.75 0.9 0.99 0.97 0.98 Cephalotrichum OTU 395 Arthrobotrys musiformis 0.25 0.46 0.29 OTU 379 microsporum 0.19 0.14 0.25 0.29 0.39 OTU 4 Monographella cucumerina 2.7 2.89 3.91 OTU 38 Chaetomiaceae 0.6 0.69 0.56 0.52 0.46 OTU 41 Lasiosphaeriaceae 1.57 0.5 0.61 OTU 384 Helotiales 1.03 1.1 1.21 1.11 1.28 Monographella OTU 42 Pezizomycetes 0.53 0.59 0.98 OTU 4 cucumerina 2.09 1.52 1.76 1.88 1.57

113

APPENDIX

Cladosporium OTU 43 Lasiosphaeriaceae 1.13 0.74 3.22 OTU 40 sphaerospermum 1.55 1.64 1.99 1.57 2.6 OTU 439 Ascomycota 0.62 0.48 0.57 OTU 407 Agaricomycetes 0.55 0.53 0 0 0.04 OTU 44 Neonectria ramulariae 0.88 0.81 0.77 OTU 41 Lasiosphaeriaceae 1.41 1.52 1.87 1.49 1.63 OTU 45* Mortierella 0.82 1.09 1.51 OTU 42 Pezizomycetes 0.86 0.98 1.97 0.75 1.02 OTU 4553 Trichosporon dulcitum 0.69 0.61 0.71 OTU 43 Lasiosphaeriaceae 1.58 1.47 1.8 1.3 1.66 OTU 4577 Hypocreales 0.36 1.69 0.46 OTU 431 Elaphomyces muricatus 0 0 0 0 0 OTU 47 Dipodascaceae 0.62 0.65 1.42 OTU 438 Fungi 0.6 0.54 0.58 0.52 0.6 OTU 48 Staphylotrichum coccosporum 1.06 1.13 1.6 OTU 439 Ascomycota 0.58 0.61 0.79 0.56 0.95 OTU 49 Myrmecridium 1.02 0.7 0.76 OTU 44 Neonectria ramulariae 0.47 0.55 1.01 0.99 0.6 OTU 5 Gibberella intricans 1.83 1.02 1.35 OTU 444 Arthrinium 0.57 0.61 0.9 0.56 0.67 OTU 51 Trichosporon scarabaeorum 0.6 1.23 0.74 OTU 45* Mortierella 1.11 0.81 0.79 1.03 0.95 OTU 524 Mycena 0 0 0 OTU 4553 Trichosporon dulcitum 1.03 1.46 1.4 1.02 1.08 Arthopyreniaceae 2 OTU 53 Pyrenochaeta inflorescentiae 0.6 0.6 0.71 OTU 46 DoF13 0.45 0.45 1.19 0.33 0.58 OTU 55 Neosetophoma samarorum 0.77 0.65 0.7 OTU 469 Trichocomaceae 0.82 1.03 1.01 0.82 0.72 OTU 56 Pleosporales 0.76 0.69 0.79 OTU 47 Dipodascaceae 0.54 0.44 0.8 1.6 0.81 Staphylotrichum OTU 59 Pyrenochaeta 0.63 0.58 0.7 OTU 48 coccosporum 0.61 0.73 0.63 0.79 0.66 OTU 6 Fusarium 1.94 1.66 1.72 OTU 480 Ophiosphaerella 0.7 0.18 0.17 0.14 0.34 OTU 60 Cylindrocarpon FKI 4602 0.6 0.86 1.53 OTU 482 Nectriaceae 0.87 0.82 0.94 0.87 0.93 OTU 61 Hypocreales 0.68 1.4 1.18 OTU 485 Saccharomycete P20 0.19 0.1 0.18 0.38 0.32 OTU 62 Peyronellaea glomerata 1.07 0.77 1.08 OTU 486 Orbiliaceae 0.57 0.97 1.15 0.42 0.49 OTU 624 Helotiales 0.53 0.56 0.6 OTU 49 Myrmecridium 0.74 0.67 0.65 0.49 0.76 OTU 627 Fungi 0.06 0 0.81 OTU 5 Gibberella intricans 1.74 1.68 1.74 1.32 1.8 Trichosporon OTU 63 Monographella cucumerina 1.05 1.34 1.87 OTU 51 scarabaeorum 0.93 0.85 0.91 0.87 0.96 Pyrenochaeta OTU 64 Mortierellaceae 0.91 0.99 0.9 OTU 53 inflorescentiae 1.03 0.88 1.35 1.08 1.15 OTU 65 Chaetomium madrasense 0.67 0.63 0.54 OTU 56 Pleosporales 0.54 0.64 0.78 0.55 0.48 OTU 676 Lecanoromycetes 0.5 0 0 OTU 586 Sordariomycetes 0 0.19 0.09 0 0.2 OTU 68 Trichosporon laibachii 1.1 1.06 1.19 OTU 59 Pyrenochaeta 0.97 0.85 1.2 0.59 1.4 OTU 69 uncultured Tetracladium 0.5 0.33 0.77 OTU 593 Halosphaeriaceae 0.76 0.7 0.85 0.64 0.81 OTU 700 Lecanicillium saksenae 0.24 0.24 0.26 OTU 6 Fusarium 1.55 0.83 1.33 1.23 3.39

114

APPENDIX

OTU 738 Arthrobotrys 0.03 0 0.07 OTU 60 Cylindrocarpon FKI 4602 0.46 0.4 0.54 0.74 0.66 OTU 74 Cercophora 1.4 1.2 1.39 OTU 61 Hypocreales 1.08 1.15 1.16 1.17 0.99 OTU 758 Ascomycota 0 0 0.17 OTU 62 Peyronellaea glomerata 0.87 0.61 0.76 0.96 0.63 OTU 769 Fungi 0.28 0.63 0.13 OTU 65 Chaetomium madrasense 0.49 0.7 0.67 0.6 0.66 OTU 77 Schizothecium carpinicola 1.32 1.35 1.46 OTU 658 Hypocreales 0.38 0.37 0.47 0.34 0.44 OTU 79 Paraphoma chrysanthemicola 0.52 0.74 0.84 OTU 675 Fungi 0 0.67 0 0.2 0 OTU 8 Chaetomiaceae 1.52 1.44 1.57 OTU 68 Trichosporon laibachii 0.94 1.13 1.23 1.92 1.17 Schizothecium OTU 82 Candida sake 0.68 0.69 0.58 OTU 77 carpinicola 1.24 1.4 1.63 1.49 1.64 OTU 847 Phaeosphaeriaceae 0 0 0 OTU 8 Chaetomiaceae 5.59 5.96 6.56 5.32 5.99 OTU 86 Alternaria tenuissima 0.95 0.88 1.24 OTU 91 Tetracladium furcatum 0.47 0.48 0.81 0.79 0.6 OTU 90 Mortierellales 0.53 0.71 0.62 OTU 92 Nectriaceae 0.44 0.27 0.59 0.42 0.61 OTU 91 Tetracladium furcatum 0.82 0.47 0.42 OTU 94 Myrothecium roridum 0.56 0.58 0.87 0.56 0.56 OTU 94 Myrothecium roridum 0.85 1.29 1.59 OTU 95 Alternaria 0.45 0.57 0.82 0.65 0.43 OTU 96 Acremonium persicinum 0.93 1.17 0.97 OTU 97 Fungi 0.5 0.62 0.8 1.36 0.64 OTU 98 Fusarium tricinctum 3.7 0.55 0.53 OTU 99 Microdochium 1.27 0.91 1.01 1.08 0.96 *… OTUs with significant overall PERMANOVA of relative sequence abundance among treatments

115

APPENDIX

Table Appendix B 5: Percent contribution of the 100 most abundant prokaryotic OTUs to differences in BC dissimilarities between a treatment and the respective control at week 7 and 15 analysed using SIMPER, and the taxonomic classification of each OTU. Treatments were selected based on a significant pairwise ANOSIM and PERMANOVA value assessing fungal community structures among treatments.

Percent contribution (SIMPER) % Week 7 Week 15 OTU Taxonomic classification Gcap FCBK+Gcap F+Gcap OTU Taxonomic classification Gcap FCBK+Gcap F+Gcap OTU 1 Hyphomicrobiaceae 0.98 0.86 0.62 OTU 1 Hyphomicrobiaceae 1.89 1.59 1.38 OTU 10 [Chthoniobacteraceae] DA101 1.02 1.15 1.15 OTU 10 [Chthoniobacteraceae] DA101 1.12 0.94 1.05 OTU 100 Kaistobacter 1.21 0.58 0.77 OTU 100 Kaistobacter 1.02 0.82 0.73 OTU 1003 Verrucomicrobiaceae 1.07 1.13 1.32 OTU 103 Solirubrobacterales 0.52 0.53 0.59 OTU 103 Solirubrobacterales 0.6 0.67 0.59 OTU 105 Thermomonas 1.06 0.92 1.08 OTU 105 Thermomonas 1.11 0.8 1.09 OTU 106 Skermanella 1.21 0.99 0.94 OTU 106 Skermanella 0.65 0.71 0.68 OTU 107 Anaerolinaceae 0.76 1.65 0.62 OTU 107 Anaerolinaceae 0.67 0.55 0.84 OTU 109* Acidobacteria-6 iii1-15 0.75 0.68 1.26 OTU 109* Acidobacteria-6 iii1-15 1.12 0.96 0.88 OTU 11 Rhodoplanes 1.32 0.7 0.93 OTU 11 Rhodoplanes 0.97 1 0.66 OTU 110 Chloroflexi Gitt-GS-136 0.79 0.57 0.81 OTU 110 Chloroflexi Gitt-GS-136 0.72 0.79 0.88 OTU 1110 Cyanobacteria 0.19 0.45 0.18 OTU 119 Acidobacteria-6 iii1-15 1.24 0.37 0.43 OTU 117 Chloroflexi Ellin6529 0.94 0.8 0.55 OTU 1191 Verrucomicrobiaceae 0.72 2.26 0.8 OTU 119 Acidobacteria-6 iii1-15 0.86 0.68 0.62 OTU 12 Nitrospira 0.67 0.68 0.57 OTU 12 Nitrospira 0.64 0.67 0.62 OTU 1222 Agrobacterium 0.77 1.21 0.89 OTU 13 Acidobacteria-6 iii1-15 0.93 0.9 1.07 OTU 13 Acidobacteria-6 iii1-15 1.15 0.88 0.78 OTU 131 Betaproteobacteria MND1 0.82 0.84 0.93 OTU 131 Betaproteobacteria MND1 1.36 1.52 0.99 OTU 137 Caldilinea 1.05 0.72 0.78 OTU 137 Caldilinea 1.2 1.34 1.27 OTU 1389* Xanthomonadaceae 3.33 2.47 3.51 OTU 14 Pedomicrobium 0.93 0.94 1.04 OTU 14 Pedomicrobium 0.77 0.81 0.77 OTU 143 [Chloracidobacteria] PK29 0.91 0.59 0.64 OTU 148 Gaiellaceae 0.89 0.64 0.72 OTU 148 Gaiellaceae 0.63 0.66 0.68 OTU 15 CandidatusNitrososphaera SCA1170 0.89 1.08 0.79 OTU 15 CandidatusNitrososphaera SCA1170 0.93 0.72 0.53 OTU 151 Acidobacteria-6 iii1-15 0.79 0.95 0.78 OTU 1501 [Pedosphaerales] auto67_4W 0.97 0.76 1.04 OTU 156 Acidobacteria-6 iii1-15 1.2 0.97 1.13 OTU 156 Acidobacteria-6 iii1-15 0.91 0.92 1.12 OTU 159 Anaerolinea 0.58 1.46 0.74 OTU 16 Geodermatophilaceae 0.6 0.58 0.78 OTU 16 Geodermatophilaceae 0.54 0.53 0.42 OTU 178 Anaerolineae S0208 0.74 0.75 0.66 OTU 178 Anaerolineae S0208 0.68 0.63 0.58

116

APPENDIX

OTU 181 Solirubrobacterales 1.41 1.52 1.21 OTU 180 Sinobacteraceae 0.97 1.04 0.86 OTU 185 Mesorhizobium 1.01 0.92 1.13 OTU 181 Solirubrobacterales 0.76 0.8 0.96 OTU 19 Gaiellaceae 0.75 0.67 0.77 OTU 185 Mesorhizobium 0.87 0.68 0.7 OTU 195 Chitinophagaceae 0.75 0.65 0.92 OTU 19 Gaiellaceae 0.92 0.65 0.67 OTU 2* Micrococcaceae 0.81 0.85 1.01 OTU 195 Chitinophagaceae 1.01 0.73 1.05 OTU 20 Chloroflexi Gitt-GS-136 0.51 0.6 0.75 OTU 2* Micrococcaceae 1.22 0.66 1.1 OTU 21 Streptomyces mirabilis 0.75 0.74 1.07 OTU 20 Chloroflexi Gitt-GS-136 0.92 0.95 1.24 OTU 22 Anaerolinea 1.13 1.22 0.97 OTU 21 Streptomyces mirabilis 1.23 1 0.73 OTU 228 Acidimicrobiales C111 0.79 0.84 1.02 OTU 22 Anaerolinea 0.85 1.99 0.65 OTU 23 Pedomicrobium 0.74 0.73 0.92 OTU 222 Phormidium 2.1 2.2 3.47 OTU 24 Gaiellaceae 0.32 0.4 0.78 OTU 2230 Opitutus 0.45 0.25 2.01 OTU 2475 Luteolibacter 1.06 1.33 1.21 OTU 228 Acidimicrobiales C111 0.72 0.79 0.78 OTU 25 Acidobacteria-6 iii1-15 1.29 0.98 0.87 OTU 23 Pedomicrobium 0.79 0.62 0.95 OTU 27 Anaerolineae GCA004 1.06 0.97 0.83 OTU 24 Gaiellaceae 0.87 0.77 0.75 OTU 271* Acidobacteria-6 iii1-15 1.19 1.42 1.28 OTU 25 Acidobacteria-6 iii1-15 1.29 1.6 1.51 OTU 28 Gaiellaceae 0.76 0.86 0.78 OTU 2564 Paenibacillus 0.87 0.92 0.72 OTU 286 Anaerolinea 1.17 1.27 0.75 OTU 27 Anaerolineae GCA004 0.76 1.38 0.83 OTU 29 Nocardioidaceae 1.01 1.15 1.04 OTU 271* Acidobacteria-6 iii1-15 0.83 1.39 1.37 OTU 3 CandidatusXiphinematobacter 1.42 1.65 1.43 OTU 28 Gaiellaceae 0.85 0.8 0.83 OTU 30 Actinobacteria MB-A2-108 0.39 0.47 0.59 OTU 286 Anaerolinea 0.91 1.86 1.27 OTU 308 Nostocaceae 2.26 2.02 2.11 OTU 29 Nocardioidaceae 0.57 0.57 0.53 OTU 31 Acidobacteria-6 iii1-15 1.26 1.11 0.96 OTU 3 CandidatusXiphinematobacter 1.15 0.85 1.35 OTU 33 Pseudonocardia 0.7 0.65 0.45 OTU 30 Actinobacteria MB-A2-108 0.63 0.54 0.56 OTU 34 Kaistobacter 2 0.98 1.31 OTU 308 Nostocaceae 2.37 2.25 2.43 OTU 346 Solirubrobacteraceae 0.57 0.95 0.72 OTU 31 Acidobacteria-6 iii1-15 0.92 1.04 1.39 OTU 35 Anaerolineae 1.37 2.06 1.13 OTU 33 Pseudonocardia 0.62 0.5 0.51 OTU 38 Acidobacteria-6 iii1-15 1.18 1.01 0.88 OTU 330 Caldilineaceae 0.92 0.62 0.86 OTU 39* Steroidobacter 0.92 0.97 0.86 OTU 34 Kaistobacter 1.69 1.25 1 OTU 40 [Pedosphaerales] 1.19 0.97 1.07 OTU 340 Anaerolinea 0.71 1.43 0.71 OTU 41 Anaerolinea 1.56 1.18 1.04 OTU 346 Solirubrobacteraceae 0.79 0.72 0.7 OTU 42 Gaiellaceae 0.97 0.96 0.85 OTU 349* Ramlibacter 2.28 1.23 1.47 OTU 44 [Pedosphaerales] auto67_4W 1.19 1.15 1.42 OTU 35 Anaerolineae 0.66 0.83 0.88

117

APPENDIX

OTU 48 Bacillus longiquaesitum 0.98 1.21 0.88 OTU 36 Rhodospirillaceae 0.61 0.58 0.55 OTU 488 Chloroflexi Ellin6529 0.64 0.66 0.49 OTU 38 Acidobacteria-6 iii1-15 0.83 0.98 1.1 OTU 49 Syntrophobacteraceae 0.61 0.92 0.6 OTU 39* Steroidobacter 0.44 0.42 0.64 OTU 492 [Chthoniobacteraceae] 1.03 0.95 1.11 OTU 40 [Pedosphaerales] 1.24 0.8 1.03 OTU 5 Bradyrhizobium elkanii 0.51 0.54 0.68 OTU 41 Anaerolinea 0.72 1.75 0.79 OTU 50 Acidobacteria-6 iii1-15 0.74 0.79 0.82 OTU 42 Gaiellaceae 0.48 0.53 0.6 OTU 51 Phormidium 2.52 2.74 3.48 OTU 44 [Pedosphaerales] auto67_4W 1.57 1.08 1.18 OTU 52 Clostridium bowmanii 1.04 0.9 1.47 OTU 48 Bacillus longiquaesitum 0.94 1.05 1.43 OTU 53 Chloroflexi Ellin6529 0.62 0.68 0.76 OTU 488 Chloroflexi Ellin6529 0.98 0.93 0.81 OTU 54 Anaerolineae envOPS12 0.92 1.17 2.72 OTU 49 Syntrophobacteraceae 0.45 0.47 0.4 OTU 55 [Chloracidobacteria] RB41 0.74 0.54 1 OTU 5 Bradyrhizobium elkanii 0.78 0.83 0.77 OTU 56 Rhodoplanes 0.73 1.24 1.23 OTU 50 Acidobacteria-6 iii1-15 1.01 0.99 0.84 OTU 57 Chloroflexi S085 0.66 0.85 0.66 OTU 51 Phormidium 2.61 3.73 3.65 OTU 58 Acidobacteria-6 iii1-15 1.18 0.77 1.09 OTU 53 Chloroflexi Ellin6529 0.86 1.06 1.04 OTU 59 Anaeromyxobacter 0.7 0.47 0.54 OTU 54 Anaerolineae envOPS12 1.23 0.93 0.94 OTU 6 Balneimonas 0.94 0.81 0.93 OTU 55 [Chloracidobacteria] RB41 0.63 0.68 0.78 OTU 60 Acidimicrobiales C111 1.13 0.85 1.1 OTU 56 Rhodoplanes 1 0.89 0.94 OTU 61 Alcaligenaceae 0.77 0.76 0.6 OTU 57 Chloroflexi S085 0.66 0.52 0.46 OTU 65 Piscirickettsiaceae 1 1.03 1.1 OTU 58 Acidobacteria-6 iii1-15 1.44 0.97 1.15 OTU 67 Rhodospirillales 1.17 0.78 1.01 OTU 6 Balneimonas 1.28 1.44 1.14 OTU 69 Acidobacteria-6 iii1-15 0.83 0.74 0.8 OTU 60 Acidimicrobiales C111 0.95 0.78 1.07 OTU 7 Knoellia subterranea 1 1.46 1.44 OTU 61 Alcaligenaceae 0.84 0.62 0.63 OTU 71* Pseudomonas umsongensis 5.41 4.4 4.35 OTU 67 Rhodospirillales 1.34 0.9 0.92 OTU 72 Anaerolineae envOPS12 0.91 1.28 0.99 OTU 69 Acidobacteria-6 iii1-15 0.62 0.73 0.64 OTU 74 Bacillaceae 1.06 1.42 1.15 OTU 7 Knoellia subterranea 1.39 1.03 1.14 OTU 740 Verrucomicrobiaceae 1.3 1.49 0.89 OTU 71* Pseudomonas umsongensis 3.22 2.36 2.59 OTU 744* Acidimicrobiales C111 0.93 1.06 1.33 OTU 72 Anaerolineae envOPS12 0.62 1.29 1.07 OTU 76 Gaiellaceae 1.14 1.08 1.3 OTU 744* Acidimicrobiales C111 0.76 0.8 0.76 OTU 7805 Chloroflexi Gitt-GS-136 0.52 0.68 0.63 OTU 7981 Gaiellaceae 0.76 0.97 0.71 OTU 7981 Gaiellaceae 0.72 0.64 0.71 OTU 8 Chloroflexi Ellin6529 1.07 0.81 0.94 OTU 8 Chloroflexi Ellin6529 0.69 0.43 0.52 OTU 80 Solirubrobacterales 0.78 0.61 0.6 OTU 80 Solirubrobacterales 0.65 1.12 1.01 OTU 81 Bacillus muralis 0.97 0.79 0.93

118

APPENDIX

OTU 81 Bacillus muralis 1.28 1.41 1.28 OTU 87 Sinobacteraceae 0.62 0.81 0.63 OTU 88 Phycisphaerae WD2101 0.52 0.58 0.79 OTU 88 Phycisphaerae WD2101 0.68 0.55 0.63 OTU 8988* Chloroflexi Ellin6529 0.89 0.61 1.14 OTU 8988* Chloroflexi Ellin6529 0.96 1.37 1.55 OTU 9 Anaerolinea 1.96 1.74 1.31 OTU 9 Anaerolinea 1.45 2.2 1.85 OTU 92* Rhodospirillales 0.58 0.4 0.52 OTU 902 Anaerolinea 0.73 1.49 0.78 OTU 934 Chloroflexi Ellin6529 0.71 0.8 0.78 OTU 92* Rhodospirillales 0.74 0.7 0.61 OTU 94 Acidobacteria-6 iii1-15 0.86 1.05 1.15 OTU 934 Chloroflexi Ellin6529 0.86 1.05 0.9 OTU 96 Acidobacteria-6 iii1-15 0.59 0.34 0.4 OTU 94 Acidobacteria-6 iii1-15 1.42 1.49 1.14 OTU 97 Rhodoplanes 0.62 0.76 0.57 OTU 97 Rhodoplanes 0.76 0.67 0.63 OTU 99 Bacillus fumarioli 0.92 1.2 0.88 OTU 99 Bacillus fumarioli 0.66 0.44 0.6 *… OTUs with significant overall PERMANOVA of relative sequence abundance among treatments

119

APPENDIX

Table Appendix B 6: Effects of treatments, distance across the long side of the field, time and interactions on fungal and prokaryotic communities in soil of the field experiment based on BC dissimilarities assessed with overall PERMANOVA and corrected p-value of pairwise PERMANOVA tests among untreated pots and treatments of the fungal communities. Degrees of Sums of Mean of sums Pseudo F- Test Organism Factor R2 P-value freedom squares of squares statistic Treatment 4 0.6326 0.1581 1.4070 0.0557 0.0029 Distance 1 0.7987 0.7987 7.1062 0.0704 0.0001 Time 2 0.9593 0.4796 4.2672 0.0845 0.0001 Fungi Treatment x time 8 0.7677 0.0960 0.8538 0.0676 0.9681 Distance x time 2 0.2100 0.1050 0.9344 0.0185 0.6174 Residuals 71 7.9804 0.1124 NA 0.7032 NA Total 88 11.3487 NA NA 1.0000 NA Overall Treatment 4 0.3122 0.0781 1.0012 0.0438 0.4105 Distance 1 0.2655 0.2655 3.4055 0.0373 0.0014 Time 2 0.4300 0.2150 2.7578 0.0604 0.0013 Prokaryota Treatment x time 8 0.3936 0.0492 0.6311 0.0553 1.0000 Distance x time 2 0.1059 0.0529 0.6791 0.0149 0.9664 Residuals 72 5.6133 0.0780 NA 0.7883 NA Total 89 7.1206 NA NA 1.0000 NA

Pairwise comparison of each treatment with untreated (p-value)

Fcap FCBK Fgran Insec Pairwise Fungi 0.522 0.522 0.124 0.522

120

APPENDIX

C) Supporting information to Chapter 4

Figure Appendix C 1: OTU richness of fungal (A) and prokaryotic (B) communities per treatment and sampling time point (n = 6). Rarefaction curve analyses of fungal sequences including OTU-3 (C) and prokaryotic sequences (D). Colours of lines correspond to sampling time points: green, blue and orange corresponds to 0, 9 and 18 weeks.

121

APPENDIX

Figure Appendix C 2: Soil fungal (A) and prokaryotic (B) community composition at phylum level per treatment and sampling time point based on mean relative sequence abundance (n = 6). Only the 10 most abundant prokaryotic phyla are displayed in colour. The letter “a” indicates archaeal phyla, while all other phyla are bacteria. Square brackets indicate proposed phyla.

122

APPENDIX

Table Appendix C 1: Number of Metarhizium isolates identified as M. brunneum EAMa01/58-Su using SSR marker analysis of six selected isolates obtained for each treatment at sampling time points week 0, 9 and 18. Number of strains identified as M. brunneum EAMa01/58-Su Treatment Week 0 Week 9 Week 18 FCBK 0 5 6 BK 0 0 0 Insec 0 1 2 Untreated 0 1 0

123

APPENDIX

D) Supporting information to General discussion

Table Appendix D 1:Allele sizes (bp) of six SSR markers and the genotype of all Metarhizium -isolates obtained from the soil of different treatments and sampling time points in the pot and field experiments to control Agriotes and D. v. virgifera in Chapter 3 and 4.

Allele size of SSR marker (bp) Experiment Sampling time point Treatment Replicate Genotype Ma2049 Ma2054 Ma2063 Ma2287 Ma327 Ma195 Agriotes -pot preappl Gcap 5 119 216 137 284 225 97 A Agriotes -pot preappl Fpowd 4 119 216 137 284 225 101 B Agriotes -pot preappl BK 1 119 216 137 284 225 NA C Agriotes -pot preappl FCBK 5 119 216 137 284 225 NA C Agriotes -field preappl FCBK 2 129 214 135 314 206 111 Q Agriotes -pot postappl1 Fcap & Gcap 4 129 216 135 314 206 111 D Agriotes -pot postappl2 Fcap 5 129 216 135 314 206 111 D Agriotes -pot preappl Fcap 2 129 216 135 314 206 111 D Agriotes -pot preappl Fcap 5 129 216 135 314 206 111 D Agriotes -pot preappl Fcap 6 129 216 135 314 206 111 D Agriotes -pot preappl FCBK 3 129 216 135 314 206 111 D Agriotes -pot preappl FCBK & Gcap 4 129 216 135 314 206 111 D Agriotes -pot preappl FCBK& Gcap 5 129 216 135 314 206 111 D Agriotes -pot preappl Fpowd 2 129 216 135 314 206 111 D Agriotes -pot preappl Fpowd 3 129 216 135 314 206 111 D Agriotes -pot preappl Gcap 4 129 216 135 314 206 111 D Agriotes -pot preappl Insec 1 129 216 135 314 206 111 D Agriotes -pot preappl Insec 6 129 216 135 314 206 111 D Agriotes -pot preappl Untreated 1 129 216 135 314 206 111 D Agriotes -field preappl Untreated 1 129 216 135 314 206 111 D Agriotes -field preappl Untreated 4 129 216 135 314 206 111 D Agriotes -field preappl Untreated 6 129 216 135 314 206 111 D Agriotes -field preappl Insec 1 129 216 135 314 206 111 D Agriotes -field preappl Insec 5 129 216 135 314 206 111 D Agriotes -field preappl Insec 6 129 216 135 314 206 111 D Agriotes -field preappl FCBK 1 129 216 135 314 206 111 D Agriotes -field preappl FCBK 3 129 216 135 314 206 111 D Agriotes -field preappl Fcap 2 129 216 135 314 206 111 D Agriotes -field preappl Fcap 4 129 216 135 314 206 111 D Agriotes -field preappl Fcap 5 129 216 135 314 206 111 D Agriotes -field preappl Fgran 1 129 216 135 314 206 111 D Agriotes -field preappl Fgran 3 129 216 135 314 206 111 D Agriotes -field preappl Fgran 6 129 216 135 314 206 111 D Agriotes -field postappl1 Untreated 1 129 216 135 314 206 111 D Agriotes -field postappl1 Untreated 2 129 216 135 314 206 111 D Agriotes -field postappl1 Insec 1 129 216 135 314 206 111 D Agriotes -field postappl1 Insec 6 129 216 135 314 206 111 D Agriotes -field postappl1 Fcap 4 129 216 135 314 206 111 D Agriotes -field postappl1 Fcap 6 129 216 135 314 206 111 D Agriotes -field postappl1 Fgran 5 129 216 135 314 206 111 D Agriotes -field postappl2 Untreated 1 129 216 135 314 206 111 D Agriotes -field postappl2 Untreated 2 129 216 135 314 206 111 D Agriotes -field postappl2 Insec 6 129 216 135 314 206 111 D Agriotes -field postappl2 Fcap 1 129 216 135 314 206 111 D 124

APPENDIX

Agriotes -field postappl2 Fgran 1 129 216 135 314 206 111 D Agriotes -field postappl2 Fgran 2 129 216 135 314 206 111 D Agriotes -field postappl2 Fgran 6 129 216 135 314 206 111 D Agriotes -field postappl1 Fgran 3 129 216 135 314 206 113 R Agriotes -field postappl2 Insec 2 129 216 135 314 206 113 R Agriotes -pot preappl Untreated 3 129 218 135 298 211 120 E Agriotes -field preappl FCBK 6 129 218 135 304 206 111 S Agriotes -field postappl2 Fcap 3 129 218 135 304 206 111 S Agriotes -pot preappl BK 5 129 218 135 314 206 111 F Agriotes -pot preappl FCBK 1 129 218 135 314 206 111 F Agriotes -pot preappl FCBK 2 129 218 135 314 206 111 F Agriotes -pot preappl FCBK & Gcap 3 129 218 135 314 206 111 F Agriotes -pot preappl FCBK & Gcap 6 129 218 135 314 206 111 F Agriotes -pot preappl Fpowd 1 129 218 135 314 206 111 F Agriotes -pot preappl Fpowd 6 129 218 135 314 206 111 F Agriotes -pot preappl Gcap 3 129 218 135 314 206 111 F Agriotes -pot preappl Gcap 6 129 218 135 314 206 111 F Agriotes -pot preappl Insec 5 129 218 135 314 206 111 F Agriotes -pot preappl Insec 3 129 239 141 284 225 122 G Agriotes -field postappl1 Fcap 5 129 239 141 294 211 120 T Agriotes -pot monitoring Fcap 4 129 239 141 298 211 120 ARSEF7524 Agriotes -pot monitoring Fcap & Gcap 3 129 239 141 298 211 120 ARSEF7524 Agriotes -pot monitoring Fpowd 4 129 239 141 298 211 120 ARSEF7524 Agriotes -pot postappl2 Fcap & Gcap 3 129 239 141 298 211 120 ARSEF7524 Agriotes -pot postappl2 Untreated 2 129 239 141 298 211 120 ARSEF7524 Agriotes -pot postappl2 Untreated 2 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Fcap & Gcap 4 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Fcap & Gcap 5 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Fcap 4 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl FCBK 4 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl FCBK 6 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl FCBK & Gcap 2 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Insec 2 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Insec 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Untreated 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Untreated 5 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Insec 2 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Insec 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Fcap 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Fcap 6 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Fgran 2 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Fgran 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field preappl Fgran 5 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Untreated 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Insec 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Insec 5 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Fcap 1 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Fcap 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl1 Fgran 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Untreated 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Untreated 4 129 239 141 298 211 120 ARSEF7524 125

APPENDIX

Agriotes -field postappl2 Untreated 5 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Untreated 6 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Insec 1 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Insec 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Fcap 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Fcap 6 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Fgran 3 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Fgran 4 129 239 141 298 211 120 ARSEF7524 Agriotes -field postappl2 Fgran 5 129 239 141 298 211 120 ARSEF7524 Agriotes -pot preappl Untreated 2 129 239 141 NA 211 120 ARSEF7524 Agriotes -pot preappl Untreated 5 129 NA 156 NA NA NA H Agriotes -field postappl1 FCBK 6 131 160 NA 302 213 111 U Agriotes -field preappl Insec 4 131 232 135 303 208 106 V Agriotes -field preappl FCBK 4 131 232 135 303 208 106 V Agriotes -field preappl FCBK 5 131 232 135 303 208 106 V Agriotes -field preappl Fcap 1 131 232 135 303 208 106 V Agriotes -field postappl1 Untreated 4 131 232 135 303 208 106 V Agriotes -field postappl1 Untreated 5 131 232 135 303 208 106 V Agriotes -field postappl1 Untreated 6 131 232 135 303 208 106 V Agriotes -field postappl1 Insec 2 131 232 135 303 208 106 V Agriotes -field postappl1 Insec 3 131 232 135 303 208 106 V Agriotes -field postappl1 Fgran 1 131 232 135 303 208 106 V Agriotes -field postappl2 Insec 3 131 232 135 303 208 106 V Agriotes -field postappl2 Insec 5 131 232 135 303 208 106 V Agriotes -field postappl2 Fcap 2 131 232 135 303 208 106 V Agriotes -field postappl2 Fcap 5 131 232 135 303 208 106 V Agriotes -pot product FCBK 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fcap 3 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fcap & Gcap 1 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fcap & Gcap 1 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring FCBK 2 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring FCBK & Gcap 5 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring FCBK & Gcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 1 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 4 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 4 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 4 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 4 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 5 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Fpowd 6 131 234 156 302 213 111 ART2825 Agriotes -pot monitoring Insec 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 Fcap 4 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 Fcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 Fcap & Gcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 FCBK 3 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 FCBK 5 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 FCBK & Gcap 2 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 FCBK & Gcap 4 131 234 156 302 213 111 ART2825 Agriotes -pot postappl1 Fpowd 2 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap & Gcap 4 131 234 156 302 213 111 ART2825 126

APPENDIX

Agriotes -pot postappl2 Fcap & Gcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap 1 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap 2 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap 3 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap 4 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 FCBK 1 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 FCBK 2 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 FCBK 4 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 FCBK 5 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 FCBK 6 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fpowd 1 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fpowd 2 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fpowd 3 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fpowd 4 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fpowd 5 131 234 156 302 213 111 ART2825 Agriotes -pot preappl Fcap & Gcap 2 131 234 156 302 213 111 ART2825 Agriotes -pot preappl Fcap & Gcap 6 131 234 156 302 213 111 ART2825 Agriotes -pot preappl Untreated 4 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 FCBK 1 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 FCBK 2 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 FCBK 3 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 FCBK 4 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 FCBK 5 131 234 156 302 213 111 ART2825 Agriotes -field postappl1 Fgran 6 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 1 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 2 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 3 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 4 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 5 131 234 156 302 213 111 ART2825 Agriotes -field postappl2 FCBK 6 131 234 156 302 213 111 ART2825 Agriotes -field product Fgran 131 234 156 302 213 111 ART2825 Agriotes -field product Fcap 131 234 156 302 213 111 ART2825 Agriotes -field product FCBK 131 234 156 302 213 111 ART2825 Agriotes -pot postappl2 Fcap & Gcap 1 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 Fcap & Gcap 2 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 Fcap & Gcap 5 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 1 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 2 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 3 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 4 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 5 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK & Gcap 6 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 Fpowd 6 131 234 156 302 213 NA ART2825 Agriotes -pot postappl2 FCBK 3 131 234 156 302 NA NA ART2825 Agriotes -pot preappl BK 4 133 234 156 302 213 111 I Agriotes -pot preappl BK 6 133 234 156 302 213 111 I Agriotes -pot preappl Fcap 1 133 234 156 302 213 111 I Agriotes -field postappl1 Fgran 2 133 234 156 302 213 111 I Agriotes -pot preappl Fcap & Gcap 3 133 248 158 292 213 111 J Agriotes-field preappl Untreated 2 137 218 135 304 208 107 W 127

APPENDIX

Agriotes -pot preappl BK 3 137 220 128 288 NA 105 K Agriotes -pot preappl Fcap & Gcap 1 137 264 128 288 204 105 L Agriotes -pot preappl Gcap 1 137 264 128 288 212 105 M Agriotes -pot monitoring Fcap 3 137 264 128 288 NA 105 N Agriotes -pot preappl FCBK & Gcap 1 137 264 128 288 NA 105 N Agriotes -pot preappl Fpowd 5 137 264 128 288 NA 105 N Agriotes -pot preappl Gcap 2 146 212 135 304 208 122 O Agriotes -pot preappl Fcap 3 167 218 135 304 208 106 P Agriotes -field postappl1 Fcap 2 NA NA NA 298 211 120 X Diabrotica preappl FCBK 4 129 216 135 314 206 111 D Diabrotica preappl FCBK 5 129 216 135 314 206 111 D Diabrotica preappl Insec 1 129 216 135 314 206 111 D Diabrotica preappl Insec 2 129 216 135 314 206 111 D Diabrotica preappl Untreated 2 129 216 135 314 206 111 D Diabrotica postappl2 BK 1 129 216 135 314 206 111 D Diabrotica postappl2 BK 6 129 216 135 314 206 111 D Diabrotica postappl2 Untreated 1 129 216 135 314 206 111 D Diabrotica postappl2 Untreated 3 129 216 135 314 206 111 D Diabrotica postappl2 Untreated 6 129 216 135 314 206 111 D Diabrotica preappl FCBK 1 129 218 135 304 206 111 S Diabrotica preappl FCBK 2 129 218 135 304 206 111 S Diabrotica preappl FCBK 3 129 218 135 304 206 111 S Diabrotica preappl FCBK 6 129 218 135 304 206 111 S Diabrotica preappl BK 1 129 218 135 304 206 111 S Diabrotica preappl BK 2 129 218 135 304 206 111 S Diabrotica preappl BK 4 129 218 135 304 206 111 S Diabrotica preappl BK 5 129 218 135 304 206 111 S Diabrotica preappl BK 6 129 218 135 304 206 111 S Diabrotica preappl Insec 3 129 218 135 304 206 111 S Diabrotica preappl Insec 4 129 218 135 304 206 111 S Diabrotica preappl Insec 5 129 218 135 304 206 111 S Diabrotica preappl Insec 6 129 218 135 304 206 111 S Diabrotica preappl Untreated 1 129 218 135 304 206 111 S Diabrotica preappl Untreated 3 129 218 135 304 206 111 S Diabrotica preappl Untreated 4 129 218 135 304 206 111 S Diabrotica preappl Untreated 5 129 218 135 304 206 111 S Diabrotica preappl Untreated 6 129 218 135 304 206 111 S Diabrotica postappl1 FCBK 1 129 218 135 304 206 111 S Diabrotica postappl1 BK 1 129 218 135 304 206 111 S Diabrotica postappl1 BK 2 129 218 135 304 206 111 S Diabrotica postappl1 BK 3 129 218 135 304 206 111 S Diabrotica postappl1 BK 4 129 218 135 304 206 111 S Diabrotica postappl1 BK 5 129 218 135 304 206 111 S Diabrotica postappl1 BK 6 129 218 135 304 206 111 S Diabrotica postappl1 Insec 1 129 218 135 304 206 111 S Diabrotica postappl1 Insec 3 129 218 135 304 206 111 S Diabrotica postappl1 Insec 4 129 218 135 304 206 111 S Diabrotica postappl1 Insec 5 129 218 135 304 206 111 S Diabrotica postappl1 Untreated 2 129 218 135 304 206 111 S Diabrotica postappl1 Untreated 3 129 218 135 304 206 111 S Diabrotica postappl1 Untreated 4 129 218 135 304 206 111 S 128

APPENDIX

Diabrotica postappl1 Untreated 6 129 218 135 304 206 111 S Diabrotica postappl2 BK 2 129 218 135 304 206 111 S Diabrotica postappl2 BK 4 129 218 135 304 206 111 S Diabrotica postappl2 Insec 1 129 218 135 304 206 111 S Diabrotica postappl2 Insec 3 129 218 135 304 206 111 S Diabrotica postappl2 Insec 4 129 218 135 304 206 111 S Diabrotica postappl2 Insec 5 129 218 135 304 206 111 S Diabrotica postappl2 Untreated 5 129 218 135 304 206 111 S Diabrotica postappl1 Insec 2 129 218 135 304 208 113 AA Diabrotica preappl BK 3 129 218 135 304 208 148 AB Diabrotica postappl2 BK 3 129 218 135 304 208 148 AB Diabrotica postappl2 BK 5 129 218 135 304 208 148 AB Diabrotica postappl2 Untreated 2 129 218 135 304 208 148 AB Diabrotica postappl2 Untreated 4 129 218 135 304 208 148 AB Diabrotica postappl1 Untreated 5 129 218 135 304 208 152 AC Diabrotica postappl1 FCBK 2 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 FCBK 3 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 FCBK 4 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 FCBK 5 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 FCBK 6 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 Insec 6 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl1 Untreated 1 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 1 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 2 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 3 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 4 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 5 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 FCBK 6 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 Insec 2 137 251 128 287 NA 105 EAMa01/58-Su Diabrotica postappl2 Insec 6 137 251 128 287 NA 105 EAMa01/58-Su

Appendix D 2: A multiple sequence alignment of the ITS2 region (310 positions) of 21 sequences including isolates of M. anisopliae , M. brunneum , M. guizhouense , M. majus , M. pingshaense and M. robertsii

>M. guizhouense ARSEF6238 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGAAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAAACCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. guizhouense CBS258.90 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGAAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAAACCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. guizhouense ARSEF4321 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. guizhouense ARSEF4303 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. majus ARSEF1015 129

APPENDIX

CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACACCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. majus ARSEF1946 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACACCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. majus ARSEF1914 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACACCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. majus ARSEF4566 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTGATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. robertsii ARSEF2527 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. robertsii ARSEF727 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. brunneum ARSEF4179 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAAACCCCCAACTTTTTATAGTTGACCTCG AATCAGGTAGGACT >M. brunneum ARSEF4152 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAAACCCCCAACTTTTTATAGTTGACCTCG AATCAGGTAGGACT >M. brunneum ARSEF7524 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCGCAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. brunneum ARSEF1066 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCGCAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. pingshaense ARSEF4342 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. pingshaense ARSEF3210 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGGGACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCA GCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCTGCGCAGTAGTAAAGCACTCGCAACAGGAGCCCGG CGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGAATCAGGTAGGACT >M. anisopliae ARSEF7487 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAGCACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT 130

APPENDIX

>M. anisopliae ARSEF7450 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAGCACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. pingshaense CBS257.90 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGCGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTTAAATTAATTGGCGGTCTCGCCGTGGCCCTCCTCT GCGCAGTAGTAAAGCACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. brunneum BIPESCO5 CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCGCAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT >M. brunneum ARSEF2825* CGAAATGCGATAAGTAATGTGAATTGCAGAATTCAGTGAATCATCGAATCTTTGAACGCACATTGCGCCCGTCAGTATTCTGG CGGGCATGCCTGTTCGAGCGTCATTACGCCCCTCAAGTCCCCTGTGG- ACTTGGTGTTGGGGATCGGCGAGGCTGGTTTTCCAGCACAGCCGTCCCTCAAATCAATTGGCGGTCTCGCCGTGGCCCTCCTC TGCGCAGTAGTAAAACACTCGCAACAGGAGCCCGGCGCGGTCCACTGCCGTAAAACCCCCCAACTTTTTATAGTTGACCTCGA ATCAGGTAGGACT

131

REFERENCES

References

Abarenkov, K., Henrik Nilsson, R., Larsson, K.-H., Alexander, I.J., Eberhardt, U., Erland, S., Høiland, K., Kjøller, R., Larsson, E., Pennanen, T., Sen, R., Taylor, A.F.S., Tedersoo, L., Ursing, B.M., Vrålstad, T., Liimatainen, K., Peintner, U., Kõljalg, U., 2010. The UNITE database for molecular identification of fungi – recent updates and future perspectives. New Phytologist 186 , 281-285. Abbott, W.S., 1987. Abbotts formula - a method of computing the effectiveness of an insecticide. Journal of American Mosquito Control 3 , 302-303. Agrawal, P.K., Agrawal, S., Shrivastava, R., 2015. Modern molecular approaches for analyzing microbial diversity from mushroom compost ecosystem. 3 Biotech 5 , 853-866. Allison, S.D., Martiny, J.B.H., 2008. Resistance, resilience, and redundancy in microbial communities. Proceedings of the National Academy of Sciences of the United States of America 105 , 11512-11519. Anderson, M.J., 2006. Distance-based tests for homogeneity of multivariate dispersions. Biometrics 62 , 245-253. Andrews, K.R., Good, J.M., Miller, M.R., Luikart, G., Hohenlohe, P.A., 2016. Harnessing the power of RADseq for ecological and evolutionary genomics. Nature Reviews Genetics 17 , 81-92. Anonymous, 2005. Wireworms. EPPO Bulletin 35 , 179-182. Ansari, M.A., Evans, M., Butt, T.M., 2009. Identification of pathogenic strains of entomopathogenic nematodes and fungi for wireworm control. Crop Protection 28 , 269-272. Ansari, M.A., Shah, F.A., Butt, T.M., 2010. The entomopathogenic nematode Steinernema kraussei and Metarhizium anisopliae work synergistically in controlling overwintering larvae of the black vine weevil, Otiorhynchus sulcatus , in strawberry growbags. Biocontrol Science and Technology 20 , 99-105. Aragón, P., Baselga, A., Lobo, J.M., 2010. Global estimation of invasion risk zones for the Western corn rootworm Diabrotica virgifera virgifera : Integrating distribution models and physiological thresholds to assess climatic favourability. Journal of Applied Ecology 47 , 1026-1035. Aregger, E., 1992. Conidia production of the fungus Beauveria brongniartii on barley and quality evaluation during storage at 2°C. Journal of Invertebrate Pathology 59 , 2-10. Auffret, M.D., Karhu, K., Khachane, A., Dungait, J.A.J., Fraser, F., Hopkins, D.W., Wookey, P.A., Singh, B.K., Freitag, T.E., Hartley, I.P., Prosser, J.I., 2016. The role of microbial community composition in controlling soil respiration responses to temperature. PloS One 11 , e0165448. Babendreier, D., Jeanneret, P., Pilz, C., Toepfer, S., 2015. Non-target effects of insecticides, entomopathogenic fungi and nematodes applied against western corn rootworm larvae in maize. Journal of Applied Entomology 139 , 457-467. Bača, F., 1994. New member of the harmful entomofauna of Yugoslavia Diabrotica virgifera virgifera LeConte (Coleoptera, Chrysomelidae). Zaštita Bilja 45 , 125-131. Badri, D.V., Vivanco, J.M., 2009. Regulation and function of root exudates. Plant, Cell and Environment 32 , 666- 681. Barelli, L., Moonjely, S., Behie, S.W., Bidochka, M.J., 2016. Fungi with multifunctional lifestyles: endophytic insect pathogenic fungi. Plant Molecular Biology 90, 657-664. Barsics, F., Haubruge, E., Francis, F., Verheggen, F.J., 2014. The role of olfaction in wireworms: a review on their foraging behavior and sensory apparatus. Biotechnology, Agronomy, Society and Environment 18 , 524- 535. Barsics, F., Haubruge, E., Verheggen, F.J., 2013. Wireworms’ management: An overview of the existing methods, with particular regards to Agriotes spp. (Coleoptera: Elateridae). Insects 4 , 117-152. Barthe, S., Gugerli, F., Barkley, N.A., Maggia, L., Cardi, C., Scotti, I., 2012. Always look on both sides: phylogenetic information conveyed by simple sequence repeat allele sequences. PloS One 7 , doi: 10.1371/journal.pone.0040699. Baverstock, J., Baverstock, K.E., Clark, S.J., Pell, J.K., 2008. Transmission of Pandora neoaphidis in the presence of co-occurring arthropods. Journal of Invertebrate Pathology 98 , 356-359. Behie, S.W., Jones, S.J., Bidochka, M.J., 2015. Plant tissue localization of the endophytic insect pathogenic fungi Metarhizium and Beauveria . Fungal Ecology 13 , 112-119. Behie, S.W., Zelisko, P.M., Bidochka, M.J., 2012. Endophytic insect-parasitic fungi translocate nitrogen directly from insects to plants. Science 336 , 1576-1577. Behle, R.W., Jackson, M.A., Flor-Weiler, L.B., 2013. Efficacy of a granular formulation containing Metarhizium brunneum F52 (Hypocreales: Clavicipitaceae) microsclerotia against nymphs of Ixodes scapularis (Acari: Ixoididae). Journal of Economic Entomology 106 , 57-63. 132

REFERENCES

Beijerinck, M.W., 1901. Ueber oligonitrophile Mikroben. Zentralblatt für Bakteriologie 7 , 561-582. Bengtsson-Palme, J., Hartmann, M., Eriksson, K.M., Pal, C., Thorell, K., Larsson, D.G., Nilsson, R.H., 2015. METAXA2: Improved identification and taxonomic classification of small and large subunit rRNA in metagenomic data. Molecular Ecology Resources 15 , 1403-1414. Bengtsson-Palme, J., Ryberg, M., Hartmann, M., Branco, S., Wang, Z., Godhe, A., De Wit, P., Sánchez-García, M., Ebersberger, I., de Sousa, F., Amend, A., Jumpponen, A., Unterseher, M., Kristiansson, E., Abarenkov, K., Bertrand, Y.J.K., Sanli, K., Eriksson, K.M., Vik, U., Veldre, V., Nilsson, R.H., 2013. Improved software detection and extraction of ITS1 and ITS2 from ribosomal ITS sequences of fungi and other eukaryotes for analysis of environmental sequencing data. Methods in Ecology and Evolution 4 , 914-919. Benjamin, E.O., Wesseler, J.H.H., 2016. A socioeconomic analysis of biocontrol in integrated pest management: A review of the effects of uncertainty, irreversibility and flexibility. NJAS - Wageningen Journal of Life Sciences 77 , 53-60. Benjamini, Y., Hochberg, Y., 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. Journal of the Royal Statistical Society 57 , 289-300. Benson, D.A., Clark, K., Karsch-Mizrachi, I., Lipman, D.J., Ostell, J., Sayers, E.W., 2015. GenBank. Nucleic Acids Research 43 , D30-35. Beys da Silva, W.O., Santi, L., Schrank, A., Vainstein, M.H., 2010. Metarhizium anisopliae lipolytic activity plays a pivotal role in Rhipicephalus (Boophilus ) microplus infection. Fungal Biology 114, 10-15. Bidochka, M.J., Kasperski, J.E., Wild, G.A.M., 1998. Occurrence of the entomopathogenic fungi Metarhizium anisopliae and Beauveria bassiana in soils from temperate and near-northern habitats. Canadian Journal of Botany 76 , 1198-1204. Bischoff, J.F., Rehner, S.A., Humber, R.A., 2009. A multilocus phylogeny of the Metarhizium anisopliae lineage. Mycologia 101 , 512-530. Blackshaw, R.P., Vernon, R.S., 2008. Spatial relationships between two Agriotes click-beetle species and wireworms in agricultural fields. Agricultural and Forest Entomology 10 , 1-11. Blagodatskaya, E., Kuzyakov, Y., 2013. Active microorganisms in soil: Critical review of estimation criteria and approaches. Soil Biology and Biochemistry 67 , 192-211. Blot, Y., Brunel, E., 1999. Survey on the infection of wheat and maize by larvae of wireworms of Agriotes and Athous genera (Coleoptera : Elateridae) in some areas of West France. Annales de la Société Entomologique de France 35 , 453-457. Bodini, S.F., Manfredini, S., Epp, M., Valentini, S., Santori, F., 2009. Quorum sensing inhibition activity of garlic extract and p-coumaric acid. Letters in Applied Microbiology 49 , 551-555. Bognar, S., 1955. P. depressa a paralysing and destructive parasite of the wireworm, A. obscurus . Novenytermeles 4, 241-252. Boldo, J.T., Junges, A., do Amaral, K.B., Staats, C.C., Vainstein, M.H., Schrank, A., 2009. Endochitinase CHI2 of the biocontrol fungus Metarhizium anisopliae affects its virulence toward the cotton stainer bug Dysdercus peruvianus . Current Genetics 55 , 551-560. Boughner, L.A., Singh, P., 2016. Microbial Ecology: Where are we now? Postdoc journal: a journal of postdoctoral research and postdoctoral affairs 4 , 3-17. Boulter, D., 1993. Insect pest control by copying nature using genetically engineered crops. Phytochemistry 34 , 1453-1466. Brandl, M.A., Schumann, M., Przyklenk, M., Patel, A., Vidal, S., 2017. Wireworm damage reduction in potatoes with an attract-and-kill strategy using Metarhizium brunneum . Journal of Pest Science 90 , 479-493. Branson, T.F., Krysan, J.L., 1981. Feeding and oviposition behavior and life cycle strategies of Diabrotica : An evolutionary view with implications for pest management 1. Environmental Entomology 10 , 826-831. Branson, T.F., Sutter, G.R., Fisher, J.R., 1982. Comparison of a tolerant and a susceptible maize inbred under artificial infestations of Diabrotica virgifera virgifera : yield and adult emergence. Environmental Entomology 11 , 371-372. Bray, J.R., Curtis, J.T., 1957. An ordination of the upland forest communities of Southern Wisconsin. Ecological Monographs 27 , 325-349. Bridge, P., Spooner, B., 2001. Soil fungi: diversity and detection. Plant and Soil 232 , 147-154. Brown, S.L., 1985. Quantitative risk assessment of environmental hazards. Annual Review of Public Health 6 , 247-267. Burges, H.D., 1998. Formulations of mycoinsecticides. In: Burges, H.D., (Ed.), Formulation of microbial pesticides: Beneficial microorganisms, nematodes and seed treatments. Kluwer Academic Publishers, Dordrecht, The Netherlands, pp. 131-185. Bürgmann, H., Pesaro, M., Widmer, F., Zeyer, J., 2001. A strategy for optimizing quality and quantity of DNA extracted from soil. Journal of Microbiological Methods 45 , 7-20. 133

REFERENCES

Burns, R.G., Stach, J.E.M., 2002. Microbial ecology of soil biofilms: Substrate bioavailability, bioremediation and complexity. Developments in Soil Science. Elsevier, pp. 17-42. CABI, 2017. Diabrotica virgifera virgifera (western corn rootworm). Invasive Species Compendium. CAB international, Wallingford, UK. Cai, Z., Rong, Y., Chen, J., Wang, J., Ma, J., Zhang, W., Zhao, X., 2016. Effects of the novel cis-nitromethylene neonicotinoid insecticide Paichongding on enzyme activities and microorganisms in yellow loam and Huangshi soils. Environmental Science and Pollution Research 23 , 7786-7793. Cai, Z., Wang, J., Ma, J., Zhu, X., Cai, J., Yang, G., 2015. Anaerobic degradation pathway of the novel chiral insecticide paichongding and its impact on bacterial communities in soils. Journal of Agricultural and Food Chemistry 63 , 7151-7160. Cao, G., Jia, M., Zhao, X., Wang, L., Tu, X., Wang, G., Nong, X., Zhang, Z., 2016a. Different effects of Metarhizium anisopliae strains IMI330189 and IBC200614 on enzymes activities and hemocytes of Locusta migratoria L. PloS One 11 , e0155257. Cao, H., Chen, R., Wang, L., Jiang, L., Yang, F., Zheng, S., Wang, G., Lin, X., 2016b. Soil pH, total phosphorus, climate and distance are the major factors influencing microbial activity at a regional spatial scale. Scientific Reports 6 , 25815. Carcinogens, U.S.I.S.G.o., 1986. Chemical carcinogens: a review of the science and its associated principles. Environmental Health Perspectives 67 , 201-282. Castro, T., Mayerhofer, J., Enkerli, J., Eilenberg, J., Meyling, N.V., Moral, R.d.A., Borges Demetrio, C.G., Delalibera, I., Jr., 2016a. Persistence of Brazilian isolates of the entomopathogenic fungi Metarhizium anisopliae and M. robertsii in strawberry crop soil after soil drench application. Agriculture Ecosystems & Environment 233 , 361-369. Castro, T., Mayerhofer, J., Enkerli, J., Eilenberg, J., Meyling, N.V., Moral, R.d.A., Demétrio, C.G.B., Delalibera Jr, I., 2016b. Persistence of Brazilian isolates of the entomopathogenic fungi Metarhizium anisopliae and M. robertsii in strawberry crop soil after soil drench application. Agriculture, Ecosystems & Environment 233 , 361-369. Cavaglieri, L., Orlando, J., Etcheverry, M., 2009. Rhizosphere microbial community structure at different maize plant growth stages and root locations. Microbiological Research 164 , 391-399. Chandler, L.D., 2003. Corn rootworm areawide management program: United States Department of Agriculture– Agricultural Research Service. Pest Management Science 59 , 605-608. Chen, H., 2016. VennDiagram: Generate high-resolution Venn and Euler plots. Chen, H., Boutros, P.C., 2011. VennDiagram: a package for the generation of highly-customizable Venn and Euler diagrams in R. BMC Bioinformatics 12 , 35. Chen, M.M., Zhu, Y.G., Su, Y.H., Chen, B.D., Fu, B.J., Marschner, P., 2007. Effects of soil moisture and plant interactions on the soil microbial community structure. European Journal of Soil Biology 43 , 31-38. Chiang, H.C., 1973. Bionomics of the Northern and Western corn rootworms. Annual Review of Entomology 18 , 44-72. Chu, H.L., Chen, W.H., Wen, T.C., Liang, Z.Q., Zheng, F.C., Liang, J.D., Han, Y.F., 2016. Delimitation of a novel member of genus Metarhizium (Clavicipitaceae) by phylogenetic and network analysis. Phytotaxa 288 , 51-60. Ciosi, M., Miller, N.J., Kim, K.S., Giordano, R., Estoup, A., Guillemaud, T., 2008. Invasion of Europe by the Western corn rootworm, Diabrotica virgifera virgifera : multiple transatlantic introductions with various reductions of genetic diversity. Molecular Ecology 17 , 3614-3627. Ciosi, M., Toepfer, S., Li, H., Haye, T., Kuhlmann, U., Wang, H., Siegfried, B., Guillemaud, T., 2009. European populations of Diabrotica virgifera virgifera are resistant to aldrin, but not to methyl-parathion. Journal of Applied Entomology 133 , 307-314. Clarke, K.R., 1993. Non-parametric multivariate analyses of changes in community structure. Australian Journal of Ecology 18 , 117-143. Clarke, K.R., Gorley, R.N., 2015. PRIMER v7: User Manual/Tutorial. PRIMER-E. Plymouth, pp. 296. Colson, I., Goldstein, D.B., 1999. Evidence for complex mutations at microsatellite loci in Drosophila . Genetics 152 , 617-627. Council Directive, 2009. Directive 2009/128/EC of the European parliament and of the council of 21 October 2009 establishing a framework for Community action to achieve the sustainable use of pesticides. Official Journal of the European Union, pp. 71-86. Crous, P.W., Gams, W., Stalpers, J.A., Robert, V., Stegehuis, G., 2004. MycoBank: an online initiative to launch mycology into the 21st century. Studies in Mycology 50 , 19-22. Curtis, H., Noll, U., Störmann, J., Slusarenko, A.J., 2004. Broad-spectrum activity of the volatile phytoanticipin allicin in extracts of garlic ( Allium sativum L.) against plant pathogenic bacteria, fungi and Oomycetes. Physiological and Molecular Plant Pathology 65 , 79-89. 134

REFERENCES de Boer, P., Caspers, M., Sanders, J.-W., Kemperman, R., Wijman, J., Lommerse, G., Roeselers, G., Montijn, R., Abee, T., Kort, R., 2015. Amplicon sequencing for the quantification of spoilage microbiota in complex foods including bacterial spores. Microbiome 3 , 30. Dematheis, F., Zimmerling, U., Flocco, C., Kurtz, B., Vidal, S., Kropf, S., Smalla, K., 2012. Multitrophic interaction in the rhizosphere of maize: Root feeding of Western corn rootworm larvae alters the microbial community composition. PloS One 7 , doi:10.1371/journal.pone.0037288. DeSantis, T.Z., Hugenholtz, P., Larsen, N., Rojas, M., Brodie, E.L., Keller, K., Huber, T., Dalevi, D., Hu, P., Andersen, G.L., 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Applied and Environmental Microbiology 72 , 5069-5072. Dessaux, Y., Chapelle, E., Faure, D., 2011. Quorum sensing and quorum quenching in soil ecosystems. In: Witzany, G., (Ed.), Biocommunication in Soil Microorganisms. Springer Berlin Heidelberg, Berlin, Heidelberg, pp. 339-367. Devkota, S., Cornejo, C., Werth, S., Chaudhary, R.P., Scheidegger, C., 2014. Characterization of microsatellite loci in the Himalayan lichen fungus Lobaria pindarensis (Lobariaceae). Applications in Plant Sciences 2, doi: 10.3732/apps.1300101. Devos, Y., Meihls, L.N., Kiss, J., Hibbard, B.E., 2013. Resistance evolution to the first generation of genetically modified Diabrotica -active Bt-maize events by western corn rootworm: Management and monitoring considerations. Transgenic Research 22 , 269-299. Domsch, K.H., Gams, W., Anderson, T.H., 1980. Compendium of soil fungi. Academic Press, London, UK, pp. 413-415. Driver, F., Milner, R.J., Trueman, J.W.H., 2000. A taxonomic revision of Metarhizium based on a phylogenetic analysis of rDNA sequence data. Mycological Research 104 , 134-150. Dromph, K.M., 2003. Collembolans as vectors of entomopathogenic fungi. Pedobiologia 47 , 245-256. Dromph, K.M., Vestergaard, S., 2002. Pathogenicity and attractiveness of entomopathogenic hyphomycete fungi to collembolans. Applied Soil Ecology 21 , 197-210. Dunn, O.J., 1964. Multiple comparisons using rank sums. Technometrics 6 , 241-252. Eckard, S., Ansari, M.A., Bacher, S., Butt, T.M., Enkerli, J., Grabenweger, G., 2014. Virulence of in vivo and in vitro produced conidia of Metarhizium brunneum strains for control of wireworms. Crop Protection 64 , 137-142. Eckard, S., Bacher, S., Enkerli, J., Grabenweger, G., 2017. A simple in vitro method to study interactions between soil insects, entomopathogenic fungi, and plant extracts. Entomologia Experimentalis et Applicata 163 , 315-327. Edgar, R.C., 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26 , 2460-2461. Edgar, R.C., 2013. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nature Methods 10 , 996-998. Edgar, R.C., Flyvbjerg, H., 2015. Error filtering, pair assembly and error correction for next-generation sequencing reads. Bioinformatics 31 , 3476-3482. Eilenberg, J., Hajek, A., Lomer, C., 2001. Suggestions for unifying the terminology in biological control. BioControl 46 , 387-400. Eldredge, N., Cracraft, J., 1980. Phylogenetic patterns and the evolutionary process: Method and theory in comparative biology. Columbia University Press, New York. Enkerli, J., Ghormade, V., Oulevey, C., Widmer, F., 2009. PCR - RFLP analysis of chitinase genes enables efficient genotyping of Metarhizium anisopliae var. a nisopliae . Journal of Invertebrate Pathology 102 , 185-188. Enkerli, J., Kolliker, R., Keller, S., Widmer, F., 2005. Isolation and characterization of microsatellite markers from the entomopathogenic fungus Metarhizium anisopliae . Molecular Ecology Notes 5 , 384-386. Enkerli, J., Schneider, S., Mayerhofer, J., Huang, Z., Sundh, I., Vidal, S., Widmer, F., 2016. Genetic diversity of Metarhizium spp. in grass, wheat, and forest habitats. 49th Annual Meeting of the Society of Invertebrate Pathology, Tours, France, pp. 108. Enkerli, J., Widmer, F., 2010. Molecular ecology of fungal entomopathogens: molecular genetic tools and their applications in population and fate studies. BioControl 55 , 17-37. EPA, 2007. Data requirements for biochemical and microbial pesticides. Ester, A., Huiting, H., 2007. Controlling wireworms ( Agriotes spp.) in a potato crop with biologicals. IOBC wprs Bulletin 30 , 189. European Commission, 2011. Commission regulation (EU) No 544/2011 of 10 June 2011 implementing Regulation (EC) No 1107/2009 of the European Parliament and of the Council as regards the data requirements for active substances. Official Journal of the European Union. European Commission, 2013. Commission Implementing Regulation (EU) No 485/2013 of 24 May 2013 amending Implementing Regulation (EU) No 540/2011, as regards the conditions of approval of the active 135

REFERENCES

substances clothianidin, thiamethoxam and imidacloprid, and prohibiting the use and sale of seeds treated with plant protection products containing those active substances. Official Journal of the European Union. European Commission, 2014. Commission Recommendation of 6 February 2014 on measures to control Diabrotica virgifera virgifera Le Conte in Union areas where its presence is confirmed (2014/63/EU). Official Journal of the European Union. Faith, D.P., Minchin, P.R., Belbin, L., 1987. Compositional dissimilarity as a robust measure of ecological distance. Vegetatio 69 , 57-68. Fang, W.G., Pei, Y., Bidochka, M.J., 2007. A regulator of a G protein signalling (RGS) gene, cag8, from the insect- pathogenic fungus Metarhizium anisopliae is involved in conidiation, virulence and hydrophobin synthesis. Microbiology Society 153 , 1017-1025. Faria, M.R.d., Wraight, S.P., 2007. Mycoinsecticides and Mycoacaricides: A comprehensive list with worldwide coverage and international classification of formulation types. Biological Control 43 , 237-256. Feng, X.M., Passoth, V., Eklund-Jonsson, C., Alminger, M.L., Schnurer, J., 2007. Rhizopus oligosporus and yeast co-cultivation during barley tempeh fermentation - Nutritional impact and real-time PCR quantification of fungal growth dynamics. Food Microbiology 24 , 393-402. Fisher, J.J., Rehner, S.A., Bruck, D.J., 2011. Diversity of rhizosphere associated entomopathogenic fungi of perennial herbs, shrubs and coniferous trees. Journal of Invertebrate Pathology 106 , 289-295. Ford, G.H., 1917. Observations on the larval and pupal stages of Agriotes obscurus , Linnaeus. Annals of Applied Biology 3 , 97-115. Franzosa, E.A., Hsu, T., Sirota-Madi, A., Shafquat, A., Abu-Ali, G., Morgan, X.C., Huttenhower, C., 2015. Sequencing and beyond: Integrating molecular 'omics' for microbial community profiling. Nature Reviews Microbiology 13 , 360-372. Freed, S., Jin, F.L., Ren, S.X., 2011. Determination of genetic variability among the isolates of Metarhizium anisopliae var. anisopliae from different geographical origins. World Journal of Microbiology & Biotechnology 27 , 359-370. Frey, B., Rime, T., Phillips, M., Stierli, B., Hajdas, I., Widmer, F., Hartmann, M., 2016. Microbial diversity in European alpine permafrost and active layers. FEMS Microbiology Ecology 92. Furlan, L., 1996. The biology of Agriotes usulatus Schäller (Col., Elateridae). I Adults and oviposition. Journal of Applied Entomology 120 , 269-274. Furlan, L., 1998. The biology of Agriotes ustulatus Schäller (Col., Elateridae). II. Larval development, pupation, whole cycle description and practical implications. Journal of Applied Entomology 122 , 71-78. Furlan, L., 2004. The biology of Agriotes sordidus Illiger (Col. Elateridae). Journal of Applied Entomology 128 , 696-706. Furlan, L., 2005. An IPM approach targeted against wireworms: what has been done and what has to be done. IOBC/wprs Bull 28 , 91-100. Furlan, L., Toffanin, F., 1998. Effectiveness of new insecticides used as seed dressing (imidacloprid and fipronil) against wireworms in controlled environment [Zea mays L.-Beta vulgaris L.-Veneto]. Atti delle Giornate Fitopatologiche (Italy). Furlan, L., Toth, M., Parker, W.E., Ivezić, M., Pančić, S., Brmež, M., Dobrinčić, R., Igrc Barčić, J., Muresan, F., Subchev, M., Toshova, T., Molnar, Z., Ditsch, B., Voigt, D., 2001. The efficacy of the new Agriotes sex pheromone traps in detecting wireworm population levels in different european countries. XXI IWGO Conference, Legnaro Italia,, pp. 293-304. Gao, Q., Jin, K., Ying, S.H., Zhang, Y., Xiao, G., Shang, Y., Duan, Z., Hu, X., Xie, X.Q., Zhou, G., Peng, G., Luo, Z., Huang, W., Wang, B., Fang, W., Wang, S., Zhong, Y., Ma, L.J., St. Leger, R.J., Zhao, G.P., Pei, Y., Feng, M.G., Xia, Y., Wang, C., 2011. Genome sequencing and comparative transcriptomics of the model entomopathogenic fungi Metarhizium anisopliae and M. acridum . PLoS Genetics 7 , e1001264. García de Salamone, I.E., Di Salvo, L.P., Escobar Ortega, J.S., Boa Sorte, P.M.F., Urquiaga, S., Teixeira, K.R.S., 2010. Field response of rice paddy crop to Azospirillum inoculation: physiology of rhizosphere bacterial communities and the genetic diversity of endophytic bacteria in different parts of the plants. Plant and Soil 336 , 351-362. Gerber, C.K., Edwards, C.R., Bledsoe, L.W., Gray, M.E., Steffey, K.L., Chandler, L.D., 2005. Application of the areawide concept using semiochemical-based insecticide baits for managing the western corn rootworm (Diabrotica virgifera virgifera LeConte) variant in the eastern Midwest. In: Vidal, S., Kuhlmann, U., Edwards, C.R., Eds.), Western corn rootworm: ecology and management. CABI publishing, Wallingford, pp. 221-238. Ginsberg, H.S., Lebrun, R.A., Heyer, K., Zhioua, E., 2002. Potential nontarget effects of Metarhizium anisopliae (Deuteromycetes) used for biological control of ticks (Acari: Ixodidae). Environmental Entomology 31 , 1191-1196.

136

REFERENCES

Glare, T.R., Moran-Diez, M.E., 2016. Microbial-Based Biopesticides - Methods and Protocols. Springer Protocols, New York, NY. Glasby, T.M., Underwood, A.J., 1996. Sampling to differentiate between pulse and press perturbations. Environmental Monitoring and Assessment 42 , 241-252. Goldstein, D.B., Pollock, D.D., 1997. Launching Microsatellites: A review of mutation processes and methods of phylogenetic inference. Journal of Heredity 88 , 335-342. Golo, P.S., Gardner, D.R., Grilley, M.M., Takemoto, J.Y., Krasnoff, S.B., Pires, M.S., Fernandes, É.K.K., Bittencourt, V.R.E.P., Roberts, D.W., 2014. Production of destruxins from Metarhizium spp. fungi in artificial medium and in endophytically colonized cowpea plants. PloS One 9 , e104946. Gonzalez, J.E., Keshavan, N.D., 2006. Messing with bacterial quorum sensing. Microbiology and Molecular Biology Reviews 70 , 859-875. Gower, J.C., 1966. Some distance properties of latent root and vector methods used in multivariate analysis. Biometrika 53 , 325-328. Gray, M.E., Sappington, T.W., Miller, N.J., Moeser, J., Bohn, M.O., 2009. Adaptation and invasiveness of western corn rootworm: Intensifying research on a worsening pest. Annual Review of Entomology, pp. 303-321. Griffiths, B.S., Philippot, L., 2013. Insights into the resistance and resilience of the soil microbial community. FEMS Microbiology Reviews 37 , 112-129. Grosch, R., Scherwinski, K., Lottmann, J., Berg, G., 2006. Fungal antagonists of the plant pathogen Rhizoctonia solani : selection, control efficacy and influence on the indigenous microbial community. Mycological Research 110 , 1464-1474. Guo, M., Gong, Z., Miao, R., Su, D., Li, X., Jia, C., Zhuang, J., 2017. The influence of root exudates of maize and soybean on polycyclic aromatic hydrocarbons degradation and soil bacterial community structure. Ecological Engineering 99 , 22-30. Haack, S.K., Garchow, H., Odelson, D.A., Forney, L.J., Klug, M.J., 1994. Accuracy, reproducibility, and interpretation of fatty acid methyl ester profiles of model bacterial communities. Applied and Environmental Microbiology 60 , 2483-2493. Hajek, A.E., 1997. Ecology of terrestrial fungal entomopathogens. In: Jones, J.G., (Ed.), Advances in Microbial Ecology, Vol 15. Plenum Publishing Corp, New York, pp. 193-249. Hajek, A.E., Humber, R.A., Elkinton, J.S., May, B., Walsh, S.R.A., Silver, J.C., 1990. Allozyme and restriction fragment length polymorphism analyses confirm Entomophaga maimaiga responsible for 1989 epizootics in North-American gypsy moth populations. Proceedings of the National Academy of Sciences, USA 87 , 6979-6982. Hajek, A.E., St Leger, R.J., 1994. Interactions between fungal pathogens and insect hosts. Annual Review of Entomology 39 , 293-322. Hall, R.A., Papierok, B., 1982. Fungi as biological control agents of arthropods of agricultural and medical importance. Parasitology 84 , 205-240. Hänel, H., 1982. The life cycle of the insect pathogenic fungus Metarhizium anisopliae in the termite Nasutitermes exitiosus . Mycopathologia 80 , 137-145. Hartmann, A., Rothballer, M., Schmid, M., 2008. Lorenz Hiltner, a pioneer in rhizosphere microbial ecology and soil bacteriology research. Plant and Soil 312 , 7-14. Hartmann, M., Frey, B., Kölliker, R., Widmer, F., 2005. Semi-automated genetic analyses of soil microbial communities: comparison of T-RFLP and RISA based on descriptive and discriminative statistical approaches. Journal of Microbiological Methods 61 , 249-360. Hartmann, M., Frey, B., Mayer, J., Mader, P., Widmer, F., 2015. Distinct soil microbial diversity under long-term organic and conventional farming. ISME Journal 9 , 1177-1194. Hartmann, M., Niklaus, P.A., Zimmermann, S., Schmutz, S., Kremer, J., Abarenkov, K., Luscher, P., Widmer, F., Frey, B., 2014. Resistance and resilience of the forest soil microbiome to logging-associated compaction. ISME Journal 8 , 226-244. Hartmann, M., Widmer, F., 2008. Reliability for detecting composition and changes of microbial communities by T-RFLP genetic profiling. FEMS Microbiology Ecology 63 , 249-260. Hattingh, M., Alexander, A., Meijering, I., van Reenen, C.A., Dicks, L.M.T., 2014. Malting of barley with combinations of Lactobacillus plantarum , Aspergillus niger , Trichoderma reesei , Rhizopus oligosporus and Geotrichum candidum to enhance malt quality. International Journal of Food Microbiology 173 , 36- 40. Herbold, C.W., Pelikan, C., Kuzyk, O., Hausmann, B., Angel, R., Berry, D., Loy, A., 2015. A flexible and economical barcoding approach for highly multiplexed amplicon sequencing of diverse target genes. Frontiers in Microbiology 6 , 731. Hervé, M., 2017. RVAideMemoire: Testing and Plotting Procedures for Biostatistics. R package version 0.9-66.

137

REFERENCES

Hicks, H., Blackshaw, R.P., 2008. Differential responses of three Agriotes click beetle species to pheromone traps. Agricultural and Forest Entomology 10 , 443-448. Hill, D.S., 1987. Agricultural insect pests of temperate regions and their control. Cambridge University Press, Cambridge, UK. Hill, G.T., Mitkowski, N.A., Aldrich-Wolfe, L., Emele, L.R., Jurkonie, D.D., Ficke, A., Maldonado-Ramirez, S., Lynch, S.T., Nelson, E.B., 2000. Methods for assessing the composition and diversity of soil microbial communities. Applied Soil Ecology 15 , 25-36. Hill, R.E., Hixson, E., Muma, M.H., 1948. Corn rootworm control tests with benzene hexachloride DDT, nitrogen fertilizers and crop rotations 1. Journal of Economic Entomology 41 , 392-401. Hiltner, L.t., 1904. Über neuere Erfahrungen und Probleme auf dem Gebiete der Bodenbakteriologie unter besonderer Berücksichtigung der Gründüngung und Brache. Arbeiten der Deutschen Landwirtschaftlichen Gesellschaft 98 , 59-78. Hirsch, J., Galidevara, S., Strohmeier, S., Devi, K.U., Reineke, A., 2013. Effects on diversity of soil fungal community and fate of an artificially applied Beauveria bassiana strain assessed through 454 pyrosequencing. Microbial Ecology 66 , 608-620. Hokkanen, H.M.T., Zeng, Q.-Q., Menzler-Hokkanen, I., 2003. Assessing the impacts of Metarhizium and Beauveria on bumblebees. In: Hokkanen, H.M.T., Hajek, A.E., Eds.), Environmental impacts of microbial insecticides: Need and methods for risk assessment. Springer Netherlands, Dordrecht, pp. 63- 71. Hollander, M., Wolfe, D.A., 1973. Nonparametric Statistical Methods. John Wiley & Sons, New York. Hozzank, A., Keller, S., Daniel, O., Schweizer, C., 2003. Impact of Beauveria brongniartii and Metarhizium anisopliae (Deuteromycotina: Hyphomycetes) on Lumbricus terrestris (Oligichaeta: Lumbricidae). International Organization for Biological and Integrated Control of Noxious Animals and Plants / wprs Bulletin in press. Hu, G., St Leger, J., 2002. Field studies using a recombinant mycoinsecticide ( Metarhizium anisopliae ) reveal that it is rhizosphere competent. Applied and Environmental Microbiology 68 , 6383-6387. Hu, X., Xiao, G., Zheng, P., Shang, Y., Su, Y., Zhang, X., Liu, X., Zhan, S., St. Leger, R.J., Wang, C., 2014. Trajectory and genomic determinants of fungal-pathogen speciation and host adaptation. Proceedings of the National Academy of Sciences 111 , 16796-16801. Huang, B., Humber, R.A., Li, S.G., Li, Z.Z., Hodge, K.T., 2005a. Further notes on the molecular taxonomy of Metarhizium . Mycotaxon 94 , 181-187. Huang, B., Li, C.R., Humber, R.A., Hodge, K.T., Fan, M.Z., Li, Z.Z., 2005b. Molecular evidence for the taxonomic status of Metarhizium taii and its teleomorph, Cordyceps taii (Hypocreales, Clavicipitaceae). Mycotaxon 94 , 137-147. Hugenholtz, P., 2002. Exploring prokaryotic diversity in the genomic era. Genome Biology 3 , 8. Humber, R.A., 2012. Identification of entomopathogenic fungi. In: Lacey, L.A., (Ed.), Manual of techniques in invertebrate pathology. Academic Press, London, pp. 151-187. Humbert, P., Przyklenk, M., Vemmer, M., Patel, A.V., 2017. Calcium gluconate as cross-linker improves survival and shelf life of encapsulated and dried Metarhizium brunneum and Saccharomyces cerevisiae for the application as biological control agents. Journal of Microencapsulation 34 , 47-56. Hummel, H.E., Dinnesen, S., Nedelev, T., Modic, S., Urek, G., Ulrichs, C., 2008. Diabrotica virgifera virgifera LeConte in confrontation mood: simultaneous geographical and host spectrum expansion in southeastern Slovenia. Mitteilungen der Deutschen Gesellschaft für Allgemeine und Angewandte Entomologie 16 , 127-130. Inglis, G.D., Duke, G.M., Goettel, M.S., Kabaluk, T.J., 2008. Genetic diversity of Metarhizium anisopliae var. anisopliae in southwestern British Columbia. Journal of Invertebrate Pathology 98 , 101-113. Inglis, G.D., Enkerli, J., Goettel, M.S., 2012. Laboratory techniques used for entomopathogenic fungi: Hypocreales. In: Lacey, L.A., (Ed.), Manual of Techniques in Invertebrate Pathology (Second Edition). Academic Press, San Diego, pp. 189-253. Jacobsen, C.S., Hjelmsø, M.H., 2014. Agricultural soils, pesticides and microbial diversity. Current Opinion in Biotechnology 27 , 15-20. Jakka, S.R.K., Shrestha, R.B., Gassmann, A.J., 2016. Broad-spectrum resistance to Bacillus thuringiensis toxins by western corn rootworm ( Diabrotica virgifera virgifera ). Scientific Reports 6 , doi:10.1038/srep27860. Jaronski, S.T., 2007. Soil ecology of the entomopathogenic Ascomycetes: a critical examination of what we (think) we know. In: Ekesi, S., Mainiania, N.K., Eds.), Use of Entomopathogenic Fungi in Biological Pest Management Research Signpost, Trivandrum India, pp. 91-143. Jaronski, S.T., 2010. Ecological factors in the inundative use of fungal entomopathogens. BioControl 55 , 159-185. Jaronski, S.T., Jackson, M.A., 2008. Efficacy of Metarhizium anisopliae microsclerotial granules. Biocontrol Science and Technology 18 , 849-863. 138

REFERENCES

Jenison, J.R., Shank, D.B., Penny, L.H., 1981. Root Characteristics of 44 Maize Inbreds Evaluated in Four Environments. Crop Science 21 , 233-237. Jones, M.R., Good, J.M., 2016. Targeted capture in evolutionary and ecological genomics. Molecular Ecology 25 , 185-202. Jones, S.E., Lennon, J.T., 2010. Dormancy contributes to the maintenance of microbial diversity. Proceedings of the National Academy of Sciences 107 , 5881-5886. Jost, L., 2006. Entropy and diversity. Oikos 113 , 363-375. Jukes, T.H., Cantor, C.R., 1969. Evolution of protein molecules. In: Munro, H.N., (Ed.), Mammalian Protein Metabolism. Academic Press, New York, pp. 21-132. Kaaya, G.P., 2000. Laboratory and field evaluation of entomogenous fungi for tick control. Annals of the New York Academy of Sciences 916 , 559-564. Kabaluk, T.J., Goettel, M.S., Erlandson, M.A., Ericsson, J.D., Duke, G.M., Vernon, B., 2005. Metarhizium anisopliae as a biological control for wireworms and a report of some other naturally-occurring parasites. IOBC/wprs Bulletin 28 , 109-115. Kabaluk, T.J., Lafontaine, J.P., Borden, J.H., 2015. An attract and kill tactic for click beetles based on Metarhizium brunneum and a new formulation of sex pheromone. Journal of Pest Science 88 , 707-716. Kabaluk, T.J., Vernon, R.S., Goettel, M.S., 2007. Mortality and infection of wireworm, Agriotes obscurus [Coleoptera: Elateridae], with inundative field applications of Metarhizium anisopliae . Phytoprotection 88 , 51-56. Kahler, A.L., Olness, A.E., Sutter, G.R., Dybing, C.D., Devine, O.J., 1985. Root Damage by Western Corn Rootworm and Nutrient Content in Maize1. Agronomy Journal 77 , 769-774. Keiser, A., 2007. Influence of farming systems, specific cultivation methods and site parameters on potato quality. ETH, Zürich. Keiser, A., Häberli, M., Stamp, P., 2012. Drycore Appears to Result from an Interaction between Rhizoctonia solani and Wireworm (Agriotes ssp.)—Evidence from a 3-Year Field Survey. Potato Research 55 , 59-67. Keller, S., Zimmermann, G., 1989. Mycopathogens of soil insects. In: Wilding, N., Collins, N.M., Hammond, P.M., Webber, J.F., Eds.), Insect-fungus interactions. Academic Press, London. Kennedy, A.C., 1999. Bacterial diversity in agroecosystems. Agriculture, Ecosystems & Environment 74 , 65-76. Kepler, R.M., Chen, Y., Kilcrease, J., Shao, J., Rehner, S.A., 2016. Independent origins of diploidy in Metarhizium . Mycologia. Kepler, R.M., Humber, R.A., Bischoff, J.F., Rehner, S.A., 2014. Clarification of generic and species boundaries for Metarhizium and related fungi through multigene phylogenetics. Mycologia 106 , 811-829. Kepler, R.M., Rehner, S.A., 2013. Genome-assisted development of nuclear intergenic sequence markers for entomopathogenic fungi of the Metarhizium anisopliae species complex. Molecular Ecology Resources 13 , 210-217. Kepler, R.M., Sung, G.H., Ban, S., Nakagiri, A., Chen, M.J., Huang, B., Li, Z., Spatafora, J.W., 2012. New teleomorph combinations in the entomopathogenic genus Metacordyceps . Mycologia 104 , 182-197. Kepler, R.M., Ugine, T.A., Maul, J.E., Cavigelli, M.A., Rehner, S.A., 2015. Community composition and population genetics of insect pathogenic fungi in the genus Metarhizium from soils of a long-term agricultural research system. Environmental Microbiology , doi: 10.1111/1462-2920.12778. Khan, A.L., Hamayun, M., Khan, S.A., Kang, S.M., Shinwari, Z.K., Kamran, M., Ur Rehman, S., Kim, J.G., Lee, I.J., 2012. Pure culture of Metarhizium anisopliae LHL07 reprograms soybean to higher growth and mitigates salt stress. World Journal of Microbiology and Biotechnology 28 , 1483-1494. Kirchmair, M., Neuhauser, S., Huber, L., Strasser, H., 2008. The impact of soil treatment on soil mycobiota. Insect Pathogens and Insect Parasitic Nematodes IOBC/wprs Bulletin 31 , 239-244. Kiss, J., Komáromi, J., Bayar, K., Edwards, C.R., Hatala-Zséllar, I., 2005. Western corn rootworm ( Diabrotica virgifera virgifera LeConte) and the crop rotation systems in Europe. In: Vidal, S., Kuhlmann, U., Edwards, C.R., Eds.), Western Corn Rootworm: Ecology and Management. CABI publishing, Wallingford, pp. 189-220. Klemm, D., Heublein, B., Fink, H.-P., Bohn, A., 2005. Cellulose: Fascinating Biopolymer and Sustainable Raw Material. Angewandte Chemie International Edition 44 , 3358-3393. Knudsen, G.R., Shotzko, D.J., Krag, C.R., 1994. Fungal entomopathogen effect on numbers and spatial patterns of the Russian wheat aphid (Homoptera: Aphididae) on preferred and nonpreferred host plants. Environmental Entomology 23 , 1558-1567. Kölliker, U., Biasio, L., Jossi, W., 2011. Potential control of Swiss wireworms with entomopathogenic fungi. IOBC/wprs Bulletin 66 , 517-520. Kölliker, U., Jossi, W., Kuske, S., 2009. Optimised protocol for wireworm rearing. IOBC/wprs Bulletin, pp. 457- 460.

139

REFERENCES

Konopka, A., Lindemann, S., Fredrickson, J., 2015. Dynamics in microbial communities: unraveling mechanisms to identify principles. ISME Journal 9 , 1488-1495. Krassilstschik, J., 1888. La production industrielle des parasites végétaux pour la destruction des insectes nuisibles. Bulletin Biologique de la France et de la Belgique 19 , 461-472. Kröber, M., Wibberg, D., Grosch, R., Eikmeyer, F., Verwaaijen, B., Chowdhury P, S.P., Hartmann, A., Pühler, A., Schlüter, A., 2014. Effect of the strain Bacillus amyloliquefaciens FZB42 on the microbial community in the rhizosphere of lettuce under field conditions analyzed by whole metagenome sequencing. Frontiers in Microbiology 5 , 252. Krysan, J.L., Jackson, J.J., Lew, A.C., 1984. Field Termination of Egg Diapause in Diabrotica with New Evidence of Extended Diapause in D. barberi (Coleoptera: Chrysomelidae). Environmental Entomology 13 , 1237- 1240. Krysan, J.L., Miller, T.A., 1986. Methods for the study of pest Diabrotica . Springer, New York. Krysan, J.L., Smith, R.F., 1987. Systematics of the virgifera species group of Diabrotica (Coleoptera: Chrysomelidae: Galerucinae). Entomography 5 , 375-484. Kudryavtsev, I., Siirde, K., LÄÄts, K., Ismailov, V., Pristavko, V., 1993. Determination of distribution of harmful click beetle species (Coleoptera, Elateridae) by synthetic sex pheromones. Journal of Chemical Ecology 19 , 1607-1611. Kuhlmann, U., Toepfer, S., Zhang, F., 2005. Is classical biological control against western corn rootworm in Europe a potential sustainable management strategy? In: Vidal, S., Kuhlmann, U., Edwards, C.R., Eds.), Western Corn Rootworm: Ecology and Management. CABI publishing, Wallingford. Lacey, L.A., Frutos, R., Kaya, H.K., Vail, P., 2001. Insect Pathogens as Biological Control Agents: Do They Have a Future? Biological Control 21 , 230-248. Lacey, L.A., Grzywacz, D., Shapiro-Ilan, D.I., Frutos, R., Brownbridge, M., Goettel, M.S., 2015. Insect pathogens as biological control agents: Back to the future. Journal of Invertebrate Pathology 132 , 1-41. Ladurner, E., Quentin, U., Franceschini, S., Benuzzi, M., Ehlers, R., 2009. Efficacy evaluation of the entomopathogenic fungus Beauveria bassiana strain ATCC 74040 against wireworms ( Agriotes spp.) on potato. IOBC/WPRS Bull 45 , 445-448. Laengle, T., Pernfuss, B., Seger, C., Strasser, H., 2005. Field efficacy evaluation of Beauveria brongniartii against Melolontha melolontha in potato cultures. Sydowia 57 , 54-93. Lafrance, J., 1968. The seasonal movements of wireworms (Coleoptera: Elateridae) in relation to soil moisture and temperature in the organic soil of southwestern Quebec. Canadian Entomologist 100 , 801-807. Landl, M., Glauninger, J., 2013. Preliminary investigations into the use of trap crops to control Agriotes spp. (Coleoptera: Elateridae) in potato crops. Journal of Pest Science 86 , 85-90. Lauber, C.L., Ramirez, K.S., Aanderud, Z., Lennon, J., Fierer, N., 2013. Temporal variability in soil microbial communities across land-use types. ISME Journal 7 , 1641-1650. Lawson, L.D., 1998. Garlic: A review of its medicinal effects and indicated active compounds. American Chemical Society. Leal, S.C.M., Bertioli, D.J., Butt, T.M., Carder, J.H., Burrows, P.R., Peberdy, J.F., 1997. Amplification and restriction endonuclease digestion of the Pr1 gene for the detection and characterization of Metarhizium strains. Mycological Research 101 , 257-265. Legendre, P., Legendre, L., 2012. Numerical Ecology. Elselvier Science, Amsterdam. Lennon, J.T., Aanderud, Z.T., Lehmkuhl, B.K., Schoolmaster, D.R., 2012. Mapping the niche space of soil microorganisms using taxonomy and traits. Ecology 93 , 1867-1879. Lerner, A., Herschkovitz, Y., Baudoin, E., Nazaret, S., Moenne-Loccoz, Y., Okon, Y., Jurkevitch, E., 2006. Effect of Azospirillum brasilense inoculation on rhizobacterial communities analyzed by denaturing gradient gel electrophoresis and automated ribosomal intergenic spacer analysis. Soil Biology and Biochemistry 38 , 1212-1218. Levine, E., Oloumi-Sadeghi, H., 1991. Management of diabroticite rootworms in corn. Annual Review of Entomology 36 , 229-255. Levine, E., Spencer, J.L., Isard, S.A., Onstad, D.W., Gray, M.E., 2002. Adaptation of the Western corn rootworm to crop rotation: Evolution of a new strain in response to a management practice. American Entomologist 48 , 94-107. Li, J., Ying, S.H., Shan, L.T., Feng, M.G., 2010. A new non-hydrophobic cell wall protein (CWP10) of Metarhizium anisopliae enhances conidial hydrophobicity when expressed in Beauveria bassiana . Applied Microbiology and Biotechnology 85 , 975-984. Lindahl, B.D., Nilsson, R.H., Tedersoo, L., Abarenkov, K., Carlsen, T., Kjoller, R., Koljalg, U., Pennanen, T., Rosendahl, S., Stenlid, J., Kauserud, H., 2013. Fungal community analysis by high-throughput sequencing of amplified markers - a user's guide. New Phytologist 199 , 288-299.

140

REFERENCES

Liu, L., Li, Y., Li, S., Hu, N., He, Y., Pong, R., Lin, D., Lu, L., Law, M., 2012. Comparison of Next-Generation Sequencing systems. Journal of Biomedicine and Biotechnology 2012 , 11. Lopes, R.B., Faria, M., Souza, D.A., Bloch, C., Silva, L.P., Humber, R.A., 2014. MALDI-TOF mass spectrometry applied to identifying species of insect-pathogenic fungi from the Metarhizium anisopliae complex. Mycologia 106 , 865-878. Lugtenberg, B., 2015a. Life of Microbes in the Rhizosphere. In: Lugtenberg, B., (Ed.), Principles of Plant-Microbe Interactions: Microbes for Sustainable Agriculture. Springer International Publishing, Cham, pp. 7-15. Lugtenberg, B., 2015b. Principles of plant-microbe interactions: Microbes for sustainable agriculture. Springer International Publishing, Cham. Lynch, J.M., Bragg, E., 1985. Microorganisms and soil aggregate stability. In: Steward, B.A., (Ed.), Advances in Soil Sciences. Springer Verlag, New York. Mamanova, L., Coffey, A.J., Scott, C.E., Kozarewa, I., Turner, E.H., Kumar, A., Howard, E., Shendure, J., Turner, D.J., 2010. Target-enrichment strategies for next-generation sequencing. Nature Methods 7 , 111-118. Marquez, M., Iturriaga, E.A., Quesada-Moraga, E., Santiago-Alvarez, C., Monte, E., Hermosa, R., 2006. Detection of potentially valuable polymorphisms in four group I intron insertion sites at the 3 '-end of the LSU rDNA genes in biocontrol isolates of Metarhizium anisopliae . BMC Microbiology 6. Martin, M., 2011. Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet.journal 17 , e10-12. Maskell, F.E., 1958. Aldrin and gamma-BHC for the control of wireworm damage to potatoes. Plant Pathology 7 , 98-100. Mayden, R.L., 1997. A hierarchy of species concepts: the denouement in the saga of the species problem. In: Claridge, M.F., Dawah, H.A., Wilson, M.R., Eds.), Species: The units of diversity. Chapman & Hall, pp. 381–423. Mayerhofer, J., Eckard, S., Hartmann, M., Grabenweger, G., Widmer, F., Leuchtmann, A., Enkerli, J., 2017. Assessing effects of the entomopathogenic fungus Metarhizium brunneum on soil microbial communities in Agriotes spp. biological pest control. FEMS Microbiology Ecology 93 , doi:10.1093/femsec/fix1117. Mayerhofer, J., Lutz, A., Widmer, F., Rehner, S.A., Leuchtmann, A., Enkerli, J., 2015. Multiplexed microsatellite markers for seven Metarhizium species. Journal of Invertebrate Pathology 132 , 132-134. McCauley, V.J.E., Zacharuk, R.Y., Tinline, R.D., 1968. Histopathology of green muscardine in larvae of four species of elateridae (Coleoptera). Journal of Invertebrate Pathology 12 , 444-459. McDonald, D., Price, M.N., Goodrich, J., Nawrocki, E.P., DeSantis, T.Z., Probst, A., Andersen, G.L., Knight, R., Hugenholtz, P., 2012. An improved Greengenes taxonomy with explicit ranks for ecological and evolutionary analyses of bacteria and archaea. ISME Journal 6 , 610-618. McNeill, J., Buck, W.R., Demoulin, V., Greuter, W., Hawkworth, D.L., Herendeen, P.S., Knapp, S., Marhold, K., Prado, J., Prud'homme van Reine, W.F., Smith, G.F., Wiersema, J.H., Turland, N.J., 2012. International code of nomenclature for algae, fungi and plants (Melbourne code); adopted by the 18th International Botanical Congress. International Botanical Congress. Koeltz Scientific Books, Melbourne, Australia. Meinke, L.J., Siegfried, B.D., Wright, R.J., Chandler, L.D., 1998. Adult susceptibility of Nebraska Western corn rootworm (Coleoptera: Chrysomelidae) populations to selected insecticides. Journal of Economic Entomology 91 , 594-600. Merrill, J.L.G., 1952. Reduction of Wireworm Damage to Potatoes. Journal of Economic Entomology 45 , 548- 549. Metcalf, R.L., 1983. Implications and Prognosis of Resistance to Insecticides. In: Georghiou, G.P., Saito, T., Eds.), Pest Resistance to Pesticides. Springer US, Boston, MA, pp. 703-733. Metcalf, R.L., 1986. Foreword. In: Krysan, J.L., Miller, T.A., Eds.), Methods for the Study of Pest Diabrotica. Springer, New York, pp. VII-XV. Metchnikoff, E., 1879. Maladies des hannetons duble. Zapiski imperatorskogo obshcestaa sel'skago Khozyaistra yuzhnoi rossii , 17-50. Meyling, N.V., Eilenberg, J., 2007. Ecology of the entomopathogenic fungi Beauveria bassiana and Metarhizium anisopliae in temperate agroecosystems: potential for conservation biological control. Biological Control 43 , 145-155. Miles, H., Cohen, M., 1941. Investigations on Wireworms and their Control. Report for the Years 1939 and 1940. Investigations on Wireworms and their Control. Report for the Years 1939 and 1940. Miles, H.W., 1942. Wireworms and agriculture, with special reference to Agriotes obscurus L. Annals of Applied Biology 29 , 176-180. Miller, N., Estoup, A., Toepfer, S., Bourguet, D., Lapchin, L., Derridj, S., Kim, K.S., Reynaud, P., Furlan, L., Guillemaud, T., 2005. Multiple transatlantic introductions of the Western corn rootworm. Science 310 , 992-992.

141

REFERENCES

Minchin, P.R., 1987. An evaluation of the relative robustness of techniques for ecological ordination. Vegetatio 69 , 89-107. Moll, J., Klingenfuss, F., Widmer, F., Gogos, A., Bucheli, T.D., Hartmann, M., van der Heijden, M.G.A., 2017. Effects of titanium dioxide nanoparticles on soil microbial communities and wheat biomass. Soil Biology and Biochemistry 111 , 85-93. Montalva, C., Collier, K., Rocha, L.F.N., Inglis, P.W., Lopes, R.B., Luz, C., Humber, R.A., 2016. A natural fungal infection of a sylvatic cockroach with Metarhizium blattodeae sp nov., a member of the M. flavouiride species complex. Fungal Biology 120 , 655-665. Mulligan, R.A., Tomco, P.L., Howard, M.W., Schempp, T.T., Stewart, D.J., Stacey, P.M., Ball, D.B., Tjeerdema, R.S., 2016. Aerobic versus anaerobic microbial degradation of clothianidin under simulated california rice field conditions. Journal of Agricultural and Food Chemistry 64 , 7059-7067. Muyzer, G., 1999. Genetic fingerprinting of microbial communities - present status and future perspectives. Methods of microbial community analysis., 8th international symposium on microbial ecology, Halifax, Canada. Nannipieri, P., Ascher, J., Ceccherini, M.T., Landi, L., Pietramellara, G., Renella, G., 2003. Microbial diversity and soil functions. European Journal of Soil Science 54 , 655-670. Narva, K.E., Siegfried, B.D., Storer, N.P., 2013. Transgenic approaches to western corn rootworm control. Advances in Biochemical Engineering/Biotechnology 136 , 135-162. Naveed, M., Herath, L., Moldrup, P., Arthur, E., Nicolaisen, M., Norgaard, T., Ferré, T.P.A., de Jonge, L.W., 2016. Spatial variability of microbial richness and diversity and relationships with soil organic carbon, texture and structure across an agricultural field. Applied Soil Ecology 103 , 44-55. Nei, M., Roychoudhury, A.K., 1974. Sampling Variances of Heterozygosity and Genetic Distance. Genetics 76 , 379-390. Ni, J., Yan, Q., Yu, Y., 2013. How much metagenomic sequencing is enough to achieve a given goal? Scientific Reports 3 , 1968. Nikolenko, S.I., Korobeynikov, A.I., Alekseyev, M.A., 2013a. BayesHammer: Bayesian clustering for error correction in single-cell sequencing. BMC Genomics 14 Suppl 1 , S7. Nikolenko, S.I., Korobeynikov, A.I., Alekseyev, M.A., 2013b. BayesHammer: Bayesian clustering for error correction in single-cell sequencing. BMC Genomics 14 , 1-11. Nishi, O., Hasegawa, K., Iiyama, K., Yasunaga-Aoki, C., Shimizu, S., 2011. Phylogenetic analysis of Metarhizium spp. isolated from soil in Japan. Applied Entomology and Zoology 46 , 301-309. Noronha, C., 2011. Crop rotations a management tool for wire-worms in potatoes. IOBC/WPRS Bull 66 , 467-471. Nurk, S., Bankevich, A., Antipov, D., Gurevich, A., Korobeynikov, A., Lapidus, A., Prjibelsky, A., Pyshkin, A., Sirotkin, A., Sirotkin, Y., Stepanauskas, R., McLean, J., Lasken, R., Clingenpeel, S.R., Woyke, T., Tesler, G., Alekseyev, M.A., Pevzner, P.A., 2013. Assembling genomes and mini-metagenomes from highly chimeric reads. In: Deng, M., Jiang, R., Sun, F., Zhang, X., Eds.), Research in Computational Molecular Biology. RECOMB 2013. Lecture Notes in Computer Science. Springer, Berlin, Heidelberg, pp. 158- 170. O'Brien, S.L., Gibbons, S.M., Owens, S.M., Hampton ‐Marcell, J., Johnston, E.R., Jastrow, J.D., Gilbert, J.A., Meyer, F., Antonopoulos, D.A., 2016. Spatial scale drives patterns in soil bacterial diversity. Environmental Microbiology 18 , 2039-2051. Oehl, F., Sieverding, E., Ineichen, K., Mäder, P., Boller, T., Wiemken, A., 2003. Impact of land use intensity on the species diversity of arbuscular mycorrhizal fungi in agroecosystems of Central Europe. Applied and Environmental Microbiology 69 , 2816-2824. Ogle, D.H., 2016. FSA: Fisheries Stock Analysis. R package version 0.8.7. Oksanen, J., Blanchet, G.F., Friendly, M., Kindt, R., Legendre, P., McGlinn, D., Michin, P.R., O'Hara, R.B., Wagner, H., 2016. vegan: Community Ecology Package. R package version 2.4-0. Oleson, J.D., Park, Y.L., Nowatzki, T.M., Tollefson, J.J., 2005. Node-injury scale to evaluate root injury by corn rootworms (Coleoptera: Chrysomelidae). Journal of Economic Entomology 98 , 1-8. Orlob, G.B., 1973. Ancient and medieval plant pathology. Bayer Pflanzenschutz, Leverkusen. Orti, G., Pearse, D.E., Avise, J.C., 1997. Phylogenetic assessment of length variation at a microsatellite locus. Proceedings of the National Academy of Sciences 94 , 10745-10749. Oulevey, C., Widmer, F., Kölliker, R., Enkerli, J., 2009. An optimized microsatellite marker set for detection of Metarhizium anisopliae genotype diversity on field and regional scales. Mycological Research 113 , 1016- 1024. Pace, N.R., Stahl, D.A., Lane, D.J., Olsen, G.J., 1986. The Analysis of Natural Microbial Populations by Ribosomal RNA Sequences. In: Marshall, K.C., (Ed.), Advances in Microbial Ecology. Springer US, Boston, MA, pp. 1-55.

142

REFERENCES

Parker, W.E., 1994. Evaluation of the use of food baits for detecting wireworms ( Agriotes spp., Coleoptera: Elateridae) in fields intended for arable crop production. Crop Protection 13 , 271-276. Parker, W.E., 1996. The development of baiting techniques to detect wireworms ( Agriotes spp., Coleoptera: Elateridae) in the field, and the relationship between bait-trap catches and wireworm damage to potato. Crop Protection 15 , 521-527. Parker, W.E., Howard, J.J., 2001. The biology and management of wireworms ( Agriotes spp.) on potato with particular reference to the U.K. Agricultural and Forest Entomology 3 , 85-98. Pattemore, J.A., Hane, J.K., Williams, A.H., Wilson, B.A., Stodart, B.J., Ash, G.J., 2014. The genome sequence of the biocontrol fungus Metarhizium anisopliae and comparative genomics of Metarhizium species. BMC Genomics 15 , doi: 10.1186/1471-2164-1115-1660. Paula, A.R., Carolino, A.T., Paula, C.O., Samuels, R.I., 2011. The combination of the entomopathogenic fungus Metarhizium anisopliae with the insecticide Imidacloprid increases virulence against the dengue vector Aedes aegypti (Diptera: Culicidae). Parasites & Vectors. BioMed Central, pp. 8. Peterson, B.K., Weber, J.N., Kay, E.H., Fisher, H.S., Hoekstra, H.E., 2012. Double digest RADseq: An inexpensive method for de novo snp discovery and genotyping in model and non-model species. PloS One 7 , e37135. Philippot, L., Čuhel, J., Saby, N.P.A., Chèneby, D., Chroňáková, A., Bru, D., Arrouays, D., Martin‐Laurent, F., Šimek, M., 2009. Mapping field ‐scale spatial patterns of size and activity of the denitrifier community. Environmental Microbiology 11 , 1518-1526. Pilz, C., Keller, S., Kuhlmann, U., Toepfer, S., 2009. Comparative efficacy assessment of fungi, nematodes and insecticides to control Western corn rootworm larvae in maize. BioControl 54 , 671-684. Pilz, C., Wegensteiner, R., Keller, S., 2008. Natural occurrence of insect pathogenic fungi and insect parasitic nematodes in Diabrotica virgifera virgifera populations. BioControl 53 , 353-359. Pimm, S.L., 1984. The complexity and stability of ecosystems. Nature 307 , 321-326. Poll, C., Marhan, S., Back, F., Niklaus, P.A., Kandeler, E., 2013. Field-scale manipulation of soil temperature and precipitation change soil CO 2 flux in a temperate agricultural ecosystem. Agriculture, Ecosystems & Environment 165 , 88-97. Polz, M.F., Cavanaugh, C.M., 1998. Bias in template-to-product ratios in multitemplate pcr. Applied and Environmental Microbiology 64 , 3724-3730. Price, P.W., 1988. An overview of organismal interactions in ecosystems in evolutionary and ecological time. Agriculture, Ecosystems & Environment 24 , 369-377. Prior, C., 1996. History, use and future of microbial insecticides in urban pest control. Second International Conference on Urban Pest, International Institute of Biological control, Ascot, UK. Quail, M.A., Smith, M., Coupland, P., Otto, T.D., Harris, S.R., Connor, T.R., Bertoni, A., Swerdlow, H.P., Gu, Y., 2012. A tale of three next generation sequencing platforms: comparison of Ion Torrent, Pacific Biosciences and Illumina MiSeq sequencers. BMC Genomics 13 , 341. R-Core-Team, 2016. R: A language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. R-Development-Core-Team, 2008. R: A language and environment for statistical computing. In: Team, R.D.C., (Ed.), R Development Core Team. R Foundation for Statistical Computing, Vienna, Austria. Rai, V., Singh, D.K., 2002. Impact of fungal biopesticide ( Beauveria bassiana and B. brongniartii ) on soil fertility in groundnut field. Pesticide Research Journal 14 , 83-91. Rastogi, G., Sani, R.K., 2011. Molecular techniques to assess microbial community structure, function, and dynamics in the environment. In: Ahmad, I., Ahmad, F., Pichtel, J., Eds.), Microbes and Microbial Technology: Agricultural and Environmental Applications. Springer New York, New York, pp. 29-57. Rauch, H., Steinwender, B.M., Mayerhofer, J., Sigsgaard, L., Eilenberg, J., Enkerli, J., Zelger, R., Strasser, H., 2017. Field efficacy of Heterorhabditis bacteriophora (Nematoda: Heterorhabditidae), Metarhizium brunneum (Hypocreales: Clavicipitaceae), and chemical insecticide combinations for Diabrotica virgifera virgifera larval management. Biological Control 107 , 1-10. Rauch, H., Zelger, R., Strasser, H., 2016. Highly efficient field emergence trap for quantitative adult Western corn rootworm monitoring. Journal of the Kansas Entomological Society 89 , 256-266. Reddy, G.V.P., Tangtrakulwanich, K., 2014. Potential application of pheromones in monitoring, mating disruption, and control of click beetles (Coleoptera: Elateridae). ISRN Entomology 2014 , 8. Rehner, S.A., 2009. Molecular systematics of entomopathogenic fungi. In: Stock, S.P., Vandenberg, J., Glazer, I., Boemare, N., Eds.), Insect pathogens. Molecular approaches and techniques. CAB International, Wallingford, UK, pp. 145-165. Rehner, S.A., Kepler, R.M., 2017. Species limits, phylogeography and reproductive mode in the Metarhizium anisopliae complex. Journal of Invertebrate Pathology 148 , 60-66.

143

REFERENCES

Rhoads, A., Au, K.F., 2015. PacBio Sequencing and Its Applications. Genomics, Proteomics & Bioinformatics 13 , 278-289. Riah-Anglet, W., Trinsoutrot-Gattin, I., Martin-Laurent, F., Laroche-Ajzenberg, E., Norini, M.P., Latour, X., Laval, K., 2015. Soil microbial community structure and function relationships: A heat stress experiment. Applied Soil Ecology 86 , 121-130. Rime, T., Hartmann, M., Stierli, B., Anesio, A.M., Frey, B., 2016. Assimilation of microbial and plant carbon by active prokaryotic and fungal populations in glacial forefields. Soil Biology and Biochemistry 98 , 30-41. Ritter, C., Katroschan, K.U., Richter, E., 2011. Alternative methods to control wireworms ( Agriotes spp. Coleoptera: Elateridae) in vegetable production - potential of calcium cyanamide and Metarhizium anisopliae . IOBC/wprs Bulletin 66 , 521-524. Ritter, C., Richter, E., 2013. Control methods and monitoring of Agriotes wireworms (Coleoptera: Elateridae). Journal of Plant Diseases and Protection 120 , 4-15. Ritz, K., 2004. Interactions between soil structure and fungi. Mycologist 18 , 52-59. Ritz, K., Young, I.M., 2004. Interactions between soil structure and fungi. Mycologist 18 , 52-59. Rocha, L.F., Inglis, P.W., Humber, R.A., Kipnis, A., Luz, C., 2013. Occurrence of Metarhizium spp. in Central Brazilian soils. Journal of Basic Microbiology 53 , 251-259. Roesch, L.F., Fulthorpe, R.R., Riva, A., Casella, G., Hadwin, A.K., Kent, A.D., Daroub, S.H., Camargo, F.A., Farmerie, W.G., Triplett, E.W., 2007. Pyrosequencing enumerates and contrasts soil microbial diversity. ISME Journal 1 , 283-290. Rogge, S.A., Mayerhofer, J., Enkerli, J., Bacher, S., Grabenweger, G., 2017. Preventive application of an entomopathogenic fungus in cover crops for wireworm control. BioControl , 1-11. Rousidou, C., Papadopoulou, E.S., Kortsinidou, M., Giannakou, I.O., Singh, B.K., Menkissoglu-Spiroudi, U., Karpouzas, D.G., 2013. Bio-pesticides: Harmful or harmless to ammonia oxidizing microorganisms? The case of a Paecilomyces lilacinus -based nematicide. Soil Biology and Biochemistry 67, 98-105. Rousk, J., Baath, E., Brookes, P.C., Lauber, C.L., Lozupone, C., Caporaso, J.G., Knight, R., Fierer, N., 2010. Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME Journal 4 , 1340-1351. RStudio-Team, 2015. RStudio: Integrated Development for R. RStudio, Inc., Boston. Rykiel, E.J., 1985. Towards a definition of ecological disturbance. Australian Journal of Ecology 10 , 361-365. Sanders, I.R., 2003. Specificity in the arbuscular mycorrhizal symbiosis. In: van der Heijden, M.G.A., Sanders, I.R., Eds.), Mycorrhizal Ecology. Springer, Berlin, Heidelberg, pp. 415-437. Sasan, R.K., Bidochka, M.J., 2012. The insect-pathogenic fungus Metarhizium robertsii (Clavicipitaceae) is also an endophyte that stimulates plant root development. American Journal of Botany 99 , 101-107. Savazzini, F., Longa, C.M.O., Pertot, I., 2009. Impact of the biocontrol agent Trichoderma atroviride SC1 on soil microbial communities of a vineyard in northern Italy. Soil Biology and Biochemistry 41 , 1457-1465. Sboner, A., Mu, X.J., Greenbaum, D., Auerbach, R.K., Gerstein, M.B., 2011. The real cost of sequencing: higher than you think! Genome Biology 12 , 125. Schabel, H.G., 1978. Percutaneous infection of Hylobius pales by Metarrhizium anisopliae . Journal of Invertebrate Pathology 31 , 180-187. Scheepmaker, J.W.A., Butt, T.M., 2010. Natural and released inoculum levels of entomopathogenic fungal biocontrol agents in soil in relation to risk assessment and in accordance with EU regulations. Biocontrol Science and Technology 20 , 503-552. Schirmer, M., Ijaz, U.Z., D'Amore, R., Hall, N., Sloan, W.T., Quince, C., 2015. Insight into biases and sequencing errors for amplicon sequencing with the Illumina MiSeq platform. Nucleic Acids Research 43 , e37-e37. Schloss, P.D., Westcott, S.L., Ryabin, T., Hall, J.R., Hartmann, M., Hollister, E.B., Lesniewski, R.A., Oakley, B.B., Parks, D.H., Robinson, C.J., Sahl, J.W., Stres, B., Thallinger, G.G., Van Horn, D.J., Weber, C.F., 2009. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Applied and Environmental Microbiology 75 , 7537- 7541. Schmidt, C.S., Alavi, M., Cardinale, M., Müller, H., Berg, G., 2012. Stenotrophomonas rhizophila DSM14405 T promotes plant growth probably by altering fungal communities in the rhizosphere. Biology and Fertility of Soils 48 , 947-960. Schneider, S., Rehner, S.A., Widmer, F., Enkerli, J., 2011. A PCR-based tool for cultivation-independent detection and quantification of Metarhizium clade 1. Journal of Invertebrate Pathology 108 , 106-114. Schneider, S., Widmer, F., Jacot, K., Kölliker, R., Enkerli, J., 2012. Spatial distribution of Metarhizium clade 1 in agricultural landscapes with arable land and different semi-natural habitats. Applied Soil Ecology 52 , 20- 28. Schoch, C.L., Seifert, K.A., Huhndorf, S., Robert, V., Spouge, J.L., Levesque, C.A., Chen, W., Consortium, F.B., 2012. Nuclear ribosomal internal transcribed spacer (ITS) region as a universal DNA barcode marker for fungi. Proceedings of the National Academy of Sciences 109 , 6241-6246. 144

REFERENCES

Schoebel, C.N., Jung, E., Prospero, S., 2013. Development of new polymorphic microsatellite markers for three closely related plant-pathogenic Phytophthora species using 454-pyrosequencing and their potential applications. Phytopathology 103 , 1020-1027. Schrank, A., Vainstein, M.H., 2010. Metarhizium anisopliae enzymes and toxins. Toxicon 56 , 1267-1274. Schumann, M., Patel, A., Vemmer, M., Vidal, S., 2014. The role of carbon dioxide as an orientation cue for western corn rootworm larvae within the maize root system: implications for an attract-and-kill approach. Pest Manag Sci 70 , 642-650. Schüβler, A., Schwarzott, D., Walker, C., 2001. A new fungal phylum, the Glomeromycota: Phylogeny and evolution. Mycological Research 105 , 1413-1421. Schwarzenbach, K., Enkerli, J., Widmer, F., 2009. Effects of biological and chemical insect control agents on fungal community structures in soil microcosms. Applied Soil Ecology 42 , 54-62. Sealy, R., Evans, M.R., Rothrock, C., 2007. The effect of a garlic extract and root substrate on soilborne fungal pathogens. HortTechnology 17 , 169-173. Shade, A., Caporaso, J.G., Handelsman, J., Knight, R., Fierer, N., 2013. A meta-analysis of changes in bacterial and archaeal communities with time. ISME Journal 7 , 1493-1506. Shade, A., Peter, H., Allison, S.D., Baho, D.L., Berga, M., Bürgmann, H., Huber, D.H., Langenheder, S., Lennon, J.T., Martiny, J.B.H., Matulich, K.L., Schmidt, T.M., Handelsman, J., 2012. Fundamentals of microbial community resistance and resilience. Frontiers in Microbiology 3 , 417. Shah, P.A., Pell, J.K., 2003. Entomopathogenic fungi as biological control agents. Applied Microbiology and Biotechnology 61 , 413-423. Shang, Y., Xiao, G., Zheng, P., Cen, K., Zhan, S., Wang, C., 2016. Divergent and Convergent Evolution of Fungal Pathogenicity. Genome Biology and Evolution 8 , 1374-1387. Shepard, R.N., 1962. The analysis of proximities: Multidimensional scaling with an unknown distance function. I. Psychometrika 27 , 125-140. Shokralla, S., Spall, J.L., Gibson, J.F., Hajibabaei, M., 2012. Next-generation sequencing technologies for environmental DNA research. Molecular Ecology 21 , 1794-1805. Simon-Delso, N., Amaral-Rogers, V., Belzunces, L.P., Bonmatin, J.M., Chagnon, M., Downs, C., Furlan, L., Gibbons, D.W., Giorio, C., Girolami, V., Goulson, D., Kreutzweiser, D.P., Krupke, C.H., Liess, M., Long, E., McField, M., Mineau, P., Mitchell, E.A.D., Morrissey, C.A., Noome, D.A., Pisa, L., Settele, J., Stark, J.D., Tapparo, A., Van Dyck, H., Van Praagh, J., Van der Sluijs, J.P., Whitehorn, P.R., Wiemers, M., 2015. Systemic insecticides (neonicotinoids and fipronil): trends, uses, mode of action and metabolites. Environmental Science and Pollution Research International 22 , 5-34. Simpson, E.H., 1949. Measurment of diversity. Nature 163 , 688. Singh, B.K., Munro, S., Reid, E., Ord, B., Potts, J.M., Paterson, E., Millard, P., 2006. Investigating microbial community structure in soils by physiological, biochemical and molecular fingerprinting methods. European Journal of Soil Science 57 , 72-82. Singh, J., Singh, D.K., 2005. Bacterial, Azotobacter , actinomycetes, and fungal population in soil after diazinon, imidacloprid, and lindane treatments in groundnut (Arachis hypogaea L.) fields. Journal of Environmental Science and Health, Part B: Pesticides, Food Contaminants and Agricultural Wastes 40 , 785-800. Spatafora, J.W., Sung, G.H., Sung, J.M., Hywel-Jones, N.L., White, J.F., 2007. Phylogenetic evidence for an animal pathogen origin of ergot and the grass endophytes. Molecular Ecology 16 , 1701-1711. Spencer, J.L., Raghu, S., 2009. Refuge or reservoir? The potential impacts of the biofuel crop Miscanthus x giganteus on a major pest of maize. PloS One 4 , e8336. St Leger, R.J., Wang, C.S., 2010. Genetic engineering of fungal biocontrol agents to achieve greater efficacy against insect pests. Applied Microbiology and Biotechnology 85 , 901-907. St. Leger, R.J., 2008. Studies on adaptations of Metarhizium anisopliae to life in the soil. Journal of Invertebrate Pathology 98 , 271-276. St. Leger, R.J., Joshi, L., Bidochka, M.J., Roberts, D.W., 1996. Construction of an improved mycoinsecticide overexpressing a toxic protease. Proceedings of the National Academy of Sciences, USA 93 , 6349-6354. St. Leger, R.J., May, B., Allee, L.L., Frank, D.C., Staples, R.C., Roberts, D.W., 1992. Genetic differences in allozymes and in formation of infection structures among isolates of the entomopathogenic fungus Metarhizium anisopliae . Journal of Invertebrate Pathology 60 , 89-101. Steinwender, B.M., Enkerli, J., Widmer, F., Eilenberg, J., Kristensen, H.L., Bidochka, M.J., Meyling, N.V., 2015. Root isolations of Metarhizium spp. from crops reflect diversity in the soil and indicate no plant specificity. Journal of Invertebrate Pathology 132 , 142-148. Steinwender, B.M., Enkerli, J., Widmer, F., Eilenberg, J., Thorup-Kristensen, K., Meyling, N.V., 2014. Molecular diversity of the entomopathogenic fungal Metarhizium community within an agroecosystem. Journal of Invertebrate Pathology 123 , 6-12. 145

REFERENCES

Šterna, V., Zute, S., Jansone, I., Kantane, I., 2017. Chemical composition of covered and naked spring barley varieties and their potential for food production. Polish Journal of Food and Nutrition Sciences 67 , 151. Strasser, H., Forer, A., Schinner, F., 1996. Development of media for the selective isolation and maintenance of virulence of Beauveria brongniartii . In: Jackson, T.A., Glare, T.R., Eds.), Microbial control of soil dwelling pests. AgResearch, Lincoln, New Zealand, pp. 125-130. Sufyan, M., Neuhoff, D., Furlan, L., 2013. Effect of male mass trapping of Agriotes species on wireworm abundance and potato tuber damage. Bulletin of Insectology 66 , 135-142. Sufyan, M., Neuhoff, D., Furlan, L., 2014. Larval development of Agriotes obscurus under laboratory and semi- natural conditions. Bulletin of Insectology 67 , 227-235. Sung, G.H., Hywel-Jones, N.L., Sung, J.M., Luangsa-Ard, J.J., Shrestha, B., Spatafora, J.W., 2007. Phylogenetic classification of Cordyceps and the clavicipitaceous fungi. Studies in Mycology , 5-59. Tamura, K., Stecher, G., Peterson, D., Filipski, A., Kumar, S., 2013. MEGA6: Molecular Evolutionary Genetics Analysis version 6.0. Molecular Biology and Evolution 30 , 2725-2729. Taylor, J.W., 2011. One Fungus = One Name: DNA and fungal nomenclature twenty years after PCR. IMA Fungus : The Global Mycological Journal 2 , 113-120. Taylor, J.W., Fisher, M., 2003. Fungal multilocus sequence typing - it's not just for bacteria. Current Opinion in Microbiology 6 , 351-356. Taylor, J.W., Jacobson, D.J., Kroken, S., Kasuga, T., Geiser, D.M., Hibbett, D.S., Fisher, M.C., 2000. Phylogenetic Species Recognition and Species Concepts in Fungi. Fungal Genetics and Biology 31 , 21-32. Tedersoo, L., Bahram, M., Põlme, S., Kõljalg, U., Yorou, N.S., Wijesundera, R., Ruiz, L.V., Vasco-Palacios, A.M., Thu, P.Q., Suija, A., Smith, M.E., Sharp, C., Saluveer, E., Saitta, A., Rosas, M., Riit, T., Ratkowsky, D., Pritsch, K., Põldmaa, K., Piepenbring, M., Phosri, C., Peterson, M., Parts, K., Pärtel, K., Otsing, E., Nouhra, E., Njouonkou, A.L., Nilsson, R.H., Morgado, L.N., Mayor, J., May, T.W., Majuakim, L., Lodge, D.J., Lee, S.S., Larsson, K.-H., Kohout, P., Hosaka, K., Hiiesalu, I., Henkel, T.W., Harend, H., Guo, L.- d., Greslebin, A., Grelet, G., Geml, J., Gates, G., Dunstan, W., Dunk, C., Drenkhan, R., Dearnaley, J., De Kesel, A., Dang, T., Chen, X., Buegger, F., Brearley, F.Q., Bonito, G., Anslan, S., Abell, S., Abarenkov, K., 2014. Global diversity and geography of soil fungi. Science 346. Tiago, P.V., Carneiro-Leao, M.P., Lima, M.L.A., Oliveira, N.T., Luna-Alves Lima, E.A., 2011. Polymorphism in Metarhizium anisopliae var. anisopliae (Hypocreales: Clavicipitaceae) based on internal transcribed spacer-RFLP, ISSR and intron markers. Genetics and Molecular Research 10 , 1565-1575. Toepfer, S., Hatala-Zseller, I., Ehlers, R.-U., Peters, A., Kuhlmann, U., 2010. The effect of application techniques on field-scale efficacy: can the use of entomopathogenic nematodes reduce damage by Western corn rootworm larvae? Agricultural and Forest Entomology 12 , 389-402. Toepfer, S., Haye, T., Erlandson, M., Goettel, M., Lundgren, J.G., Kleespies, R.G., Weber, D.C., Walsh, G.C., Peters, A., Ehlers, R.U., Strasser, H., Moore, D., Keller, S., Vidal, S., Kuhlmann, U., 2009. A review of the natural enemies of beetles in the subtribe Diabroticina (Coleoptera: Chrysomelidae): Implications for sustainable pest management. Biocontrol Science and Technology 19 , 1-65. Toepfer, S., Kuhlmann, U., 2004. IPM strategies for the Western Corn Rootworm (Coleoptera: Chrysomelidae): current status and potential adoption to the European maize production systems. Mitteilungen der Deutschen Gesellschaft für Allgemeine und Angewandte Entomologie 14 , 391-398. Toepfer, S., Kuhlmann, U., 2006. Constructing life-tables for the invasive maize pest Diabrotica virgifera virgifera (Col.; Chrysomelidae) in Europe. Journal of Applied Entomology 130 , 193-205. Torsvik, V., Goksoyr, J., Daae, F.L., 1990. High diversity in DNA of soil bacteria. Applied and Environmental Microbiology 56 , 782-787. Tóth, M., Furlan, L., 2005. Pheromone composition of European click beetle pests (Coleoptera, Elateridae): common components-selective lures. IOBC/WPRS Bull 28 , 133-142. Tóth, M., Furlan, L., Yatsynin, V.G., Ujváry, I., Szarukán, I., Imrei, Z., Tolasch, T., Francke, W., Jossi, W., 2003. Identification of pheromones and optimization of bait composition for click beetle pests (Coleoptera: Elateridae) in Central and Western Europe. Pest Management Science 59 , 417-425. Trabelsi, D., Mengoni, A., Ben Ammar, H., Mhamdi, R., 2011. Effect of on-field inoculation of Phaseolus vulgaris with rhizobia on soil bacterial communities. FEMS Microbiology Ecology 77 , 211-222. Trabelsi, D., Mhamdi, R., 2013. Microbial inoculants and their impact on soil microbial communities: A review. BioMed Research International 2013. Traugott, M., Benefer, C.M., Blackshaw, R.P., van Herk, W.G., Vernon, R.S., 2015. Biology, ecology, and control of elaterid beetles in agricultural land. Annual Review of Entomology 60 , 313-334. Treseder, K.K., Lennon, J.T., 2015. Fungal traits that drive ecosystem dynamics on land. Microbiology and Molecular Biology Reviews 79 , 243-262. Trevors, J.T., 2009. One gram of soil: a microbial biochemical gene library. Antonie van Leeuwenhoek Journal of Microbiology 97 , 99. 146

REFERENCES van Dillewijn, P., Villadas, P.J., Toro, N., 2002. Effect of a Sinorhizobium meliloti strain with a modified puta gene on the rhizosphere microbial community of alfalfa. Applied and Environmental Microbiology 68 , 4201-4208. van Herk, W.G., Vernon, R.S., Moffat, C., Harding, C., 2008a. Response of the pacific coast wireworm, Limonius canus , and the dusky wireworm, Agriotes obscurus (Coleopotera: Elateridae), to insecticide-treated wheat seeds in a soil bioassay. Phytoprotection 89 , 7-19. van Herk, W.G., Vernon, R.S., Roitberg, B.D., 2008b. Repellency of a wireworm, Agriotes obscurus (Coleoptera: Elateridae), on exposure to synthetic insecticides in a soil-less bioassay. Environmental Entomology 37 , 534-545. van Herk, W.G., Vernon, R.S., Tolman, J.H., Saavedra, H.O., 2008c. Mortality of a wireworm, Agriotes obscurus (Coleoptera: Elateridae), after topical application of various insecticides. Journal of Economic Entomology 101 , 375-383. van Rozen, K., Ester, A., 2010. Chemical control of Diabrotica virgifera virgifera LeConte. Journal of Applied Entomology 134 , 376-384. Veen, K.H., 1968. Recherches sur la maladie, due al Metarrhizium anisopliae chez la criquet pelerin. Mededelingen Landbouwhogeschool, Wageningen, pp. 77. Vega, F.E., Goettel, M., Blackwell, M., Chandler, D., Jackson, M.A., Keller, S., Koike, M., Maniania, N.K., Monzon, A., Ownley, B.H., Pell, J., Rangel, D.E.N., Roy, H.E., 2009. Fungal entomopathogens: new insights on their ecology. Fungal Ecology 2 , 149-159. Velásquez, V.B., Carcamo, M.P., Merino, C.R., Iglesias, A.F., Duran, J.F., 2007. Intraspecific differentiation of Chilean isolates of the entomopathogenic fungi Metarhizium anisopliae var. anisopliae as revealed by RAPD, SSR and ITS markers. Genetics and Molecular Biology 30 , 89-99. Vernon, B., Lagasa, E., Philip, H., 2001. Geographic and temporal distribution of Agriotes obscurus and A. lineatus (Coleoptera: Elateridae) in British Columbia and Washington as determined by pheromone trap surveys. Journal of the Entomological Society of British Columbia 98 , 257-265. Vernon, R.S., Van Herk, W., Tolman, J., Ortiz Saavedra, H., Clodius, M., Gage, B., 2008. Transitional sublethal and lethal effects of insecticides after dermal exposures to five economic species of wireworms (Coleoptera: Elateridae). Journal of Economic Entomology 101 , 365-374. Vernon, R.S., Van Herk, W.G., 2012. Wireworms as pests of potato. In: Alyokhin, A., Giordanengo, P., Vincent, C., Eds.), Insect pests of potato: Global perspectives on biology and management. Academic Press, Amsterdam, pp. 103-164. Vernon, R.S., Van herk, W.G., Clodius, M., Harding, C., 2009. Wireworm management I: Stand protection versus wireworm mortality with wheat seed treatments. Journal of Economic Entomology 102 , 2126-2136. Vernon, R.S., van Herk, W.G., Clodius, M., Harding, C., 2013. Crop protection and mortality of Agriotes obscurus wireworms with blended insecticidal wheat seed treatments. Journal of Pest Science 86 , 137-150. Vestergaard, S., Cherry, A., Keller, S., Goettel, M., 2003. Safety of hyphomycete fungi as microbial control agents. In: Hokkanen, H.M.T., Hajek, A.E., Eds.), Environmental impacts of microbial insecticides: Need and methods for risk assessment. Springer Netherlands, Dordrecht, pp. 35-62. Vieira, F.C.S., Nahas, E., 2005. Comparison of microbial numbers in soils by using various culture media and temperatures. Microbiological Research 160 , 197-202. Walstad, J.D., Anderson, R.F., Stambaugh, W.J., 1970. Effects of environmental conditions on two species of muscardine fungi ( Beauveria bassiana and Metarrhizium anisopliae ). Journal of Invertebrate Pathology 16 , 221-226. Wang, B., Kang, Q., Lu, Y., Bai, L., Wang, C., 2012. Unveiling the biosynthetic puzzle of destruxins in Metarhizium species. PNAS 109 , 1287-1292. Wang, C., Feng, M.G., 2014. Advances in fundamental and applied studies in China of fungal biocontrol agents for use against arthropod pests. Biological Control 68 , 129-134. Wang, C.S., Hu, G., St Leger, R.J., 2005. Differential gene expression by Metarhizium anisopliae growing in root exudate and host ( Manduca sexta ) cuticle or hemolymph reveals mechanisms of physiological adaptation. Fungal Genetics and Biology 42 , 704-718. Wang, C.S., St Leger, R.J., 2007. The MAD1 adhesin of Metarhizium anisopliae links adhesion with blastospore production and virulence to insects, and the MAD2 adhesin enables attachment to plants. Eukaryotic Cell 6, 808-816. Wang, C.S., Typas, M.A., Butt, T.M., 2002. Detection and characterisation of pr1 virulent gene deficiencies in the insect pathogenic fungus Metarhizium anisopliae. FEMS Microbiology Letters 213 , 251-255. Wellington, E.M.H., Berry, A., Krsek, M., 2003. Resolving functional diversity in relation to microbial community structure in soil: exploiting genomics and stable isotope probing. Current Opinion in Microbiology 6 , 295-301.

147

REFERENCES

Wesseler, J., Fall, E.H., 2010. Potential damage costs of Diabrotica virgifera virgifera infestation in Europe – the ‘no control’ scenario. Journal of Applied Entomology 134 , 385-394. Wetterstrand, M.S., 2016. DNA Sequencing costs: Data from the NHGRI Genome Sequencing Program (GSP) Whipps, J.M., 2001. Microbial interactions and biocontrol in the rhizosphere. Journal of Experimental Botany 52 , 487-511. Wieland, G., Neumann, R., Backhaus, H., 2001. Variation of microbial communities in soil, rhizosphere, and rhizoplane in response to crop species, soil type, and crop development. Applied and Environmental Microbiology 67 , 5849-5854. Wiley, E.O., 1978. The evolutionary species concept reconsidered. Systematic Zoology 27 , 17-26. Woese, C.R., 1987. Bacterial Evolution. Microbiological Reviews 51 , 221-271. Wraight, S.P., Ramos, M.E., 2015. Delayed efficacy of Beauveria bassiana foliar spray applications against Colorado potato beetle: Impacts of number and timing of applications on larval and next-generation adult populations. Biological Control 83 , 51-67. Wyrebek, M., Huber, C., Sasan, R.K., Bidochka, M.J., 2011. Three sympatrically occurring species of Metarhizium show plant rhizosphere specificity. Microbiology 157 , 2904-2911. Zaller, J.G., König, N., Tiefenbacher, A., Muraoka, Y., Querner, P., Ratzenböck, A., Bonkowski, M., Koller, R., 2016. Pesticide seed dressings can affect the activity of various soil organisms and reduce decomposition of plant material. BMC Ecology 16 , 37. Zaluski, R., Kadri, S.M., Alonso, D.P., Martins Ribolla, P.E., de Oliveira Orsi, R., 2015. Fipronil promotes motor and behavioral changes in honey bees ( Apis mellifera ) and affects the development of colonies exposed to sublethal doses. Environmental Toxicology Chemistry 34 , 1062-1069. Zhao, H., Wang, Z.K., Yin, Y.P., Li, Y.L., Li, Z.L., Peng, G.X., Xia, Y.X., 2007. Trehalose and trehalose- hydrolyzing enzyme in the haemolymph of Locusta migratoria infected with Metarhizium anisopliae strain CQMa102. Insect Science 14 , 277-282. Zimmermann, G., 1986. The Galleria bait method for detection of entomopathogenic fungi in soil. Journal of Applied Entomology 102 , 213-215. Zimmermann, G., 2007. Review on safety of the entomopathogenic fungus Metarhizium anisopliae . Biocontrol Science and Technology 17 , 879-920. Zimmermann, J., Musyoki, M.K., Cadisch, G., Rasche, F., 2016. Proliferation of the biocontrol agent Fusarium oxysporum f. sp. strigae and its impact on indigenous rhizosphere fungal communities in maize under different agro-ecologies. Rhizosphere 1 , 17-25.

148

ACKNOWLEDGEMENTS

Acknowledgements

I would like to address a great “thank you” to my main supervisor Jürg Enkerli for his support and well-planned frame of the project, for making it possible that I could attend several conferences and for teaching me an efficient, precise and fore-seeing way of working as a scientist, while not forgetting breaks and leisure time to re-energize. In the same manner I would like to thank Prof. Adrian Leuchtmann for the constructive inputs during the project meetings where I could benefit from his enormous knowledge about fungi and biology in general. I would like to express my gratitude to Prof. Bruce McDonald for his interest into my PhD thesis and his very valuable comments during the PhD committee meetings. All committee members including Prof. Stefan Vidal I would like to thank for the interesting questions and discussions during the PhD defense.

I would like to thank the head of the Molecular ecology group, Dr. Franco Widmer, for his critical questions which are so important for scientific progress, his creative ideas to assess and design control experiments and his supportive way of leading our group. I would like to address my gratitude to Dr. Martin Hartmann who introduced and explained the computational aspect of the PhD thesis which was a very important part for most results to me and thereby helped me to develop from a bioinformatic-newby to an interested user.

I would like to thank my colleagues of the group Molecular Ecology, shortly called “Mökis”. In the earlier days there were the PhD students Lena Hersemann, Verena Knorst, Katharina Kempf followed by Martina Birrer, Florian Gschwend and Miguel Angel Loera Sanchez, the Postdocs Dr. Tina Wunderlin and Dr. Aaron Fox and the technicians Stephanie Pfister and Sonja Reinhard, Urs Büchler and many “Zivis” and “Azubis”. All of them I would like to thank for the help in the lab, their input in assessing results, the wide variety of cakes in the coffee brakes, the ski-weekends and hiking tours, jointly visited conferences, cheer-ups in difficult moments and shared jokes.

This work would not have succeeded without three research groups with whom we performed the two main experiments of the thesis: The researchers at the University of Innsbruck and Laimburg including Dr. Hermann Strasser, Dr. Roland Zelger, Dr. Hannes Rauch, Christopher Oliver Pabst and Maria Zottele which were involved in a pot and field trial conducted in Laimburg and Styria. In the same way I would like to thank Giselher Grabenweger, Sonja Eckard, Christian Schweitzer and Christian Gees for our joint pot and field experiments at Agroscope and Tägerig. This work was part of the EU project “Innovative Biological products for Soil pest control” (INBIOSOIL) and therefore I am very grateful to all project members.

A great hug and thank you to all my friends here in Zurich and back home in Austria and to my parents and sisters who gave me freedom to decide what I would like to do and supported me anyway, who always had an open ear and good advice from an outside point and who helped organizing an unforgettable PhD defense party.

149