JUSTIN Y. CHAN

Towards a mechanistic understanding of

fungal life history strategies:

Dispersal and competition as critical components of saprotrophic fungal ecology

A THESIS IN FULFILMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

SCHOOL OF BIOLOGICAL, EARTH & ENVIRONMENTAL SCIENCES FACULTY OF SCIENCE FEB 2021

1

Abstract

Saprotrophic fungi perform the vital role of cycling nutrients and carbon back into the environment through the decomposition of organic matter. But in a rapidly changing global environment, we do not fully understand how these environmental changes will affect the process of decomposition by saprotrophic fungi. To understand decomposition and carbon cycling at a global scale, we must first begin to identify the life history strategies that saprotrophic fungi employ. For saprotrophic fungi, resources are arrayed in patches in the environment, much like an archipelago of islands. On these resource islands, interspecific competition is intense and available resources are continually depleted as a consequence of fungal metabolism. Here, airborne dispersal is a key factor that allows fungi to avoid being restricted to an island experiencing total resource patch collapse and to persist within the environment. In this thesis, I take an experimental approach and explore how living in finite resource patches shapes allocation to dispersal in fungi, and how a resource patch can alter the course of interspecific competition. In Chapters 2 and 3, I tracked allocation to dispersal in saprotrophic fungi using Phacidium lacerum, a novel model species. In Chapter 2, I ran a Petri-dish experiment varying resource island size (Petri-dish size) and nutrient concentration to test how environmental quality influenced allocation to dispersal. In Chapter 3, I included both interspecific and intraspecific competitors to see how negative interactions changed patterns of allocation to dispersal. In Chapters 4 and 5, I focused on competition between saprotrophic fungi. In Chapter 4, I explored how the strength of competition varied across simple and complex substrates, and in

Chapter 5, I tracked how wood decay was influenced by interspecific competition in a paired wood block experiment. Altogether, these studies highlight the importance of considering dispersal and competition in saprotrophic fungi, especially in the context of finite resource patches as islands.

2

Contents

Abstract 2 Acknowledgements 7 List of Figures 9 List of Tables 11 List of Published Works 12

Chapter 1: Introduction: Investigating the strategies for life of wood decay fungi

1.1 Fungi 14 1.2 The global importance of wood decomposition 15 1.3 Wood as both habitat and resource for fungi 17 1.4 Life history theory and fungi 18 1.5 Resource patches as islands and fungal meta-populations 19 1.6 Dispersal and fungal life history 20 1.7 Colonisation, competition, and fungal combat 22 1.8 Community assembly and interactions 24 1.9 Understanding competition and dispersal in saprotrophic fungi 26 1.10 Competition in saprotrophic fungi 28 1.11 An experimental approach: Thesis outline 29 1.12 References 32

Chapter 2: Environmental cues for dispersal in a filamentous fungus in simulated islands

2.1 Abstract 41

2.2 Introduction 42

2.3 Methods 46

2.3.1 Species isolation and identification 46 2.3.2 Experimental Design 46

3

2.3.3 Image Analysis 47 2.3.4 Statistical Analysis 48

2.4 Results 49

2.4.1 Colony growth rate and patch edge detection 49 2.4.2 Effect of resource level on pycnidia production 50 2.4.3 Patterns of allocation to growth and dispersal across different islands 54

2.5 Discussion 54

2.5.1 Cues for dispersal 54 2.5.2 Patterns in dispersal allocation between island sizes 55 2.5.3 A two-cue model for dispersal in fungi 57 2.5.4 Conclusion 60

2.6 References 62 2.7 Supplementary Material 67

Chapter 3: When to cut your losses: Dispersal allocation in an asexual filamentous fungus in response to competition

3.1 Abstract 73 3.2 Introduction 74

3.3 Materials and methods 76

3.3.1 Study Species 76 3.3.2 Isolating Competitors 77 3.3.3 Experimental Design 77 3.3.4 Image Analysis 78 3.3.5 Statistical Analysis 79

3.4 Results 80

3.4.1 Colony growth and contact with competitors 80 3.4.2 Allocation to dispersal as a response to competition 82 3.4.3 Trade-offs in allocation 84

4

3.5 Discussion 87

3.5.1 Trade-offs in allocation 87

3.6 Conclusion 90 3.7 References 91 3.8 Supplementary Material 96

Chapter 4: Complex environments alter competitive dynamics in fungi

4.1 Abstract 98

4.2 Introduction 99

4.3 Methods 102

4.3.1 Species selection 102 4.3.2 Microbial respiration assay 103 4.3.3 Experimental design 104 4.3.4 Establishment of simple competitive environments 104 4.3.5 Establishment of wood block interactions 105 4.3.6 Statistical analysis 106

4.4 Results 107

4.4.1 Respiration rate and growth rate 107 4.4.2 Patterns of competition between fungi on agar and wood 108 4.4.3 Competitive hierarchies on agar and wood 111

4.5 Discussion 115

4.5.1 Patterns of competition across simple and complex substrates 115 4.5.2 Metabolic rate and competitive ability 117 4.5.3 Conclusion 118

4.6 References 120 4.7 Supplementary Material 126

5

Chapter 5: The effect of fungal dynamics on early wood decomposition: insights from a laboratory experiment

5.1 Abstract 129

5.2 Introduction 130

5.3 Methods 133

5.3.1 Species selection 133 5.3.2 Ability to respire cellulose and lignin 134 5.3.3 Establishing wood block interactions 135 5.3.4 Statistical analysis 137

5.4 Results 138

5.4.1 Lag in early wood decay 138 5.4.2 The effect of competition on mass change in wood blocks during early wood decay 138 5.4.3 Biomass import in early wood decay 139

5.5 Discussion 142

5.5.1 Lag in early wood decay 142 5.5.2 The effect of fungal competition on wood decay 143 5.5.3 Priority effects and implications on early wood decay 144 5.5.4 Conclusion 145

5.6 References 147

5.7 Supplementary Material 152

Chapter 6: General Conclusions

6.1 Fungal life history in the context of declining resource patches 160 6.2 The need to disperse 161 6.3 Fungus eat fungus world 163 6.4 An archipelago of ephemeral resource islands? 166 6.5 A need for theory 168 6.6 References 170

6

Acknowledgements

This work has been the culmination of my academic journey thus far. I remember looking up to all the postgraduate students ahead of me, inspired by their experiments and their presentations.

When I had finally started my honours year, I knew I would take the path of the PhD. The PhD has been the most intellectually rigorous and at times, was a vertical learning curve. During this process,

I remember trying so hard to be what I thought the model fungal ecology researcher. The first year was challenging – learning how to make agar, use an autoclave (I am still terrified of that machine), run PCRs, and a whole gamut of microbiological lab skills. Later, I would relax from that personal mission, and I learnt to enjoy the PhD as a process. This journey has taught me many things, of which

I would learn that giving presentations weren’t so scary, and that it would become one of my favourite things to do. Suffice to say, my PhD has been the biggest learning opportunity I have undertaken, and in this endeavour, I owe an enormous debt of gratitude to all the people that have made it possible.

My primary supervisor, Will, has been a pillar for my PhD journey. Forever patient, his mentorship and guidance has been a driving force for me even when experiments go awry. I am eternally grateful for his wisdom and advice throughout this entire process, and I cannot thank him enough.

Stephen, whom I’ve worked with since my undergraduate days, has been a constant presence in my academic journey. Always ready for a chinwag, he has always given me time when I knock on his door unannounced. His mentorship will always be appreciated. Jeff - my last supervisor.

My first year was made possible because of your generosity. And Jess, thank you for bearing with me when I asked basic questions about microbiological lab technique. You are a superstar. To Lynne and the Boddy Lab, thank you for allowing me to visit your lab. Thank you, Melanie, for providing me the guides, the how-to’s, and instructions for the experimental set-ups, this process would’ve taken me far longer if it weren’t for your generosity.

7

Thank you to the Australian Research Council (DP160103765), whose granted funded the studies in this thesis. And I acknowledge the support by the government through the Australian

Government Research Training Program Scholarship.

Thanks to Jono, for saving my hide more times than I could count. To Hayley, thank you for all the opportunities you have given me. To MK! Thank you for collaborating with me! To Charlotte, my Bonser lab companion since before we both took this journey. Steph, for inspiring me and pushing me forward. Gina, for being the best cheerleader I could ever ask for. Maya, for all the times

I’d roped you into being a research assistant with promise of dumplings, and for being the baddest B since 2008.

To my parents, thank you for all the sacrifices you have made to make this possible. There are not enough “thank-yous” in my lungs that can approach the gratitude necessary for all you have done for me. To Jerold, you’re alright.

And finally, to a certain someone. You have been my guiding light. Thank you for all the words of encouragement and loving support. May we walk this path together.

JYC

8

List of Figures

Chapter 2 Figure 1 51 Relative growth rate and pycnidia produced (per 12 hours) between treatments over time

Figure 2 53 Comparison of pycnidia production per 12 hours for P. lacerum against relative growth rate

Figure 3 59 A conceptual model of the two-cue switch in resource allocation to dispersal

Chapter 3 Figure 1 81 Spline ± SE of pycnidia density in mm2 over 7 days.

Figure 2 83 Pycnidial density per mm2 among treatments at 180 hours.

Figure 3 85 Comparison of relationship between pycnidia development per 12 hours for P. lacerum against relative growth rate.

Figure 4 86 Phacidium lacerum interactions at 180 hours

9

Chapter 4 Figure 1 113 2 Comparison of total average growth rate ±SE (mm per hour) before and after colony meeting.

Figure 2 114 Proportion of territory (% occupied) captured for the Helotiales isolate (He), Omphalotus sp. (Om), Peniophora sp. (Pe), and Phacidium lacerum (Pl) during interactions on agar (right) and in wood (left) over time.

Chapter 5 Figure 1 136 Experimental set up of wood block interactions and Wood block interaction

Figure 2 140 Corrected mass change in wood (against control).

Figure 3 141 Estimated biomass at time point 0 relative to control blocks.

10

List of Tables

Chapter 2 Table 1 52 Tukey’s HSD test comparisons of treatment effects on pycnidial density

Chapter 3 Table 1 84 Longitudinal two-level mixed model with repeated measures of individual Petri-dishes to test for differences in the relationship between dispersal allocation and RGR.

Chapter 4

Table 1 109 Mixed model with time and competitive environment as fixed factors, with proportion of area occupied as the response variable.

Table 2 112 Competitive hierarchy established between pairwise interactions on Wood and Agar.

Chapter 5

Table 1 141 Tukey’s HSD post-hoc test comparing estimated biomass at time point 0 relative to control blocks.

11

List of Published works

Chapter 2

Environmental cues for dispersal in a filamentous fungus in simulated islands

Chan, J. Y., Bonser, S. P., Powell, J. R., & Cornwell, W. K. (2020). Environmental cues for dispersal in a filamentous fungus in simulated islands. Oikos, 129(7), 1084-1092.

JC and WC conceived the ideas and designed methodology; WC, SB, and JP supervised the work of JC;

JC collected data; JC and WC analysed data; JP assisted with statistical analysis; JC lead the writing of the manuscript; All authors contributed critically to the drafts and gave final approval for publication.

Chapter 3

When to cut your losses: Dispersal allocation in an asexual filamentous fungus in response to competition

Chan, J. Y., Bonser, S. P., Powell, J. R., & Cornwell, W. K. (2019). When to cut your losses: Dispersal allocation in an asexual filamentous fungus in response to competition. Ecology and evolution, 9(7),

4129-4137.

JC, SB, and WC conceived the ideas and designed methodology. WC, SB, and JP supervised the work of JC. JC collected data. JC and WC analysed data. JC led the writing of the manuscript. All authors contributed critically to the drafts and gave final approval for publication.

12

CHAPTER 1

Introduction: Investigating the strategies for life

of wood decay fungi

Justin Y. Chan

13

1.1 Fungi

The kingdom Fungi is a diverse clade of eukaryotes separated from other organisms by the presence of chitin in their cell walls. Fungi are vital to the functioning of many ecosystems, particularly as carbon and nutrient cyclers (Schimel 1995). Many fungi are of commercial importance in the production of medicine, food, and other resources (Bennett 1998, Alastruey-Izquierdo et al.

2015) and are also significant as plant pathogens, causing major losses in global agriculture (Boyd et al. 2013). Globally, Larsen et al. 2017 estimated that there are 2.4 million soil fungi, with as many as a combined 165.6 million species when considering all parasitic fungi (Larsen et al. 2017), with the vast majority of these being cryptic species. Despite the ecological and commercial significance of fungi, the threats towards fungal species at a global level remain poorly understood, with many red- listed fungi worldwide as a consequence of continued habitat loss and degradation of forested ecosystems (Dahlberg et al. 2010).

Traditionally, fungi are difficult to quantify due to the amorphous and non-unitary nature of fungi as organisms. The lack of precise species definitions for fungi have hampered progress for taxonomic reform (Hibbett et al. 2007). Many mycelial saprotrophic fungi lack clearly defined body- forms (barring fruiting bodies), with clonal populations or colonies (Todd and Rayner 1980, Walker and White 2017). This lack of clear distinction between fungi in the environment has often contributed much to the “black box” approach to fungal ecology, where assumptions about the ecological functioning of fungi are made (Fernandez and Kennedy 2015).

With a global distribution, fungi can be found in all terrestrial ecosystems (Peay et al. 2016), with some fungi specialising in aquatic and marine ecosystems (Grossart et al. 2019). To face the myriad challenges of the natural world, fungi have evolved and adapted to various ecological niches.

Fungi have developed relationships with both animals and plants, ranging from parasitic to symbiotic

14

(Naranjo-Ortiz and Gabaldón 2019). One of the most well-known mutualistic relationships between fungi and plants is the mycorrhizal association between fungi and plant roots, where the fungus is able to exchange mineral nutrients for photosynthate from the host plant (Brundrett 1991). Some lineages of fungi have developed intimate symbioses with algae, coevolving together and forming lichen (Honegger 2009). Another group of fungi turned to saprotrophy - or the decay of organic matter.

Saprotrophic fungi are heterotrophs and most members of either the Basidiomycete and

Ascomycete clades, with a few members from the Zygomycete and Mucoromycete fungal clades

(Cornwell et al. 2009, Eichlerová et al. 2015). The distribution of saprotrophic fungi throughout terrestrial ecosystems is heavily influenced by how they derive their nutrition (Cooke et al. 1984).

Unlike plants and other autotrophs, saprotrophic fungi are unable to photosynthesise and as a consequence, saprotrophic fungi must derive their nutrition from the organic carbon found in the environment. This reliance on organic carbon restricts the pattern of distribution for fungi to habitats where they are readily available (Cooke et al. 1984), and imposes unique challenges to their ecology dissimilar to the challenges faced by plants, with the constant threat of resource pools drawing down to non-existence.

The ecological needs of fungi drastically change depending on their lifestyle (Peay et al.

2016, Naranjo-Ortiz and Gabaldón 2019). In this thesis, I focus on the decomposers, who perform the critical ecosystem function of cycling nutrients and carbon back into the environment. I explore how a life of saprotrophy shapes the ecological needs of saprotrophic fungi, and perhaps fundamentally, I aim to answer a more essential question: What do saprotrophic fungi do?

15

1.2 The global importance of wood decomposition

Woodland vegetation communities constitute a large pool of organic carbon within the global carbon cycle (Swift 1977, Cornwell et al. 2009). Forests contain ~360 PgC within plant biomass

(Dixon et al. 1994), with 36-72 PgC locked within woody debris (Brown 2002, Goodale et al. 2002,

Cornwell et al. 2009). When trees and shrubs die, the organic carbon enters the pool of plant detritus such as leaf litter or woody debris, the latter of which can have residence times in the environment that can span many years or decades (Pietsch et al. 2014, Oberle et al. 2018). In a changing climate, there is a great deal of uncertainty regarding how temperature will affect this pool of carbon (Gower 2003, Bardgett et al. 2008). While abiotic factors and plant traits have often been used to predict carbon dioxide evolution, the activity of microbial decomposers such as fungi have a huge impact on how carbon dioxide is released from woody debris (Hiscox et al. 2015, 2016).

Understanding how this pool of carbon moves within the greater carbon cycle is important to have accurate predictions and models for projections of future climate (Cramer et al. 2001, Cornwell et al.

2009, McGuire and Treseder 2010).

One of the reasons why wood has such a long residence time is due to the complex of molecules collectively known as the lignocellulose complex. The lignocellulose complex is made up of lignin, cellulose, and hemicellulose, and together forms the physical structure of wood (Blanchette

1995). Lignin is a complex phenolic compound, and its resistant nature is due to the aromatic rings that form its molecular structure (Ruiz-Dueñas and Martínez 2009). Unlike lignin, cellulose and hemicellulose are polysaccharide chains that are vulnerable to degradation by common enzymes such as hydrolases and gluccanases (Eriksson 1978, Béguin and Aubert 1994, Bhat and Bhat 1997).

The carbohydrate rich cellulose and hemicellulose are locked by lignin (Béguin and Aubert 1994), and access to this resource is limited to organisms able to degrade lignin, thus releasing the cellulose and hemicellulose, or organisms with the enzymatic capacity to modify lignin, allowing decay of the

16 cellulose and hemicellulose leaving the lignin largely intact (Lopez et al. 2007, Sánchez 2009).

However, the ability to degrade lignin is comparatively not common (Cragg et al. 2015, Purahong et al. 2016, Sista Kameshwar and Qin 2018). Even within the kingdom Fungi, the ability to degrade lignin is restricted to certain clades (Nagy et al. 2016).

The evolution of carbon dioxide from woody debris can follow many pathways, involving consumption by invertebrates, combustion, and physical degradation, but microbial decomposition

(chiefly by fungi) represents a major pathway of carbon dioxide evolution within the greater carbon cycle in terrestrial ecosystems (Cornwell et al. 2009). The rate by which carbon dioxide is released by fungal decomposition from plant detritus is largely controlled by the interactions between saprotrophic fungi (Maynard et al. 2017). Understanding how the life history strategies of fungi influence the interactions between saprotrophic fungi represents an important step towards elucidating this vital part of the global carbon cycle.

1.3 Wood as both habitat and resource for fungi

Woody debris as a pool of organic carbon is an important resource for saprotrophic fungi, whose metabolism and respiration frees the locked carbon within the plant detritus. As such, fungi are responsible for a major flux of carbon in terrestrial ecosystems (Cornwell et al. 2009,

Lustenhouwer et al. 2020). Additionally, the activity of saprotrophic fungi performs the vital task of recycling nutrients back into their environment (Harmon et al. 1990, Jönsson et al. 2008, Fukami et al. 2010), producing new humus from the mineralisation of organic matter (Cooke et al. 1984,

Crowther et al. 2012).

17

Wood decay fungi play an important role in decomposition because they can decompose the recalcitrant lignocellulose complex present in woody debris (Sánchez 2009, Hiscox et al. 2018).

Fungal lineages able to tap into this resource are broadly grouped into three distinct classes, separated by their ability to decay wood. These groups are separated between broader taxa and classified further by their enzymatic ability to break down wood components (Cooke et al. 1984).

The most common of the wood decay fungi are separated into two groups - the white rot fungi, and the brown rot fungi. White rot fungi are those that are able to completely decompose all components of the lignocellulose complex, while brown rot fungi degrade the cellulose and hemicellulose found in wood, leaving lignin behind during the decomposition process (Goodell et al.

2008). One notable group causes soft rot. Characterised by decay in the presence of high moisture

(Sista Kameshwar and Qin 2018), soft rot fungi are able to degrade wood without the help of peroxidase-based enzymes. While emerging research has identified the inadequacy of grouping wood decay fungi by decay mode (Riley et al. 2014), it remains a useful guide to understand the role that different saprotrophic fungi play in our environment.

1.4 Life history theory and fungi

The aim of life history theory is to try and understand how organisms achieve reproductive success in response to different ecological contexts, and how optimisation of reproductive success is shaped in turn by evolution over time (Stearns 2000). Much of our understanding of fungal ecology is built on the theoretical framework of plant and animal ecology (Stearns 2000, Gilchrist et al.

2006). But unlike either of the aforementioned groups, fungal communities are indeterminate and lack quantifiable units for individuals (Todd and Rayner 1980, Pringle and Taylor 2002). As a result, direct application of the same life history strategies may be limited in its ability to explain phenomena seen in microbes like fungi without directly analogous components within the life cycle

18 between fungi and higher eukaryotes like traditional animal and plant models. Fungi have, compared to plants and animals, fewer tissue types and lack complex anatomical organisation, and yet have a great deal of diversity in growth forms ranging from unicellular yeasts (Van Dyke et al. 2019) to expansive multicellular networks spanning upwards of 15 hectares (Smith et al. 1992). While life history theory can break down when extrapolated from one organism to another, especially where there are fundamental differences of trophic mode, elements of the theory may still be applicable.

When spatial dynamics are considered, marine sessile organisms like corals represent a solid analogue for fungi and other microbes, with similar life history with respect to life cycles and dispersal (McDougald et al. 2011).

Saprotrophic fungi are typically sessile and often appear as mycelia, or a network of filamentous hyphal cells (Walker and White 2017). Hereafter, the similarities to other higher organisms are limited. While the sessile nature of fungi can be compared to plants or sessile marine animals, the resources that fungi utilise are not renewed. For a fungus, access to resources allows for growth and ultimately reproduction, but use of these resources directly reduces the resources available for other biological processes (Gilchrist et al. 2006). The finite nature of the resources that saprotrophic fungi occupy poses the first of many challenges towards the completion of the fungal life cycle. Unlike autotrophs, where critical resources are eventually replenished (e.g. light and nutrients) within the environment, the completion of the fungal life cycle must be finalised prior to the depletion of the local resource patch.

Trade-offs in resource allocation within an organism underpin the strategies for maximising fitness in a given environment (Stearns 1992, 2000, Sæther and Engen 2015). With a finite resource pool, trade-offs occur as an organism is unable to simultaneously optimise two or more traits given limited resources (Stearns 1992). For a saprotrophic fungus, the organism must be able to allocate a finite pool of resources correctly to growth or dispersal. As such, the ecological needs of

19 saprotrophic fungi are best understood by viewing the fungal life cycle within the context of finite resources. Saprotrophic fungi need to juggle the need to grow, withstand competition, extract resources, and reproduce within an environment of diminishing resources. Where failure to successfully reproduce will result in the local extinction of the organism as the resources fall to levels below where reproduction is viable.

1.5 Resource patches as islands and fungal meta-populations

As decomposers, the metabolism of saprotrophic fungi necessarily draws down resources from the local environment (Boddy and Hiscox 2016). As such, dispersal is a fundamental ecological process and a critically important component of the ecology of saprotrophic fungi. Meta-population dynamics focuses on the interconnectivity of smaller populations of species within a greater population at a larger scale (Hanski 1998). Resource patches appear as spatially disparate islands.

Dispersal and patch dynamics connect fungal meta-populations within a greater population within the greater landscape, whose interactions can shape the progression of communities within patches or lead to localised extinction of a species (Hanksi 1998). Without a way to physically escape an environment with dwindling resources, dispersal is the only way for many fungi to access new resources (Lancaster and Downes 2017). As sessile organisms, direct movement from resource to resource is only achievable via hyphal extension, a process that can be energetically costly or impossible (Edman et al. 2004). While some species are able to produce expansive networks of long cords (Boddy 1999), the majority of wood-decay fungi solve the issue of distance with a distinct airborne dispersal phase (Edman et al. 2004, Norros et al. 2012).

Like other sessile organisms, fungi traverse from patch to patch within the environment via the production of dispersal propagules (Kinlan and Gaines 2003). This process maintains connectivity

20 between resource patches of different ages (Jonsson et al. 2005, Norros et al. 2012) and fungi can move distances far longer than hyphal extension can allow (Edman et al. 2004). From a fungus’s perspective, suitable resource patches (e.g. woody debris) are arrayed within a matrix of otherwise inhospitable terrain. These resource patches can be thought of as resource-rich islands for saprotrophic fungi against a backdrop of a resource-poor environment (Andrews et al. 1987). To access new islands and to escape islands with diminishing resources, fungi must disperse via airborne spores. During dispersal events, fungi can produce an enormous amount of spores that can travel great distances (Peay and Bruns 2014). While many of these spores perish, enough propagules will reach suitable sites for the fungus to locate and colonise new resources (Calhim et al. 2018).

Fungi can also disperse via the activity of insects, but this occurs incidentally via the symbiotic relationship between fungi and some saproxylic insects (Persson et al. 2011). In some ectomycorrhizal fungi, ornamentation on the surface of spores aids transport by soil organisms through the soil column, but ornamentation is rare amongst saprotrophic fungi (Calhim et al. 2018).

1.6 Dispersal and fungal life history

Saprotrophic fungi exist in a constantly changing environment and the ability to respond appropriately to changes in the environment is key to continued survival (Clobert et al. 2009). Like marine sessile invertebrates, reproduction and dispersal are linked, with dispersal the only way to escape unfavourable conditions (Kinlan and Gaines 2003). For fungi, the ability to detect local environmental conditions for competitors and resource level is necessary in preparation for dispersal. Changes in environmental quality have been observed to trigger dispersal events in microorganisms (McDougald et al. 2011), with a reduction in available nutrients being the primary driver of dispersal in microbes (Delaquis et al. 1989, Sawyer and Hermanowicz 2000, Schleheck et al.

2009). This is in line with informed dispersal theory (Clobert et al. 2009). Informed dispersal theory predicts that organisms can take information from their local environment to make decisions to

21 disperse if benefits of dispersal outweigh the risks of remaining stationary. Given that saprotrophic fungi have evolved within an environment of predictably declining resources, allocation to dispersal is expected in the face of worsening environmental conditions (Allison and Martiny 2008, Hibbing et al. 2010).

For saprotrophic fungi, declining environmental conditions should be a powerful cue to trigger dispersal, with the resource drawdown and the constant threat of interspecific competition being candidates as cues for dispersal. This is supported by theoretical models, where current resource level, patch size, and competition are predicted to trigger dispersal (Gilchrist et al. 2006).

As fungi exist on finite resource patches, allocation to dispersal at some point is necessary to avoid local extinction when resources fall to critically low levels. Failure of a fungus to respond accurately to these environmental cues may limit the ability of a fungus to reach new resource patches, and it is likely that strategies that promote dispersal to maximise fungal fitness have evolved under these conditions (Sæther and Engen 2015).

1.7 Colonisation, competition, and fungal combat

Competition is the most common type of interaction between saprotrophic fungi (Hiscox et al. 2018). While broadly similar to the concept of exploitation and interference competition between other organisms, competition between fungi is centred around the capture and defence of territory within finite resources. Unlike direct predation between animals or scramble competition over nutrients or light between plants, exploitation and interference competition over territory often occur in tandem, making it difficult to sensibly differentiate from one another (Cooke et al.

1984). For saprotrophic fungi, competition over space is simultaneously competition over resources

22 as capture of territory allows a fungus to have access to all nutrients therein (Hiscox et al. 2015).

Fungal competition can occur at different temporal and spatial scales, beginning with fungi accessing unoccupied resources, termed primary resource capture, and more direct antagonistic competition over held territory between fungi, termed either combat or secondary resource capture (Boddy

2000).

Control over territory within a resource patch is paramount in competition between fungi. In primary resource capture, fungi compete through time to reach a vacant resource patch before other competitors (Boddy 2001), with species with high dispersal potential more likely to succeed in this endeavour (Edman et al. 2004, Jonsson et al. 2005). Rapid spore germination and efficient resource uptake are both traits common in fungi that dominate during this phase of competition

(Boddy and Hiscox 2016). This phase is a critical point in fungal competition as early access to unoccupied space allows less combative fungi to rapidly utilise what resources they can prior to the arrival of often more competitive secondary arriving fungi. As such, early colonisation into a substrate can provide a fungus an advantage in access to nutrients in the absence of competitors.

When competition arises over territory within a resource patch that is already occupied, direct antagonistic interactions occur (Hiscox et al. 2015). Also known as fungal combat or secondary resource capture, this stage of competition is typified by the production of secondary metabolites and direct hyphal interference (Boddy and Hiscox 2016). Secondary metabolites encompass a wide range of organic compounds not used in the degradation of lignocellulose (Boddy and Hiscox 2016).

The presence of these compounds can negatively affect the functioning of competitor’s mycelia or outright inhibit mycelial growth (Hiscox et al. 2018).

In fungal competition over territory, there are generally two main outcomes: deadlock and replacement (Cooke et al. 1984). Deadlock occurs when neither fungus is able to breach into the territory of a competitor, resulting in a competitive standstill of similarly matched fungi.

23

Replacement occurs when one fungus is able to displace the territory of a competitor. Between these two extremes, partial replacement can occur, where both competitors swap territories (Boddy and Hiscox 2016). The outcomes of fungal combat are largely determined by the relative competitive abilities of the participants of fungi (Hiscox et al. 2015). Fungi can be better at defending against competitors, but poor at breaching into existing territory, or vice versa. But the competitive ability of fungi is not absolute and changes in environmental conditions can temper the competitive abilities of previously competitively superior fungi (van der Wal et al. 2013, Hiscox et al. 2018). As fungi extract nutrients from their local environment, the quality of the environment changes (Boddy

2001). Changes in the water potential or gaseous state of the resource patch as a result of the degradation can shift the balance of competition towards later arriving species (Fukami 2015), driving the progression of fungal community development.

1.8 Community assembly and interactions

Community ecology theory has largely been formulated with animal and plant systems in mind. This has hindered the development of fungal community ecology theory (Bruns 2019). This lag is in part due to the cryptic nature of fungi and the development of modern technology. For the purposes of this thesis, I refer to fungal communities as spatio-temporally grouped fungi within a spatially restricted substrate. That is to say, an interacting group of fungi bound within the lifetime of a singular resource patch. While this definition restricts fungal population dynamics at a landscape scale, the definition encompasses the scale by which individual fungi interact with other fungi through space and time.

Saprotrophic fungal communities begin to form as early as the non-interacting latent endophytic stage within living plants (Rodriguez et al. 2009, Lee et al. 2019). In wood, these

24 endophytic fungi can persist after tree death and become the first decomposers within the newly formed plant detritus, being the first to colonise a novel resource patch (Song et al. 2017). As later arrivals colonise the resource patch, all the available space becomes occupied. Early arriving species will give way to later arriving species due to competitive displacement, thus driving community development over time (Fukami et al. 2010, Dickie et al. 2012). While the progression of fungal community development in a resource patch can mimic aspects of ecological succession following disturbance, saprotrophic fungal communities can never reach a climax state (Boddy 2001). Due to the nature of finite resource patches, the end state for a fungal community results in the inevitable collapse of the patch as all resources are utilised.

Fungal community development is controlled by antagonistic interactions, but facilitative interactions are also a driver for community development (Niemelä et al. 1995, Boddy 2001, Lee et al. 2019). The production of specific organic compounds by fungal predecessors may alter the environment alongside physical modification of the substrate that may allow later successors to establish in the resource patch (Niemelä et al. 1995). Together, antagonistic and facilitative interactions between fungi can shape the trajectory of fungal community development (Fukami

2015). While there are generally early arriving species with high colonisation ability and later arriving competitive species (Boddy and Hiscox 2016), the progression of community development is still largely stochastic (Fukami et al. 2010). Here, the immigration history of fungi entering into the resource patch can drastically shift the end stage composition of the fungal community (Dickie et al.

2012, Fukami 2015). If a highly competitive species can enter into the community at an early- successional stage, it may deny entry to later arriving species (Heilmann-Clausen and Boddy 2005, van der Wal et al. 2015). Such stochasticity in the assembly of saprotrophic fungal communities is responsible for the great variation in community structures observed in nature, with the consequences of this stochastic assembly influencing the function of these decomposer communities (Fukami et al. 2010, Dickie et al. 2012).

25

The combined interactions of a community of saprotrophic fungi jointly degrade a resource patch, and in wood these interactions jointly result in wood decay (Jönsson et al. 2008). The rate of decay is largely controlled by the interactions between the idiosyncratic communities that form within wood (Fukami et al. 2010). The stochastic assembly that is a key feature of wood decay fungal communities can influence how quickly wood decays, with the order and timing of species immigration having large functional consequences (Fukami et al. 2010, Dickie et al. 2012). Known as priority effects, the order and sequence of species arrival within a community can shape how a community develops and subsequently affect the rate of decay and carbon evolution from wood

(Fukami et al. 2010). Experimental evidence suggests that such changes can be attributable to the identity of specific early arrivals (Fukami et al. 2010, Dickie et al. 2012, Hiscox et al. 2015). The identity of early arrivals is likely to alter community development because different fungal species vary in both the rate of resource use, but also the selective degradation of lignocellulose limited by enzymatic capacity. Few studies have compared how competitive ability varies with enzymatic decay mode, but if an early arriving fungus is a potent competitor, but weaker decayer, weaker competitors would be inhibited from establishing (Heilmann-Clausen and Boddy 2005, van der Wal et al. 2015). It stands to reason that a consequence of this simplified scenario would result in a reduced rate of wood decay if later arriving fungi are unable to establish within the resource patch.

1.9 Understanding competition and dispersal in saprotrophic fungi

Wood decay fungi live in a complex web of interactions, and the results of these interactions affect the amount of carbon that is released back into the environment from plant detritus. In the finite resource patches that saprotrophic fungi reside, fungi have evolved under conditions of depreciating resources (Allison and Martiny 2008, Hibbing et al. 2010). Here, competition over

26 territory, and dispersal in search for new resources have emerged as two key components of the ecology of saprotrophic fungi.

The life cycles of fungi can be incredibly complex (Pringle and Taylor 2002), but central to fungal life cycles is a need to reproduce and disperse (Gilchrist et al. 2006, Heaton et al. 2016). While any single metric for fitness is insufficient to cover all types of fungi (Gilchrist et al. 2006), fungal life history strategies should ultimately maximise reproductive potential (Stearns 2000, Sæther and

Engen 2015). With dispersal as reproduction, the optimal allocation strategy to maximise reproduction must be informed by the current state of the occupied resource patch - with the declining resource level, and current competitive pressures both providing valuable information to the organism in question (Clobert et al. 2009). The resources extracted from the finite resource patches that fungi reside should fuel reproduction but given the intense competition that surrounds the finite resource, failure to adequately defend against competitors may result in localised extinction before successful reproduction. For saprotrophic fungi, the factors that influence optimal allocation strategies are largely stochastic rather than deterministic. The finite resource patches are inherently unstable, but predictably, they decline over time. Between the resource drawdown as a result of fungal metabolism, and the constant pressure from competition, a fungus must take in cues from the changing environment to affect internal allocation switches with respect to their life history

(Clobert et al. 2009). A key question arises here:

What are the correct dispersal allocation decisions to make in an environment in flux?

To understand dispersal in fungi, we must begin by looking into the trade-offs that underpin resource allocation in fungi. To this end, I developed a model system using the ascomycetous saprotroph, Phacidium lacerum. (Order: Helotiales). Phacidium lacerum is a filamentous fungus and plant pathogen of pine trees and Rosaceae fruit (Wiseman et al. 2016, Nawrot-Chorabik et al. 2016)

27 with no previous documented presence in rotting wood. The main form of dispersal for P. lacerum is the production of pycnidial conidiomata (henceforth: pycnidia) on the surface of infect plant material, following with the release of ascospores into the environment (Crous et al. 2014). I serendipitously extracted P. lacerum from rotting wood and discovered that it had the rare quality of producing fruiting bodies in-vitro on agar plates. This gave me the opportunity to explore allocation to dispersal in a filamentous saprotrophic fungus in response to different environmental cues. In this thesis, I address the question of optimal resource allocation by considering allocation to dispersal in

P. lacerum in response to both the quality of a resource patch and the pressures of competition separately.

1.10 Competition in saprotrophic fungi

The competitive dynamics between fungi have consequences for ecological function with respect to decomposition and carbon dioxide evolution (Venugopal et al. 2017). But the environment can influence the strength of the competitive dynamics between saprotrophic fungi

(Hiscox et al. 2016). Niche theory and the competitive exclusion principle together predict that populations of any two species are unable to coexist using the same resource when competing together in space and time (Hardin 1960, Holmer and Stenlid 1997). The organism able to extract available resources at a higher rate than its competitor will eventually outcompete and exclude the other organism (Tilman et al. 1982, Wilson and Tilman 1993). These findings are supported by simple experiments that demonstrate how higher resource extraction rate can predict which species will be competitively superior during species interactions (Tilman et al. 1982, Wilson and Tilman 1993).

When applying these theories to saprotrophic fungal communities, we should expect to see few superior competitive fungi dominate and exclude less competitive fungi. But despite the incredibly high numbers of interacting saprotrophic fungi, there remains a staggeringly high diversity of species found within these resource patches (Mäkipää et al. 2017). Another key question arises here:

28

How does the environment influence competition between saprotrophic fungi?

To understand how the environment can influence competition between fungi, I compared the competitive dynamics between wood decay fungi across a simple substrate and a complex substrate. By comparing how competition progresses between interspecific interactions between fungi across simple and complex substrates, I could begin to break down the role the environment plays in mediating the competitive interactions between fungi in a more ecologically relevant setting.

In this thesis, I focused on how the environment affected saprotrophic fungi, with particular focus on two critically important components of fungal ecology: dispersal and competition. But for the last component of my thesis, I wanted to explore how interactions between saprotrophic fungi could affect the environment instead. As stated above, saprotrophic fungi (especially wood-decay fungi) are critically important in the greater carbon cycle (Cornwell et al. 2009). Herein lay the last question I had set out to answer:

How does competition between saprotrophic fungi affect resource use?

Not all fungi utilise resources at the same rate and different fungi have different competitive abilities (Hiscox et al. 2016). With a limited resource pool, fungi may have to make resource allocation decisions in response to competition, possibly downregulating resource extraction due to the energetic cost of defending against an opponent or attacking a competitor’s territory (Gilchrist et al. 2006, Heaton et al. 2016). To see how competition affects resource extraction rate in fungi, I measured mass loss in wood colonised by saprotrophic fungi alone, and mass loss in wood colonised by saprotrophic fungi exposed to interspecific competition. This allowed me to compare how

29 resource extraction rate of saprotrophic fungi differs when a fungus is met with competition than when growing alone.

1.11 An experimental approach: Thesis outline

In my thesis, I took a very first-principles approach in examining the life history strategies and ecology of wood decay fungi. By isolating how fungi alter their allocation to dispersal, and how the environment can alter competitive dynamics in fungal interactions, I can begin to understand how fungi may predictably respond to their environment. There is a dearth of theory that centres around fungi. My overarching aim was to examine how the processes of dispersal and competition fit within the greater context of fungal life history and how the environment may shape components of the ecology of fungi. To this end, I took an experimental approach informed by general ecological theory. In my research, I have highlighted dispersal and competition as what I believe to be critical components of fungal life history and ecology.

Chapter 2: In this study, I focused on how the quality of a resource patch can affect dispersal allocation in P. lacerum. I considered resource patches as islands and islands of various size and nutrient concentration using two different sized Petri-dishes and tracked the development and rate of production of pycnidia over the course of the experiment. Based on theory developed in Gilchrist et al. 2006, we predicted that resource level and patch size will both act as cues to trigger dispersal in P. lacerum.

Chapter 3: Following on from the experimental work laid out in Chapter 2, I tested how competition would affect dispersal allocation in P. lacerum. Using a similar Petri-dish approach, I set up individual, intraspecific, and interspecific treatments for P. lacerum without changing Petri-dish

30 size or nutrient quality. I predicted that P. lacerum will increase dispersal allocation to avoid intense interspecific competition.

Chapter 4: In this chapter, I focus on how the environment can affect competitive dynamics between saprotrophic fungi. The theory of competitive exclusion predicts that where two species share the same resources in time and space, a superior competitor will outcompete and exclude the weaker competitor. In this experiment, I tested whether competitive dynamics established on simple substrates (agar) would transfer over to more complex substrates (wood). To do this, I set up competitive interactions between saprotrophic fungi on Petri-dishes filled with a simple nutrient medium (malt extract agar) and mirrored the same competitive interactions in a paired wood block experiment, where wood blocks were colonised by individual fungi.

Chapter 5: In the final data chapter of this thesis, I tested how competition between saprotrophic fungi affected wood decomposition. From the same experimental set-up of Chapter 4, I measured the mass loss of wood blocks from the competition treatments in that experiment. I compared mass loss across wood blocks colonised by a single species without interaction against wood blocks colonised by a single species but placed in a paired wood block set-up. In this experiment, I focused on how competition between species and colonisation of the wood block can impact early wood decay dynamics.

With the studies in this thesis, I investigate the ecological processes of dispersal and competition in saprotrophic fungi. Within the context of finite resource patches, I draw inspiration from island biogeography to inform how dispersal and competition may shape resource allocation in life history strategies on finite resource patches. Furthermore, I study the impact of the environment on competitive dynamics between fungi and consider the reciprocal impact of competition on wood decomposition by saprotrophic fungi. Taken together, the studies presented in this thesis contribute

31 to the developing field of fungal ecology, with focus on simple experiments informed by theory.

Here, the experimental approach can provide insights on how dispersal and competition may shape life history strategies in fungi, and provide a steppingstone for future studies.

32

1.12 References

Alastruey-Izquierdo, A. et al. 2015. Susceptibility test for fungi: Clinical and laboratorial correlations in medical mycology. - Rev. Inst. Med. Trop. Sao Paulo 57 Suppl 19: 57–64.

Allison, S. D. and Martiny, J. B. H. 2008. Resistance, resilience, and redundancy in microbial communities. - Proceedings of the National in press.

Andrews, J. H. et al. 1987. Fungi, leaves, and the theory of island biogeography. - Microb. Ecol. 14: 277– 290.

Bardgett, R. D. et al. 2008. Microbial contributions to climate change through carbon cycle feedbacks. - ISME J. 2: 805–814.

Béguin, P. and Aubert, J. P. 1994. The biological degradation of cellulose. - FEMS Microbiol. Rev. 13: 25– 58.

Bennett, J. W. 1998. Mycotechnology: the role of fungi in biotechnology1Based on a lecture held at the symposium, `Progress in US Biotechnology’, at the 8th European Congress on Biotechnology (ECB8) in Budapest, Hungary, August 1997.1. - J. Biotechnol. 66: 101–107.

Bhat, M. K. and Bhat, S. 1997. Cellulose degrading enzymes and their potential industrial applications. - Biotechnol. Adv. 15: 583–620.

Blanchette, R. A. 1995. Degradation of the lignocellulose complex in wood. - Can. J. Bot. 73: 999–1010.

Boddy, L. 1999. Saprotrophic Cord-Forming Fungi: Meeting the Challenge of Heterogeneous Environments. - Mycologia 91: 13–32.

Boddy, L. 2000. Interspecific combative interactions between wood-decaying basidiomycetes. - FEMS Microbiol. Ecol. 31: 185–194.

Boddy, L. 2001. Fungal Community Ecology and Wood Decomposition Processes in Angiosperms: From Standing Tree to Complete Decay of Coarse Woody Debris. - Ecol. Bull.: 43–56.

Boddy, L. and Hiscox, J. 2016. Fungal Ecology: Principles and Mechanisms of Colonization and Competition by Saprotrophic Fungi. - Microbiol Spectr in press.

Boyd, L. A. et al. 2013. Plant-pathogen interactions: disease resistance in modern agriculture. - Trends Genet. 29: 233–240.

33

Brown, S. 2002. Measuring carbon in forests: current status and future challenges. - Environ. Pollut. 116: 363–372.

Brundrett, M. C. 1991. Mycorrhizas in Natural Ecosystems. 21: 171–313.

Bruns, T. D. 2019. The developing relationship between the study of fungal communities and community ecology theory. - Fungal Ecol. 39: 393–402.

Calhim, S. et al. 2018. Fungal spore diversity reflects substrate-specific deposition challenges. - Sci. Rep. 8: 5356.

Clobert, J. et al. 2009. Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations. - Ecol. Lett. 12: 197–209.

Cooke, R. C. et al. 1984. Ecology of saprotrophic fungi. - Longman.

Cornwell, W. K. et al. 2009. Plant traits and wood fates across the globe: rotted, burned, or consumed? - Glob. Chang. Biol. 15: 2431–2449.

Cragg, S. M. et al. 2015. Lignocellulose degradation mechanisms across the Tree of Life. - Curr. Opin. Chem. Biol. 29: 108–119.

Cramer, W. et al. 2001. Global response of terrestrial ecosystem structure and function to CO2 and climate change: results from six dynamic global vegetation models: ECOSYSTEM DYNAMICS, CO2 and CLIMATE CHANGE. - Glob. Chang. Biol. 7: 357–373.

Crous, P. W. et al. 2014. Phacidium and Ceuthospora (Phacidiaceae) are congeneric: taxonomic and nomenclatural implications. - IMA Fungus 5: 173–193.

Crowther, T. W. et al. 2012. Functional and ecological consequences of saprotrophic fungus–grazer interactions. - ISME J. 6: 1992–2001.

Dahlberg, A., Genney, D. R., & Heilmann-Clausen, J. (2010). Developing a comprehensive strategy for fungal conservation in Europe: current status and future needs. Fungal Ecology, 3(2), 50-64.

Delaquis, P. J. et al. 1989. Detachment of Pseudomonas fluorescens from biofilms on glass surfaces in response to nutrient stress. - Microb. Ecol. 18: 199–210.

Dickie, I. A. et al. 2012. Do assembly history effects attenuate from species to ecosystem properties? A field test with wood-inhabiting fungi. - Ecol. Lett. 15: 133–141.

34

Dixon, R. K. et al. 1994. Carbon pools and flux of global forest ecosystems. - Science 263: 185–190.

Edman, M. et al. 2004. Spore deposition of wood-decaying fungi: importance of landscape composition. - Ecography 27: 103–111.

Eichlerová, I. et al. 2015. Enzymatic systems involved in decomposition reflects the ecology and taxonomy of saprotrophic fungi. - Fungal Ecol. 13: 10–22.

Eriksson, K.-E. 1978. Enzyme mechanisms involved in cellulose hydrolysis by the rot fungus Sporotrichum pulverulentum. - Biotechnol. Bioeng. 20: 317–332.

Fernandez, C. W. and Kennedy, P. G. 2015. Moving beyond the black-box: fungal traits, community structure, and carbon sequestration in forest soils. - New Phytol. 205: 1378–1380.

Fukami, T. 2015. Historical Contingency in Community Assembly: Integrating Niches, Species Pools, and Priority Effects. - Annu. Rev. Ecol. Evol. Syst. 46: 1–23.

Fukami, T. et al. 2010. Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. - Ecol. Lett. 13: 675–684.

Gilchrist, M. A. et al. 2006. Identifying fitness and optimal life-history strategies for an asexual filamentous fungus. - Evolution 60: 970–979.

Goodale, C. L. et al. 2002. Forest carbon sinks in the northern hemisphere. - Ecol. Appl. 12: 891–899.

Goodell, B. et al. 2008. Fungal Decay of Wood: Soft Rot—Brown Rot—White Rot. - In: Development of Commercial Wood Preservatives. ACS Symposium Series. American Chemical Society, pp. 9–31.

Gower, S. T. 2003. Patterns and Mechanisms of the Forest Carbon Cycle. - Annu. Rev. Environ. Resour. 28: 169–204.

Grossart, H.-P. et al. 2019. Fungi in aquatic ecosystems. - Nat. Rev. Microbiol. 17: 339–354.

Hanski, I. 1998. Metapopulation dynamics. - Nature 396: 41–49.

Hardin, G. 1960. The competitive exclusion principle. - Science 131: 1292–1297.

Harmon, M. E. et al. 1990. Effects on carbon storage of conversion of old-growth forests to young forests. - Science 247: 699–702.

Heaton, L. L. M. et al. 2016. Energetic Constraints on Fungal Growth. - Am. Nat. 187: E27–40.

35

Heilmann-Clausen, J. and Boddy, L. 2005. Inhibition and stimulation effects in communities of wood decay fungi: exudates from colonized wood influence growth by other species. - Microb. Ecol. 49: 399–406.

Hibbett, D. S. et al. 2007. A higher-level phylogenetic classification of the Fungi. - Mycol. Res. 111: 509– 547.

Hibbing, M. E. et al. 2010. Bacterial competition: surviving and thriving in the microbial jungle. - Nat. Rev. Microbiol. 8: 15–25.

Hiscox, J. et al. 2015. Antagonistic fungal interactions influence carbon dioxide evolution from decomposing wood. - Fungal Ecol. 14: 24–32.

Hiscox, J. et al. 2016. Effects of pre-colonisation and temperature on interspecific fungal interactions in wood. - Fungal Ecol. 21: 32–42.

Hiscox, J. et al. 2018. Fungus wars: basidiomycete battles in wood decay. - Stud. Mycol. 89: 117–124.

Holmer, L. and Stenlid, J. 1997. Competitive Hierarchies of Wood Decomposing Basidiomycetes in Artificial Systems Based on Variable Inoculum Sizes. - Oikos 79: 77–84.

Honegger, R. 2009. Lichen-Forming Fungi and Their Photobionts. - In: Deising, H. B. (ed), Plant Relationships. Springer Berlin Heidelberg, pp. 307–333.

Jonsson, B. et al. 2005. Ecology of species living on dead wood – lessons for dead wood management. - Silva Fenn. in press.

Jönsson, M. T. et al. 2008. Colonization and extinction patterns of wood-decaying fungi in a boreal old- growth Picea abies forest: Colonization-extinction dynamics of wood-decaying fungi. - J. Ecol. 96: 1065–1075.

Kinlan, B. P. and Gaines, S. D. 2003. Propagule dispersal in marine and terrestrial environments: A community perspective. - Ecology 84: 2007–2020.

Lancaster, J. and Downes, B. J. 2017. A landscape-scale field experiment reveals the importance of dispersal in a resource-limited metacommunity. - Ecology 98: 565–575.

Larsen, B. B. et al. 2017. Inordinate Fondness Multiplied and Redistributed: the Number of Species on Earth and the New Pie of Life. - Q. Rev. Biol. 92: 229–265.

36

Lee, M. R. et al. 2019. Good neighbors aplenty: fungal endophytes rarely exhibit competitive exclusion patterns across a span of woody habitats. - Ecology 100: e02790.

Lopez, M. J. et al. 2007. Lignocellulose-degrading enzymes produced by the ascomycete Coniochaeta ligniaria and related species: Application for a lignocellulosic substrate treatment. - Enzyme Microb. Technol. 40: 794–800.

Lustenhouwer, N. et al. 2020. A trait-based understanding of wood decomposition by fungi. - Proc. Natl. Acad. Sci. U. S. A. 117: 11551–11558.

Mäkipää, R. et al. 2017. Interactions between soil- and dead wood-inhabiting fungal communities during the decay of Norway spruce logs. - ISME J. 11: 1964–1974.

Maynard, D. S. et al. 2017. Fungal interactions reduce carbon use efficiency. - Ecol. Lett. 20: 1034–1042.

McDougald, D. et al. 2011. Should we stay or should we go: mechanisms and ecological consequences for biofilm dispersal. - Nat. Rev. Microbiol. 10: 39–50.

McGuire, K. L. and Treseder, K. K. 2010. Microbial communities and their relevance for ecosystem models: Decomposition as a case study. - Soil Biol. Biochem. 42: 529–535.

Nagy, L. G. et al. 2016. Comparative Genomics of Early-Diverging Mushroom-Forming Fungi Provides Insights into the Origins of Lignocellulose Decay Capabilities. - Mol. Biol. Evol. 33: 959–970.

Naranjo-Ortiz, M. A. and Gabaldón, T. 2019. Fungal evolution: major ecological adaptations and evolutionary transitions. - Biol. Rev. Camb. Philos. Soc. 94: 1443–1476.

Nawrot-Chorabik, K. et al. 2016. Interactions between callus cultures ofPinus sylvestrisand pine fungi with different trophic properties. - For. Pathol. 46: 179–186.

Niemelä, T. et al. 1995. Interactions of fungi at late stages of wood decomposition. - Ann. Bot. Fenn. 32: 141–152.

Norros, V. et al. 2012. Dispersal may limit the occurrence of specialist wood decay fungi already at small spatial scales. - Oikos 121: 961–974.

Oberle, B. et al. 2018. When a tree falls: Controls on wood decay predict standing dead tree fall and new risks in changing forests. - PLoS One 13: e0196712.

37

Peay, K. G. and Bruns, T. D. 2014. Spore dispersal of basidiomycete fungi at the landscape scale is driven by stochastic and deterministic processes and generates variability in plant-fungal interactions. - New Phytol. 204: 180–191.

Peay, K. G. et al. 2016. Dimensions of biodiversity in the Earth mycobiome. - Nat. Rev. Microbiol. 14: 434– 447.

Persson, Y. et al. 2011. Do bark beetles facilitate the establishment of rot fungi in Norway spruce? - Fungal Ecol. 4: 262–269.

Pietsch, K. A. et al. 2014. Global relationship of wood and leaf litter decomposability: the role of functional traits within and across plant organs: Global relationship of wood and leaf litter decomposability. - Glob. Ecol. Biogeogr. 23: 1046–1057.

Pringle, A. and Taylor, J. 2002. The fitness of filamentous fungi. - Trends Microbiol. 10: 474–481.

Purahong, W. et al. 2016. Are correlations between deadwood fungal community structure, wood physico-chemical properties and lignin-modifying enzymes stable across different geographical regions? - Fungal Ecol. 22: 98–105.

Riley, R. et al. 2014. Extensive sampling of basidiomycete genomes demonstrates inadequacy of the white-rot/brown-rot paradigm for wood decay fungi. - Proc. Natl. Acad. Sci. U. S. A. 111: 9923–9928.

Rodriguez, R. J. et al. 2009. Fungal endophytes: diversity and functional roles. - New Phytol. 182: 314– 330.

Ruiz-Dueñas, F. J. and Martínez, A. T. 2009. Microbial degradation of lignin: how a bulky recalcitrant polymer is efficiently recycled in nature and how we can take advantage of this. - Microb. Biotechnol. 2: 164–177.

Sæther, B.-E. and Engen, S. 2015. The concept of fitness in fluctuating environments. - Trends Ecol. Evol. 30: 273–281.

Sánchez, C. 2009. Lignocellulosic residues: biodegradation and bioconversion by fungi. - Biotechnol. Adv. 27: 185–194.

Sawyer, L. K. and Hermanowicz, S. W. 2000. Detachment of Aeromonas hydrophila and Pseudomonas aeruginosa due to variations in nutrient supply. - Water Sci. Technol. 41: 139–145.

Schimel, D. S. 1995. Terrestrial ecosystems and the carbon cycle. - Glob. Chang. Biol. 1: 77–91.

38

Schleheck, D. et al. 2009. Pseudomonas aeruginosa PAO1 preferentially grows as aggregates in liquid batch cultures and disperses upon starvation. - PLoS One 4: e5513.

Sista Kameshwar, A. K. and Qin, W. 2018. Comparative study of genome-wide plant biomass-degrading CAZymes in white rot, brown rot and soft rot fungi. - Mycology 9: 93–105.

Smith, M. L. et al. 1992. The fungus Armillaria bulbosa is among the largest and oldest living organisms. - Nature 356: 428–431.

Song, Z. et al. 2017. Fungal endophytes as priority colonizers initiating wood decomposition. - Funct. Ecol. 31: 407–418.

Stearns, S. C. 1992. The evolution of life histories. - sidalc.net.

Stearns, S. C. 2000. Life history evolution: successes, limitations, and prospects. - Naturwissenschaften 87: 476–486.

Swift, M. J. 1977. The Roles of Fungi and Animals in the Immobilisation and Release of Nutrient Elements from Decomposing Branch-Wood. - Ecol. Bull.: 193–202.

Tilman, D. et al. 1982. Phytoplankton Community Ecology: The Role of Limiting Nutrients. - Annu. Rev. Ecol. Syst. 13: 349–372.

Todd, N. K. and Rayner, A. D. M. 1980. Fungal individualism. - Sci. Prog. 66: 331–354. van der Wal, A. et al. 2013. A thready affair: linking fungal diversity and community dynamics to terrestrial decomposition processes. - FEMS Microbiol. Rev. 37: 477–494. van der Wal, A. et al. 2015. Neglected role of fungal community composition in explaining variation in wood decay rates.

Van Dyke, M. C. C. et al. 2019. Fantastic yeasts and where to find them: the hidden diversity of dimorphic fungal pathogens. - Curr. Opin. Microbiol. 52: 55–63.

Venugopal, P. et al. 2017. Assemblage composition of fungal wood-decay species has a major influence on how climate and wood quality modify decomposition. - FEMS Microbiol. Ecol. in press.

Walker, G. M. and White, N. A. 2017. Introduction to fungal physiology. - In: Fungi. John Wiley & Sons, Inc., pp. 1–35.

39

Wilson, S. D. and Tilman, D. 1993. Plant Competition and Resource Availability in Response to Disturbance and Fertilization. - Ecology 74: 599–611.

Wiseman, M. S. et al. 2016. A New Postharvest Fruit Rot in Apple and Pear Caused by Phacidium lacerum. - Plant Dis. 100: 32–39.

40

CHAPTER 2

Environmental cues to dispersal in a filamentous fungus in simulated islands

Justin Y. Chan, Stephen P. Bonser, Jeff R. Powell, William K. Cornwell

41

2.1 Abstract

Airborne dispersal is a key part of the life history of many saprotrophic fungi. Theory suggests a transition from growth and resource capture to airborne dispersal at some point as the resource availability in a patch declines, but in the absence of an experimental model system this theory has remained abstract. For saprobes, resources are arrayed in an ever-shifting archipelago of islands with the quality of each island being defined by patch size and resource density. We tracked how Phacidium lacerum, a saprotrophic fungus, allocated to dispersal in small and large islands of varying resource density. We found that Phacidium altered the timing and rate of dispersal allocation in response to both patch size and resource density. On small resource islands, Phacidium drastically increased dispersal allocation after reaching the edge of the patch; if resource density was sufficient, on larger resource islands, Phacidium began allocation to dispersal prior to reaching the edge of the island, suggesting an additional absolute total resource level cue. These results are consistent with a two-cue model for the switch to allocation to airborne dispersal: 1) absolute resource level controlled by the fungus, 2) the fungus’ perception of patch size. This can be thought of as a mix between a full resource allocation switch (bang-bang) if the fungus perceives the patch is fully occupied with a smaller magnitude early shift (bet-hedging) if absolute resource level crosses a threshold.

42

2.2 Introduction

Microorganisms are in every habitat on earth, inhabiting every nook and cranny in available niche space (Baas-Becking 1934, Fuhrman 2009, Blasche et al. 2017). As part of all major biogeochemical cycles, microbial communities perform vital ecosystem functions including nitrogen- fixation (Finlay et al. 1997, Muller et al. 2018) and the decomposition of organic matter (Boer et al.

2005, Schimel and Schaeffer 2012, Wilhelm et al. 2019). The environment for microbial life is constantly changing and microorganisms have evolved under variable environmental conditions

(Allison and Martiny 2008, Hibbing et al. 2010). Resource allocation to growth, resource capture, and dispersal in microorganisms must be informed by changes in the environment (Congdon et al. 2001).

But a question remains: how does a microorganism make appropriate resource allocation decisions in an environment in flux?

Within a given environment, distinct microbial communities occupy restricted resource patches. For heterotrophic microorganisms like fungi, individual metabolic processes and competitive interactions deplete local resources resulting in a decline of patch quality through time

(Finn 2001, Gilchrist et al. 2006). Taken together, heterotrophic microorganisms exist in a spatially restricted declining-resource environment. Notably, both mutualistic ectomycorrhizal fungi and saprotrophic fungi address this problem with a distinct airborne dispersal phase (Edman et al. 2004b,

Norros et al. 2012). Airborne dispersal plays a critical role in the assembly of saprotrophic fungal communities across resource patches (Stenlid and Gustafsson 2001, Johst et al. 2002, Edman et al.

2004a), and fungi are considered a key part of airborne biodiversity (Womack et al. 2010).

Saprotrophic fungi derive their nutrition from detritus, and the sources of that detritus is episodic in the form of the death of plants, animals and other microbes. Each of these pieces of detritus is then inexorably reduced to nothing as a result of saprotrophic metabolism, a process collectively known as decomposition (Jonsson et al. 2005). Together the patchy creation of detritus and the decomposition of each patch creates an ever-shifting archipelago of resource-rich islands within a

43 resource-poor matrix. In this patchy world, fungi have two means of dispersing from patch to patch: hyphal extension and air-borne spores. Fungal hyphae are well adapted when the resource-poor matrix is not too extensive but dispersal via hyphal extension over very long distances may be costly or impossible (Edman et al. 2004b). To bridge extensive gaps and find new resource islands before competitors, many fungi disperse via air-borne spores.

Theoretical consideration of fungal allocation between hyphae and spores has suggested that resource level, patch size, and increased competitive interactions on an island should act as cues (Gilchrist et al. 2006). Changes in environmental quality have also been observed to be a cue to trigger dispersal in microorganisms (McDougald et al. 2011). Critically, a reduction in nutrient levels is observed to be a primary driver of dispersal allocation (Delaquis et al. 1989, Sawyer and

Hermanowicz 2000, Schleheck et al. 2009). This observation in dispersal behaviour in microorganisms is in line with the “informed dispersal” framework (sensu (Clobert et al. 2009)).

Informed dispersal theory posits that organisms are able to gather information about their local environment and their internal condition. The combined information from the external and internal environment can induce a dispersal response if the expected risk/benefit balance of remaining stationary exceeds the risk/benefit balance of dispersal. Fungi are responsive to external cues and are able to alter their colony growth form in response to resources (Heaton et al. 2012), fungivory

(Crowther et al. 2011), and mycelial interactions (Rotheray et al. 2008). But the extent to which environmental cues shape allocation to dispersal in fungi is unclear. When resources become limiting, a reduced allocation to the production of reproductive propagules would limit the ability of a fungus from reaching new resource sites. Thus, an upregulation of resource allocation to dispersal in response to declining resources may maximise fitness in certain conditions.

Allocation to reproductive dispersal should be optimised under life history trade-offs to maximise the production of reproductive propagules dependent on current environmental conditions (Kozłowski 1992, Heaton et al. 2016). Under the Gilchrist patch array model, a

44 filamentous fungus is predicted to employ a “bang-bang” strategy of resource allocation: allocation of all resources towards mycelial growth followed by total allocation towards dispersal (Vincent and

Pulliam 1980, Gilchrist et al. 2006). The first phase of the “bang-bang” strategy should increase the capacity of a fungus to capture resources within a patch through mycelial expansion. The subsequent conversion of captured resources into air-borne propagules will, in theory, maximise the fitness of the fungus (Gilchrist et al. 2006). But this simple model leaves out two key aspects of variable resource patches: the size of the patch relative to the fungus and the resource density of the patch. The variability that surrounds the quality of a patch can drive the evolution of life history strategies towards risk spreading or bet hedging strategies (Wilbur and Rudolf 2006, Venable 2007,

Childs et al. 2010). Altering the timing of reproduction may be an adaptive strategy to ensure successful dispersal under variable environmental conditions, where a risk-reduction strategy may provide fitness benefits compared to a rigid strategy in a changing environment (Meyers and Bull

2002, Sæther and Engen 2015). Both patch size and resource density alter the quality of an island and can affect the total resource available to the fungus, and thus shape the pattern of resource allocation within the confines of life history trade-offs.

An organism is unable to simultaneously optimise two or more traits that require the same limiting resource(s), leading to observed restrictions on the evolution and expression of life history allocation strategies (Stearns 1992). Allocation to airborne dispersal comes at an opportunity cost – those resources allocated to dispersal structures and spores could be used for defence, resource acquisition, or growth (Kozłowski 1992, Gilchrist et al. 2006, Heaton et al. 2016, Chan et al. 2019). To test how an environment can trigger dispersal allocation in fungi and alter the patterns in trade-offs in resource allocation under restricted nutrient environments, we tested how a filamentous fungus alters its dispersal behaviour when grown under varying levels of nutrient concentration in a Petri- dish experiment.

45

Previous work has developed the asexual filamentous fungus Phacidium lacerum (Order:

Helotiales), an ascomycetous saprotroph found in rotting wood, as a tractable model system for research on airborne dispersal in fungi (Chan et al. 2019). It is also a pathogen of pine trees (Nawrot-

Chorabik et al. 2016) and Rosaceae fruits (Wiseman et al. 2016). The primary form of dispersal of P. lacerum is through the production of fruiting bodies called pycnidial conidiomata (henceforth, pycnidia) on the surface of infected plant material (Crous et al. 2014). Pycnidia are macroscopic capsules that contain copious pycnidiospores (asexual conidia) that are released into the environment. We chose P. lacerum as our focal species as it has the property of producing fruiting bodies in-vitro. This quality is rare in saprotrophic fungi, and consistent induced sporulation in basidiomycetes is difficult to achieve in a lab environment (Brandt 2013), which makes P. lacerum an attractive model species for studying dispersal allocation cues in a saprotrophic fungus.

With this model system, we have the opportunity to observe patterns in dispersal in response to cues in the environment. We performed targeted tests on the resource allocation trade- offs that underpin the life history of the fungus and vary growth-medium concentration and Petri- dish size as proxies for resource level and island size, respectively, to test interactions with life history allocation trade-offs. Here, we quantified allocation to dispersal by measuring the pycnidial development in the fungal colony on Petri-dishes in a lab environment. We predicted that: 1) total resource level and resource island edge detection will act as cues to trigger dispersal allocation in a fungal colony, and 2) this fungal species will employ a mixed pattern of allocation to dispersal dependent on the quality of its island.

46

2.3 Methods

2.3.1 Species isolation and identification

We isolated a single culture of Phacidium lacerum from rotting Eucalyptus tereticornis logs by extracting wood chips from the centre of rotting logs with a flame-sterilised chisel. Logs were collected from a woodland in Richmond, NSW (33°37’04.0”S 150°44’25.3”E) in February 2016. We placed the wood chips onto 2% malt extract agar (MEA) and subcultured from the edge of the growing colony until we obtained a pure culture. We extracted DNA from the growing tips of fungal hyphae at the edge of the culture using the DNeasy Plant Mini Kit (Qiagen, Chadstone, Victoria,

Australia) as per the manufacturer’s instruction. We amplified the ITS (ITS1F & ITS4) region of rDNA

(White et al. 1990) through PCR amplification and analysed the amplicons using a ABI3500 Genetic

Analyser (Applied Biosystems, Life Technologies, Mulgrave, Victoria, Australia). The species identity was assigned by conducting a BLAST search against the NCBI Nucleotide database. From the original isolate, we subcultured onto five new Petri dishes of 2% MEA from which we inoculated from to create our experimental plates. We maintained cultures at 4oC in the dark.

2.3.2 Experimental Design

We prepared malt extract agar media at three concentrations to represent a high, mid, and low resource treatment (1%, 0.5%, and 0.1% MEA respectively, Supplementary material). We used small and large Petri-dishes (9 cm and 14 cm diameter respectively) to test how the size of a resource island affected allocation to dispersal. We combined nutrient quality and Petri-dish size for a total of six treatments. For each treatment, we had six replicates for a total of 36 treatment plates with an added 12 plates with two sizes of plates with water agar to act as our control. Prior to the experiment, we inoculated our focal species on water agar (10mg agar /L water) to normalise hyphal

47 density among treatment inocula. The fungi were allowed to establish on the H2O agar for 7 days before 5mm diameter plugs were taken from the growing edge of the colonies and inoculated onto the centre of the treatment plates. All Petri-dishes were sealed with parafilm and incubated in the dark at 25oC.

2.3.3 Image Analysis

We measured colony size as a measure of fungal growth rate. We observed colony size visually and we tracked colony growth rate every 24 hours by tracing the perimeter of the colony on an acetate sheet. Colony relative growth rate (RGR) was then calculated as the proportional (ie. log) difference in colony area (Lambers and Poorter 1992). We calculated the difference in colony growth rate between treatments by comparing radial extension rate to avoid the impact of increasing radius on colony area. Pycnidia emerged after five days and we began tracking pycnidia development every

12 hours after their first emergence. We marked each pycnidium on an acetate sheet and then counted the marks by scanning the sheet using a Canoscan LiDE 210 scanner (300dpi resolution), thresholding the image and using the automated counting functionality of ImageJ (NIH, USA). We measured the number of pycnidia every 12 hours for the first 10 days, before switching to tracking pycnidia every 24 hours after the initial reproductive allocation phase. The experiment ran for a total of 20 days (480 hours). Pycnidia densities were measured as number of pycnidia per colony area.

48

2.3.4 Statistical Analysis

To compare how treatment affected colony growth rate, we ran a two-way analysis of variance (ANOVA) on radial extension rate. Tukey’s HSD post hoc tests were used to assess any significant differences. To test the timing and initiation of dispersal allocation in P. lacerum, we fitted a survival analysis/time-to-event model to data for when colonies produced at least 50 pycnidia. This model compares when P. lacerum colonies produces 50 pycnidia across different treatments as a variable. We used the survival package (Therneau 2015) in R (v.3.4.2) to fit a Cox proportional hazards (CoxPH) model, along with a significance test for treatment, nutrient, and island size effects.

To rule out coarse total-resource-uptake as a possible cue for dispersal, we ran a two-way ANOVA in

R (v.3.4.2) comparing resource concentration of the colony area at initiation of pycnidia production across nutrient concentration and Petri dish size. Colonies in the low nutrient treatments were excluded from analysis as they did not meet the minimum pycnidia production threshold. To test how pycnidial density varied across island size and nutrient concentration by the end of the experiment, we ran a two-way ANOVA in R (v.3.4.2) with treatment as a fixed effect for final pycnidial density at day 20. To examine the relationship between allocation to colony growth and dispersal, we fitted a generalised linear mixed-effects model accounting for repeated measures of the same Petri dish with pycnidia production as the response term. The model was fit to a Poisson distribution, and Petri dish replicates were treated as a random effect, with treatment and relative growth rate as fixed effects.

49

2.4 Results

2.4.1 Colony growth rate and patch edge detection

Treatment had a significant effect on colony growth rate, with a significant interaction

between nutrient concentration and size of Petri dish (p < 0.0001, ANOVA, F2,29=109.1). Overall,

Phacidium lacerum colonies grown on the high nutrient treatments grew more rapidly than fungi in lower nutrient treatments (p<0.001, Tukey’s HSD). All colonies on the 1% nutrient concentration reached the edge of their Petri dish on days 7 and 10 for the small and large Petri dish respectively

(Fig. 1). Colonies on the 1% nutrient concentration had an average radial growth rate of 0.24 mm/h

(±0.0007 SE) (small Petri dish) and 0.27 mm/h (±0.0006 SE) (large Petri dish) during active mycelial expansion of the colony. Colonies on the 0.5% nutrient concentration reached the edge of their Petri dish on days 8 and 12 for the small and large Petri Dish respectively, and colonies on the 0.1% reached the edge of their Petri dish on days 9 and 14 for the small and large Petri dish respectively.

Colonies growing on 0.5% nutrient concentration had a radial growth rate of 0.21 mm/h (±0.0004

SE) (small Petri dish) and 0.22 mm/h (±0.0006 SE) (large Petri dish) during colony expansion, while colonies growing on the 0.1% nutrient concentration had a radial growth rate of 0.18 mm/h

(±0.0008 SE and ±0.0005 SE) (small and large Petri dishes) during colony expansion. Colonies on the

H2O agar did not reach the edge of the Petri-dish by day 20. P. lacerum colonies began the production of pycnidia (<50 pycnidia) prior to reaching the edge of the Petri dish across all treatments, but colonies on the small islands had a marked increase in pycnidia production following complete colonisation of the Petri dish (Fig. 1).

50

2.4.2 Effect of resource level on pycnidia production

We found that nutrient concentration had a significant effect on the timing of pycnidia production, with colonies on the 1% nutrient treatment producing pycnidia significantly earlier than other nutrient treatments (p<0.0001, CoxPH, Fig. 2). We also found that P. lacerum colonies on larger islands consistently produced pycnidia earlier than the small island counterparts on the same

nutrient level (p<0.0001, CoxPH, Fig. 2). Colonies on the H2O agar did not produce pycnidia over the course of the experiment (20 days). By the end of the experiment, colonies on the 1% nutrient treatment had significantly higher pycnidial density than colonies on other nutrient treatments

(p<0.0001, Tukey’s HSD), with colonies on the 0.5% nutrient treatment having significantly higher pycnidial density than colonies on the 0.1% nutrient treatment (p<0.0001, Tukey’s HSD, Table 1).

There were overall significant differences in resources captured by the colonies across size (p <

0.0001, ANOVA, F1,20=74.21) and nutrient concentration (p < 0.0001, ANOVA, F1,20=383.21) of the Petri dish for initiation of dispersal allocation. There was no significant interaction between size and

nutrient concentration (p=0.23, ANOVA, F1,20=1.53).

51

Figure 1. The relative growth rate (RGR) and pycnidia produced (per 12 hours) between treatments. Axes are standardised across graphs. Lines denote when P. lacerum colonies reached the edge of the Petri-dish. Colonies on the small islands in the 1% and 0.5% nutrient treatment had a dramatic increase in pycnidia production following the colony reaching the edge of the Petri dish.

Colonies on the larger islands in the 1% and 0.5% nutrient treatment began producing pycnidia at comparable rates prior to reaching the edge of the Petri dish. Colonies in the 0.1% nutrient treatment produced very few pycnidia.

52

Table 1. Tukey’s HSD test comparisons of treatment effects on pycnidial density. Pycnidial density was not significantly different between P. lacerum colonies on the 0.1% and 1% nutrient concentration between Petri dish size (S0.1-L0.1 and (S1-L1). Pycnidial density was significantly different across all other treatments.

Tukey HSD p L0.5-L0.1 <0.0001 L1-L0.1 <0.0001 S0.1-L0.1 0.999999 S0.5-L0.1 <0.0001 S1-L0.1 <0.0001 L1-L0.5 <0.0001 S0.1-L0.5 <0.0001 S0.5-L0.5 <0.0001 S1-L0.5 <0.0001 S0.1-L1 <0.0001 S0.5-L1 <0.0001 S1-L1 0.782728 S0.5-S0.1 <0.0001 S1-S0.1 <0.0001 S1-S0.5 <0.0001

53

Figure 2. Comparison of pycnidia production per 12 hours for P. lacerum against relative growth rate. There was a consistent negative relationship between RGR and pycnidia production (P <

0.001, GLM). Colonies on the small islands had negligible production of pycnidia during active growth (RGR>0). Colonies on the large islands began producing pycnidia prior to the cessation of active growth of the colony (RGR=0).

54

2.4.3 Patterns of allocation to growth and dispersal across different islands

Phacidium lacerum colonies on small islands had greatly reduced pycnidial development rates during active growth prior to reaching the patch edge (RGR > 0, Fig. 2), while colonies on larger islands began pycnidial development prior to reaching the edge of the plate and before cessation of colony growth (RGR > 0, Fig. 2). Across all treatments, colonies on the larger plate began pycnidia production earlier than colonies on smaller islands at the same resource concentration (P<0.0001,

CoxPH, Fig. 2). We observed a consistent negative relationship between relative growth rate and pycnidia production across both island size and nutrient concentration (p < 0.0001, GLM, Fig. 2).

2.5 Discussion

2.5.1 Cues for dispersal

Increased total nutrient level hastened allocation to dispersal in Phacidium lacerum colonies.

We found that between island sizes, increasing nutrient concentration consistently resulted in an earlier allocation to dispersal, with colonies on the 1% nutrient concentration producing pycnidia earlier than colonies on the 0.5% and 0.1% nutrient treatments. Colonies on larger islands also produced pycnidia significantly earlier across all nutrient treatments compared to their counterparts on smaller islands. We found that colonies began production of pycnidia at different points of total resource uptake, suggesting that variation across dispersal time cannot be solely attributed to a critical resource satiation point for the fungus.

Overall, Phacidium lacerum colonies on smaller islands markedly increased pycnidia production following the colony reaching the edge of the Petri-dish (Fig. 1). This is expected as allocation to growth will stop after complete patch occupancy, resulting in a complete allocation switch to dispersal. On the other hand, P. lacerum colonies on larger islands produced pycnidia prior

55 to the colony reaching the edge of the dish. While nutrient concentration per area on the Petri-dish is the same between island sizes, total nutrient level is greater on larger islands and may contribute to greater availability of resources via diffusion of nutrients to the colony (Read and Stribley 1975).

This is in line with our predictions that increasing total resource availability will trigger earlier allocation to dispersal. We can interpret this early allocation to dispersal as the fungus meeting a minimum resource threshold for reproduction. A greater availability of resources will allow for faster growth and earlier conversion of resources to reproductive material. This suggests that absolute resource level control by the fungal colony could act as a cue or trigger for early dispersal if conditions are met.

Two cues within the patch may thus induce a switch from growth to dispersal allocation within a fungal colony. Firstly, edge detection and total patch occupancy will result in a complete switch to dispersal in the absence of further available space for growth. Secondly, increased resource availability through a higher resource density may trigger an early switch to dispersal during active colony growth. For a fungus growing in a declining resource island, differences in resource quality and size of an immediate patch may inform the decision to upregulate allocation towards dispersal (Clobert et al. 2009). Taking both total resource level and patch size as cues may alter pattern of allocation to dispersal to maximise likelihood of successful dispersal in an uncertain environment (Venable and Brown 1988).

2.5.2 Patterns in dispersal allocation between island sizes

In the model presented in Gilchrist et al. (2006), the optimal allocation strategy is to prioritise growth and maximise colony size, subsequently enhancing resource removal, before switching allocation to maximise spore production following total patch occupancy. Fungi on smaller islands exhibit the optimal dispersal allocation pattern congruent with the model predicted pattern

56

(Fig. 1). In our experiment, we observe P. lacerum colonies on smaller islands growing to the edge of the Petri-dish prior to a switch to complete allocation to dispersal. Our data supports this hypothesis, with colonies on smaller islands producing minimal pycnidia during active colony growth

(Fig. 2), followed by a massive production of pycnidia after occupying the entire Petri-dish (Fig.1).

This pattern of resource allocation mirrors the bang-bang strategy for dispersal (Gilchrist et al. 2006).

There is strong evidence for a trade-off here in the pattern of resource allocation, with colonies on small islands unable to simultaneously maximise growth and dispersal under the resource constraints (Fig. 2).

The pattern of resource allocation is not consistent across island sizes. Phacidium lacerum colonies on larger islands began allocation to dispersal prior to reaching the limit of their patch (Fig.

1) and did not exhibit a bang-bang resource allocation switch towards dispersal. We found that P. lacerum colonies on larger islands had an earlier allocation to dispersal compared to colonies on smaller islands. The early production of pycnidia may be a bet-hedging strategy and is expected under the variable dynamics in ephemeral resource patches. However, in our experiment, resources within the Petri dish are not heterogeneously distributed. The early allocation to dispersal during colony growth may be attributed to nutrient translocation through diffusion and capillary action by the leading hyphal front of a colony (Read and Stribley 1975). Increased access to resources may lead to an allocation switch to dispersal if the fungus reaches a minimum threshold (Kozłowski

1992), and edge detection may play a lesser role in larger patches as fungi have evolved under variable patch conditions. Under these circumstances, a risk-reduction or bet-hedging strategy with an early allocation to dispersal may be maintained even if the local patch is homogeneous (Moran

1992), with early allocation to dispersal providing an advantage when trying to capture new non- spatially contiguous resources.

Under natural conditions, increasing the size of a resource island may introduce more within-patch heterogeneity at the scale of the fungal hyphae. On the forest floor, detritus represents

57 a resource island for saprotrophic fungi, and fungi growing on smaller resource islands (i.e. a single leaf) can reach the limit of the patch and follow a bang-bang strategy of resource allocation to dispersal (Gilchrist et al. 2006). But fungi establishing on larger resource islands (i.e. a fallen tree) have more space to cover which could introduce more stochasticity and variability in the patch

(Wiens 1989). For saprotrophic fungi, failure to produce reproductive propagules for passive dispersal would result in local extinction. Early allocation to dispersal may then be an adaptive response to an unpredictable environment. In line with our predictions, this mixed pattern of allocation to dispersal may thus be an adaptive response to the stochastic creation of resource islands of varying quality.

2.5.3 A two-cue model for dispersal in fungi

One interpretation of the results of this experiment is a two-cue model of allocation to dispersal (Fig. 3). In an archipelago of spatially restricted declining-resource islands, a fungus has to detect the quality of its immediate environment. Our results suggest that at least patch-size and resource concentration are integrated to trigger a dispersal allocation response. Edge detection and total patch occupancy would result in a total switch in allocation from growth to dispersal in the absence of space to grow. While a high resource concentration in a patch may meet a minimum resource threshold for dispersal. Immediate local resource availability alone may be insufficient as a cue as it lacks information on the longevity of a patch without detection of the patch size (Finn 2001,

Clobert et al. 2009). While delayed dispersal allocation prior to patch edge detection may be maladaptive if the patch is very large and has imminent neighbouring competitors or collapses before successful allocation to dispersal (Clobert et al. 2009, Bonte et al. 2012, Smith et al. 2018). An integration of detection of both patch size and resource concentration may provide a fungal colony the most information to trigger allocation to dispersal in a variable environment (Fig. 3). In the

Gilchrist model (2006), the focus on mycelial density and resource extraction provides valuable

58 insight to optimal dispersal allocation in a fungus, wherein increasing somatic growth, and subsequently resource patch removal, maximises spore production. But the inclusion of variable elements to patch quality is an important addition to this model.

In our experiment, P. lacerum colonies on small experimental islands have a drastic upregulation of dispersal allocation following complete occupancy of the Petri-dish (Fig. 1). While this reflects the optimal allocation strategy presented in the Gilchrist model, it also reflects constraints in resource allocation. Upon complete occupancy of the resource island (the Petri-dish), this frees the fungus to completely allocate to dispersal in the absence of other trade-offs. Our data suggests that fungi are able to convert gathered resources to pycnidia production to varying levels based on the different nutrient levels available. Total conversion of resources to dispersal allocation may then be restricted by available nutrients, but internal trade-offs like somatic maintenance

(Doust 1989, Heaton et al. 2016) or defence (Chan et al. 2019) may limit the absolute total allocation to dispersal. We also observe the early production of pycnidia prior before complete patch occupancy in fungal colonies on large experimental islands. This early production of pycnidia may reflect an initial allocation to dispersal once a minimum resource level is attained by the colony and may be a bet-hedging strategy. This early allocation to dispersal may shift earlier or later depending on the resource concentration of a given resource island. Minimum resource level for dispersal may act as a one of the vital cues that triggers initial dispersal allocation in fungi to ensure a degree of dispersal success in the unpredictable environment of ephemeral resource islands where many fungi reside (Fig. 3).

59

Figure 3. A conceptual model of the two-cue switch in resource allocation to dispersal. A minimum resource level attained by the fungal colony triggers an initial allocation swap to dispersal from growth (solid red line). A total switch in allocation to dispersal follows detection of a patch edge (dashed blue line). Resource allocation is shared between growth and dispersal until a complete switch to dispersal allocation following total patch occupancy (solid red line).

60

Fungi exist in a patchy world, with many new patches forming randomly in an environment, but with a degree of predictability. In saprotrophic fungi, these islands occur in a landscape randomly (Jonsson et al. 2005) and with variable quality (Abrego and Salcedo 2013). The inclusion of the resource level and patch size detection into the model of our understanding of dispersal allocation in fungi can help explain the patterns that we are observing in the environment. As saprobes, metabolic processes and competitive interactions inevitably result in resource drawdown that directly reduces the lifetime of a given resource island (Gilchrist et al. 2006). In this arena, a fungus must be ready to disperse as the costs of failure-to-disperse is high (Sæther and Engen 2015).

Here, we observe fungal colonies employ both a full resource allocation switch (bang-bang) on small experimental islands, and an early switch to dispersal prior to complete patch occupancy (bet- hedging) on large experimental islands. Our data suggest that fungi can take information from their island and pursue a mixed strategy of allocation to dispersal given different environmental contexts which may reflect an evolutionary history under unpredictable conditions.

2.5.4 Conclusion

The patterns of resource allocation in fungi remains a central question to the basic life history of fungi (Pringle and Taylor 2002, Klein and Paschke 2004, Prosser et al. 2007, Ho et al. 2017).

Fungi have a dizzying array of mating types and strategies for sexual and asexual reproduction

(Chamberlain and Ingram 1997, Taylor et al. 1999, Heitman 2015, Ojeda-López et al. 2018), but life history evolution remains a fundamental unknown in fungi. The development of first-principles theory is critically needed for a better understanding of this hidden kingdom. Here, we have taken advantage of an asexual filamentous fungus to demonstrate how the local environment can present cues that can influence fungal life history allocation. The conceptual model from this work may be used to investigate other species with more complex sexual reproductive strategies. We believe that the basic principles presented in this study will provide a valuable springboard towards building a

61 greater understanding of the patterns in fungal life history we are observing. Further focus on how the environment shapes fungal evolution within ephemeral resource islands will provide valuable insight into the trade-offs that structure life history allocation in fungi.

62

2.6 References

Abrego, N. and Salcedo, I. 2013. Variety of woody debris as the factor influencing wood-inhabiting fungal richness and assemblages: Is it a question of quantity or quality? - For. Ecol. Manage. 291: 377–385.

Allison, S. D. and Martiny, J. B. H. 2008. Resistance, resilience, and redundancy in microbial communities. - Proceedings of the National in press.

Baas-Becking, L. G. M. 1934. Geobiologie; of inleiding tot de milieukunde. - WP Van Stockum & Zoon NV.

Blasche, S. et al. 2017. Model microbial communities for ecosystems biology. - Current Opinion in Systems Biology 6: 51–57.

Boer, W. de et al. 2005. Living in a fungal world: impact of fungi on soil bacterial niche development. - FEMS Microbiol. Rev. 29: 795–811.

Bonte, D. et al. 2012. Costs of dispersal. - Biol. Rev. Camb. Philos. Soc. 87: 290–312.

Brandt, M. E. 2013. Filamentous Basidiomycetes in the Clinical Laboratory. - Curr. Fungal Infect. Rep. 7: 219–223.

Chamberlain, M. and Ingram, D. S. 1997. The Balance and Interplay between Asexual and Sexual Reproduction in Fungi. - In: Andrews, J. H. et al. (eds), Advances in Botanical Research. Academic Press, pp. 71–87.

Chan, J. Y. et al. 2019. When to cut your losses: Dispersal allocation in an asexual filamentous fungus in response to competition. - Ecol. Evol. 9: 4129–4137.

Childs, D. Z. et al. 2010. Evolutionary bet-hedging in the real world: empirical evidence and challenges revealed by plants. - Proc. Biol. Sci. 277: 3055–3064.

Clobert, J. et al. 2009. Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations. - Ecol. Lett. 12: 197–209.

Congdon, J. D. et al. 2001. Resource allocation-based life histories: a conceptual basis for studies of ecological toxicology. - Environmental Toxicology and Chemistry: An International Journal 20: 1698– 1703.

63

Crous, P. W. et al. 2014. Phacidium and Ceuthospora (Phacidiaceae) are congeneric: taxonomic and nomenclatural implications. - IMA Fungus 5: 173–193.

Crowther, T. W. et al. 2011. Species-specific effects of grazing invertebrates on mycelial emergence and growth from woody resources into soil. - Fungal Ecol. 4: 333–341.

Delaquis, P. J. et al. 1989. Detachment ofPseudomonas fluorescens from biofilms on glass surfaces in response to nutrient stress. - Microb. Ecol. 18: 199–210.

Doust, J. L. 1989. Plant reproductive strategies and resource allocation. - Trends Ecol. Evol. 4: 230–234.

Edman, M. et al. 2004a. Abundance and viability of fungal spores along a forestry gradient - responses to habitat loss and isolation? - Oikos 104: 35–42.

Edman, M. et al. 2004b. Spore deposition of wood-decaying fungi: importance of landscape composition. - Ecography 27: 103–111.

Finlay, B. J. et al. 1997. Microbial Diversity and Ecosystem Function. - Oikos 80: 209–213.

Finn, J. A. 2001. Ephemeral Resource Patches as Model Systems for Diversity-Function Experiments. - Oikos 92: 363–366.

Fuhrman, J. A. 2009. Microbial community structure and its functional implications. - Nature 459: 193– 199.

Gilchrist, M. A. et al. 2006. Identifying fitness and optimal life-history strategies for an asexual filamentous fungus. - Evolution 60: 970–979.

Heaton, L. et al. 2012. Analysis of fungal networks. - Fungal Biol. Rev. 26: 12–29.

Heaton, L. L. M. et al. 2016. Energetic Constraints on Fungal Growth. - Am. Nat. 187: E27–40.

Heitman, J. 2015. Evolution of sexual reproduction: a view from the Fungal Kingdom supports an evolutionary epoch with sex before sexes. - Fungal Biol. Rev. 29: 108–117.

Hibbing, M. E. et al. 2010. Bacterial competition: surviving and thriving in the microbial jungle. - Nat. Rev. Microbiol. 8: 15–25.

Ho, A. et al. 2017. Revisiting life strategy concepts in environmental microbial ecology. - FEMS Microbiol. Ecol. in press.

64

Johst, K. et al. 2002. Metapopulation persistence in dynamic landscapes: the role of dispersal distance. - Oikos 98: 263–270.

Jonsson, B. et al. 2005. Ecology of species living on dead wood – lessons for dead wood management. - Silva Fenn. in press.

Klein, D. A. and Paschke, M. W. 2004. Filamentous fungi: the indeterminate lifestyle and microbial ecology. - Microb. Ecol. 47: 224–235.

Kozłowski, J. 1992. Optimal allocation of resources to growth and reproduction: Implications for age and size at maturity. - Trends Ecol. Evol. 7: 15–19.

Lambers, H. and Poorter, H. 1992. Inherent Variation in Growth Rate Between Higher Plants: A Search for Physiological Causes and Ecological Consequences. - In: Begon, M. and Fitter, A. H. (eds), Advances in Ecological Research. Academic Press, pp. 187–261.

McDougald, D. et al. 2011. Should we stay or should we go: mechanisms and ecological consequences for biofilm dispersal. - Nat. Rev. Microbiol. 10: 39–50.

Meyers, L. A. and Bull, J. J. 2002. Fighting change with change: adaptive variation in an uncertain world. - Trends Ecol. Evol. 17: 551–557.

Moran, N. A. 1992. The Evolutionary Maintenance of Alternative Phenotypes. - Am. Nat. 139: 971–989.

Muller, E. E. L. et al. 2018. Using metabolic networks to resolve ecological properties of microbiomes. - Current Opinion in Systems Biology 8: 73–80.

Nawrot-Chorabik, K. et al. 2016. Interactions between callus cultures ofPinus sylvestrisand pine fungi with different trophic properties. - For. Pathol. 46: 179–186.

Norros, V. et al. 2012. Dispersal may limit the occurrence of specialist wood decay fungi already at small spatial scales. - Oikos 121: 961–974.

Ojeda-López, M. et al. 2018. Evolution of asexual and sexual reproduction in the aspergilli. - Stud. Mycol. 91: 37–59.

Pringle, A. and Taylor, J. 2002. The fitness of filamentous fungi. - Trends Microbiol. 10: 474–481.

Prosser, J. I. et al. 2007. The role of ecological theory in microbial ecology. - Nat. Rev. Microbiol. 5: 384– 392.

65

Read, D. J. and Stribley, D. P. 1975. Diffusion and translocation in some fungal culture systems. - Trans. Br. Mycol. Soc. 64: 381–388.

Rotheray, T. D. et al. 2008. Grazing alters network architecture during interspecific mycelial interactions. - Fungal Ecol. 1: 124–132.

Sæther, B.-E. and Engen, S. 2015. The concept of fitness in fluctuating environments. - Trends Ecol. Evol. 30: 273–281.

Sawyer, L. K. and Hermanowicz, S. W. 2000. Detachment of Aeromonas hydrophila and Pseudomonas aeruginosa due to variations in nutrient supply. - Water Sci. Technol. 41: 139–145.

Schimel, J. P. and Schaeffer, S. M. 2012. Microbial control over carbon cycling in soil. - Front. Microbiol. 3: 348.

Schleheck, D. et al. 2009. Pseudomonas aeruginosa PAO1 preferentially grows as aggregates in liquid batch cultures and disperses upon starvation. - PLoS One 4: e5513.

Smith, G. R. et al. 2018. Competition–colonization tradeoffs structure fungal diversity. - ISME J. 12: 1758– 1767.

Stearns, S. C. 1992. The evolution of life histories. - sidalc.net.

Stenlid, J. and Gustafsson, M. 2001. Are Rare Wood Decay Fungi Threatened by Inability to Spread? - Ecol. Bull.: 85–91.

Taylor, J. W. et al. 1999. THE EVOLUTION OF ASEXUAL FUNGI: Reproduction, Speciation and Classification. - Annu. Rev. Phytopathol. 37: 197–246.

Therneau, T. 2015. A Package for Survival Analysis in S. version 2.38. in press.

Venable, D. L. 2007. Bet hedging in a guild of desert annuals. - Ecology 88: 1086–1090.

Venable, D. L. and Brown, J. S. 1988. The Selective Interactions of Dispersal, Dormancy, and Seed Size as Adaptations for Reducing Risk in Variable Environments. - Am. Nat. 131: 360–384.

Vincent, T. L. and Pulliam, H. R. 1980. Evolution of life history strategies for an asexual annual plant model. - Theor. Popul. Biol. 17: 215–231.

66

White, T. J. et al. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. - PCR protocols: a guide to methods and applications 18: 315–322.

Wiens, J. A. 1989. Spatial Scaling in Ecology. - Funct. Ecol. 3: 385–397.

Wilbur, H. M. and Rudolf, V. H. W. 2006. Life-history evolution in uncertain environments: bet hedging in time. - Am. Nat. 168: 398–411.

Wilhelm, R. C. et al. 2019. Bacterial contributions to delignification and lignocellulose degradation in forest soils with metagenomic and quantitative stable isotope probing. - ISME J. 13: 413–429.

Wiseman, M. S. et al. 2016. A New Postharvest Fruit Rot in Apple and Pear Caused by Phacidium lacerum. - Plant Dis. 100: 32–39.

Womack, A. M. et al. 2010. Biodiversity and biogeography of the atmosphere. - Philos. Trans. R. Soc. Lond. B Biol. Sci. 365: 3645–3653.

67

2.7 Supplementary Material

Supplementary Table 1. Comparison of substrates included at different malt extract agar media concentrations (g/L of water)

(g/L) 1% 0.5% 0.1% H2O

Agar 10 10 10 10

D glucose 10 5 1 0

Malt Extract 10 5 1 0

Peptone 0.5 0.25 0.05 0

Supplementary Table 2. Analysis of Variance (ANOVA) with nutrient concentration and Petri dish size as treatment effects on colony growth rate. Overall, colony growth rate was significantly different between treatments.

ANOVA DF SS Mean Sq F P Nutrient 2 0.029 0.014 5228.2 <0.0001* Size 1 0.003 0.003 914.8 <0.0001* Nutrient : Size 2 0.001 0.001 109.1 <0.0001* Residuals 29 0.000 0.000

68

Supplementary Table 3. Tukey’s HSD test comparisons of treatment effects on colony growth rate. Colony growth rate was significantly different across all treatments.

Tukey HSD p L0.5-L0.1 <0.0001* L1-L0.1 <0.0001* S0.1-L0.1 <0.0001* S0.5-L0.1 <0.0001* S1-L0.1 <0.0001* L1-L0.5 <0.0001* S0.1-L0.5 <0.0001* S0.5-L0.5 <0.0001* S1-L0.5 <0.0001* S0.1-L1 <0.0001* S0.5-L1 <0.0001* S1-L1 <0.0001* S0.5-S0.1 <0.0001* S1-S0.1 <0.0001* S1-S0.5 <0.0001*

Supplementary Table 4. Analysis of Variance (ANOVA) with nutrient concentration and Petri dish size as treatment effects on coarse resource uptake between treatments at initiation of pycnidia production. Overall, coarse resource uptake was significantly different between colonies.

ANOVA DF SS Mean Sq F P Nutrient 1 0.008 0.014 74.21 <0.0001* Size 1 0.042 0.003 383.21 <0.0001* Nutrient : Size 1 0.001 0.000 1.531 0.23 Residuals 20 0.000 0.000

69

Supplementary Table 5. Tukey’s HSD test comparisons of coarse resource uptake between nutrient concentration and Petri dish size

Tukey HSD p S0.5-L0.5 <0.0005* L1-L0.5 <0.0001* S1-L0.5 <0.0001* L0.5-S0.5 <0.0001* S1-S0.5 <0.0001* S1-L1 <0.0001*

Supplementary Table 6. Analysis of Variance (ANOVA) with nutrient concentration and Petri dish size as treatment effects on pycnidia density. Overall, pycnidial densities were significantly different between treatments.

ANOVA DF SS Mean Sq F P Nutrient 2 11.086 5.543 670.66 <0.0001* Size 1 0.336 0.336 40.65 <0.0001* Nutrient : Size 2 0.421 0.211 25.48 <0.0001* Residuals 29 0.240 0.008

70

Supplementary Table 7. Generalised linear mixed-effects model with pycnidia production as the response variable. The model was fit with a Poisson distribution with Petri dish replicate as a random effect, and treatment and relative growth rate as fixed effects. There is an overall negative correlation with significant interaction between relative growth rate (RGR) and treatment.

Dependent variable: Pycnidia Produced L0.5 3.263*** (0.126) L1 4.427*** (0.126) S0.1 -1.171*** (0.128) S0.5 2.526*** (0.126) S1 2.298*** (0.126) RGR -106.321*** (1.810) L0.5: RGR -34.485*** (1.880) L1: RGR -36.337*** (1.837) S0.1: RGR -84.050*** (3.332) S0.5: RGR -476.980*** (9.705) S1: RGR -434.802*** (6.708) Constant 4.562*** (0.094) Observations 839 Log Likelihood -272,706.400

71

Akaike Inf. Crit. 545,438.800 Bayesian Inf. Crit. 545,500.300 Note* *p<0.1; **p<0.05; ***p<0.01

Supplementary Figure 1. A) Phacidium lacerum pycnidial conidiomata, B) Spores extracted from pycnidia.

Supplementary Figure 2. Visual comparison between pycnidial density between treatments at day 20 on small islands.

72

CHAPTER 3

When to cut your losses: Dispersal allocation in an asexual filamentous fungus in response to competition

Justin Y. Chan, Stephen P. Bonser, Jeff R. Powell, William K. Cornwell

73

3.1 Abstract

1. Fungal communities often form on ephemeral substrates and dispersal is critical for the

persistence of fungi among the islands that form these meta-communities. Within each

substrate, competition for space and resources is vital for the local persistence of fungi. The

capacity to detect and respond by dispersal away from unfavourable conditions may confer

higher fitness in fungi. Informed dispersal theory posits that organisms are predicted to

detect information about their surroundings which may trigger a dispersal response. As

such, we expect that fungi will increase allocation to dispersal in the presence of a strong

competitor.

2. In a lab setting, we tested how competition with other filamentous fungi affected the

development of conidial pycnidiomata (asexual fruiting bodies) in Phacidium lacerum over

ten days.

3. Phacidium lacerum was not observed to produce more asexual fruiting bodies or produce

them earlier when experiencing interspecific competition with other filamentous fungi.

However, we found that a trade-off existed between growth-rate and allocation to dispersal.

We also observed a defensive response to specific interspecific competitors in the form of

hyphal melanisation of the colony which may have an impact on the growth rate and

dispersal trade-off.

4. Our results suggest that P. lacerum have the capacity to detect and respond to competitors

by changing their allocation to dispersal and growth. However, allocation to defence may

come at a cost to growth and dispersal. Thus, it is likely that optimal life history allocation in

fungi constrained to ephemeral resources will depend on the competitive strength of

neighbours surrounding them

74

3.2 Introduction

Dispersal is a fundamental part of the life history of an organism. Effective dispersal can be an adaptive response to declining resource levels (Abrams 2000, Holt 2008) or increasing competition

(Martorell and Martínez-López 2014). Dispersal can also be achieved through the production of reproductive propagules, a mode of dispersal vital to sessile organisms (Kinlan and Gaines 2003).

There are dispersal events seen in most species, and this phenomenon drives many spatial dynamics seen in the environment (Spiegel and Nathan 2010). Dispersal, patch dynamics and species interactions together occur within communities and shape the colonisation and extinction of species within a local patch (Hanski 1998, Leibold et al. 2004, Lancaster and Downes 2017). Thus, dispersal influences the ability of individual species to persist within a landscape (Johst et al. 2002). For most organisms, the ability to disperse is integral to persist in a fundamentally dynamic and often competitive environment (Edman et al. 2004). However, the drivers that shape patterns in dispersal remain unclear for many groups of organisms, particularly for microorganisms such as fungi (Cadotte et al. 2006, Fuhrman 2009, Nemergut et al. 2013, Lancaster and Downes 2017).

Informed dispersal theory posits that organisms integrate information about their internal condition and external environment, with the dispersal option weighed against the costs of remaining in the current environment and triggered when costs reach a certain level (Clobert et al.

2009). The increase in allocation to dispersal in plants can be influenced by environmental stress

(Martorell and Martínez-López 2014) and competitive conditions (French et al. 2017, Tabassum and

Bonser 2017), where these adversities induce higher allocation to reproduction. However, these results may not be directly applicable to understanding the drivers of dispersal in microbial systems.

For instance, in many situations, key resources in plants do not deplete over the course of an individual’s lifespan (e.g., light) or are accessed by organs exploring new patches (e.g., nutrients). In contrast, for many microbial systems, the fundamental resources for growth (such as carbon, nitrogen, phosphorus) deplete at the scale of the spatial extent and lifespan of an organism. It is not

75 clear the extent to which predictions from plant systems may be extrapolated to much smaller, heterotrophic organisms like fungi.

Competition for resources is common in saprotrophic fungi (Boddy 2000). Fungi primarily compete over space, and the control of resources within territories (Hiscox et al. 2015a). Access to resources held by other fungi often requires combative antagonism to breach an already occupied space. Variation in competitive ability is common, with distinct hierarchies formed by interactions bound in a substrate (Boddy 2000, Stenlid and Gustafsson 2001, van der Wal et al. 2015, Haňáčková et al. 2015). Thus, delaying dispersal through delaying reproduction may result in low fitness if a fungal colony cannot withstand the combative assault of a competitively superior fungus. This presents a scenario where the benefits of remaining stationary diminish over time as resources become depleted, and the costs of defending a territory overtake the costs of dispersal (Clobert et al. 2009). Increased allocation to dispersal in response to competition has been demonstrated in plants (Bonser 2013, Fazlioglu et al. 2016), and this response is associated with a strategy of escaping intense competition, but whether fungi increase allocation to dispersal in response to competition is still unknown.

Fungi are similar to plants in that they are a modular organism with the capacity to respond to stimuli at the growing point of each hyphal tube (Lee et al. 2016). While little is known about the shift to dispersal, fungi have the ability to dynamically allocate resources and translocate nutrients through their mycelia(Tlalka et al. 2008, Philpott et al. 2014). Fungi tend to be highly responsive to environmental factors, altering their colony growth form in response to resource pools (Heaton et al.

2012), fungivore attack (Crowther et al. 2011), and interspecific mycelial interaction (Rotheray et al.

2008). We tested if fungi are able to alter allocation of resources to dispersal as a response to a decrease in environment quality or increasing competition as a strategy to persist as a sessile organism in an unstable environment.

76

Here, we take advantage of a system for studying allocation to air-borne dispersal using an asexual filamentous fungus, Phacidium lacerum. Phacidium lacerum, an ascomycetous pathogen and filamentous fungus of the order Helotiales, is known to infect pine trees (Nawrot-Chorabik et al.

2016) and Rosaceae fruits (Wiseman et al. 2016) with no previously documented presence in decaying wood. The primary form of dispersal is through the production of ascomata or pycnidial conidiomata (SFig. 1) emerging from the surface of infected plant material, and the subsequent release of ascospores or asexual conidia (Crous et al. 2014). It was chosen as a suitable species as it is capable of developing pycnidia in vitro in a Petri dish on growth media. This property allows us to quantify the density of pycnidia over colony area as a proxy of allocation to air-borne dispersal, as such, it is the focal species of this study. With this new model system for fungal allocation to air- borne dispersal, we can quantify allocation to dispersal with and without a competitor present and begin to connect dispersal to an emerging understanding of fungal life histories. We predict that: 1) competition will induce early reproduction in a fungal colony, and 2) competition will induce high allocation to reproduction in a fungal colony.

3.3 Materials and methods

3.3.1 Study Species

We extracted isolates of Phacidium lacerum (Isolate Face008) from rotting Eucalyptus tereticornis logs in Richmond, NSW (33°37'04.0"S 150°44'25.3"E) in February 2016. The site is in remnant Cumberland Plain woodland dominated by Eucalyptus tereticornis. The isolates were collected by splitting the wood with a sterilised chisel and extracting wood chips from the centre of the logs. We placed the wood chips on 2% malt extract agar (MEA) and subcultured from the emerging hyphae until we attained a pure culture. We extracted DNA from the growing hyphae from fungal cultures using DNeasy Plant Mini Kit (Qiagen, Chadstone, Victoria, Australia) as per the manufacturer’s instruction. We amplified the ITS (ITS1F & ITS4) region of rDNA (Thompson et al.

77

2012) through PCR amplification and analysed the amplicons using a ABI3500 Genetic Analyser

(Applied Biosystems, Life Technologies, Mulgrave, Victoria, Australia). The species identity was confirmed by conducting a BLAST search against the NCBI Nucleotide database.

3.3.2 Isolating Competitors

To test whether P. lacerum responds to competitive interactions, we selected three basidiomycete fungi capable of degrading wood. A Phanerochaete sp. (Face061) was extracted from the same wood blocks as P. lacerum. We obtained two more fungal isolates from existing collections in the

Hawkesbury Institute of the Environment (Western Sydney University, Richmond), Omphalotus sp.

(MT5A) and an unidentified cord-forming basidiomycete (HWK05). We selected these species as they are likely able to impose competitive pressures on P. lacerum over the course of the experiment as they tested positive with a lignin-guaiacol test, indicating that they were able to produce oxidative enzymes (Hiscox & Boddy, 2017). All fungal isolates were maintained on 2% MEA at 4oC (see Fig. 4 for colony morphology).

3.3.3 Experimental Design

To test whether a fungus alters dispersal allocation as a result of competitive interactions, we set up multiple pairwise interspecific and intraspecific interactions with the focal species P. lacerum on Petri dishes and growing on 2% MEA. Our experiment consisted of five treatments with

P. lacerum: 1) P. lacerum alone, 2) P. lacerum in intraspecific competition with a genetic clone of itself, 3) the focal species in interspecific competition with the Phanerochaete sp. isolate, 4)

Omphalotus sp. isolate, or 5) with the isolate HWK5. We had eight replicates of each treatment with a total of 40 treatment plates.

78

Prior to the experiment, all species were inoculated onto H2O agar (10mg/L) to normalise

hyphal density between species. The cultures were allowed to establish on the H2O agar for 7 days before 5mm diameter plugs were taken from the growing edge of the colonies and inoculated onto

Petri dishes (9 cm diameter) with 2% MEA. Focal species in the alone treatment were inoculated onto the centre of the petri dish. For each of the competition treatments, P. lacerum was inoculated

2.25 cm away from the edge of the Petri dish, while the competitor was inoculated 2.25 cm from the centre, opposite to the P. lacerum inoculation such that both inocula had equal area to develop (see

Fig. 4). All Petri dishes were incubated in the dark at 25oC.

3.3.4 Image Analysis

We assessed the Petri dishes every day for colony growth and measured colony growth rate every day by tracking colony size. We measured colony growth rate by measuring the log difference in colony area every 24 hours to calculate relative growth rate (Lambers and Poorter 1992). After the first emergence of pycnidia, two days following inoculation, we conducted a census of the number and location of pycnidia every 12 hours by marking the position of pycnidia on the back of the Petri dish with an acetate sheet and black permanent marker. For each experimental plate, we measured the number of pycnidia, and colony radius to estimate reproductive allocation as number of pycnidia relative to colony size to avoid any artifacts of sampling pycnidial density. This allowed us to obtain a density measure to account for differences in colony size as a result of interaction. An image of this sheet was captured after every time point using a Canoscan LiDE 210 scanner (300dpi resolution) to track accumulation of pycnidia over the course of the experiment. We monitored pycnidial formation over 8 days (180 hours), which was the time taken for pycnidial development to reach the edge of the colony when the focal species grown alone. This way, we could compare both the timing and the extent of dispersal allocation across the experimental treatments against a colony growing alone. The experiment ran for a total of 10 days.

79

We processed the scans of the pycnidia after every sampling period using ImageJ (NIH, USA).

We manually selected for every black dot using colour thresholding in the RGB colour space, thus creating a binary image. We then used the particle analysis package in ImageJ to automate counts of pycnidia from the binary image. We specified a minimum size of 20-pixel units to discount any spurious marks.

3.3.5 Statistical Analysis

To test the hypothesis of a shift in the timing of allocation to dispersal in response to competition, we fit a survival analysis model (also called a time-to-event model) to data for when P. lacerum produced at least 100 pycnidia. This model estimated the time to at least 100 pycnidia, along with a treatment effect for this variable and a significance test for the treatment effect. Analysis was carried out in the survival package (Therneau 2015) in R (v.3.4.2). To test if P. lacerum produce more pycnidia as a response to competition, we ran a one-way analysis of variance

(ANOVA) with treatment as the competition block in R (v.3.4.2) for final pycnidial density at hour

180. Tukey’s HSD post hoc tests were used to assess significant differences between treatments. To estimate the slopes of the trade-off between growth and dispersal for the different treatments, we fit a longitudinal two-level mixed model in which we have repeated measures of the same petri dish

(see (Diggle et al. 2002). The goal of the model was to test for differences in the relationship between dispersal allocation and RGR We tested this model both with and without first order temporal correlation structure, and results were very similar. Results are reported for the simpler model without correlation structure. RGR values were transformed to z-scores to facilitate model convergence. The number of pycnidia was the response variable and model fitting was done with a

Poisson error term via penalised quasi-likelihood in R (v.3.4.2) within the MASS package (Venables and Ripley 2013). Reference level for the model was the control treatment and post-hoc significance

80 testing for differences in intercepts and slopes were done via summary.glmmPQL function in the

MASS library (Venables and Ripley 2013).

3.4 Results

3.4.1 Colony growth and contact with competitors

The hyphal front of the colonies of P. lacerum came into first contact earliest in the intraspecific competition treatment at two days (Fig. 1). This was followed by the focal species in interspecific treatments at three (Omphalotus sp. and Isolate HWK5) and four days (Phanerochaete sp.) for colony contact. When P. lacerum was grown alone, it took five days on average for the hyphal front of the colony to reach the edge of the Petri dish from the inoculum in the centre of the

Petri dish.

We observed different competitive outcomes with the focal species and its competitors. By the end of the experiment, the P. lacerum colonies grew larger than Phanerochaete sp. colonies. The hyphal front of the colonies of P. lacerum did not breach into the space held by the neighbouring P. lacerum colony during intraspecific competition. We did not observe colonies of P. lacerum breaching into the space of the neighbouring colony during interspecific competition with Isolate

HWK5. Phacidium lacerum was reduced in size when in competition with Omphalotus sp., with

Omphalotus sp. growing and extending into the space held by the colony of P. lacerum (Fig. 4).

Phacidium lacerum responded to colony contact with Isolate HWK5 and Omphalotus sp. with a conspicuous pigmentation of the leading hyphal front at the point of contact (Fig 4).

81

Figure 1. Spline ± SE of pycnidia density in mm2 over 7 days. P. lacerum came into contact with other colonies of P. lacerum at 48 hours. Omphalotus sp. and Isolate HWK5 came into contact with P. lacerum at 72 hours. Phanerochaete sp. came into contact with P. lacerum at 96 hours. P. lacerum growing alone came into contact with the Petri dish edge at 120 hours.

82

3.4.2 Allocation to dispersal as a response to competition

On average, pycnidia appeared at 67 (± 1.05 SE) hours across all treatments (Fig. 1), with no significant difference in timing of emergence of pycnidia as a response to competitor presence or identity (survival analysis, P>0.05 for treatment). Pycnidial density varied among treatments at the end of the experiment (180 Hours). Overall, we found a significant difference in pycnidial density

between treatments (Fig. 2, ANOVA, F4,42= 40.95, P<0.001). Post hoc Tukey’s HSD tests showed that P. lacerum had significantly reduced pycnidial density when experiencing interspecific competition with the unidentified basidiomycete (Tukey’s HSD, P<0.001), Phanerochaete sp. (Tukey’s HSD, P<0.01) and Omphalotus sp. (Tukey’s HSD, P<0.001). We did not observe a reduced allocation to dispersal when P. lacerum was in intraspecific competition with a genetic clone of itself, with no significant difference in the density of pycnidia in the colony (Tukey’s HSD, P=0.96).

83

Figure 2. Pycnidial density per mm2 among treatments at 180 hours. Significant differences in pycnidial density denoted by letters determined by Tukey’s post-hoc tests. Pycnidial density in P. lacerum colonies are significantly lower when in interspecific competition than P. lacerum colonies in intraspecific competition or in the absence of interaction.

84

3.4.3 Trade-offs in allocation

We observed a negative relationship between growth rate and dispersal allocation for all treatments (Table 1). We found that there was no significant difference in the slope of this relationship between when P. lacerum was grown alone versus when P. lacerum was experiencing either intraspecific competition (P=0.852, GLM), and interspecific competition with Omphalotus sp.

(P=0.058, GLM) or interspecific competition with Isolate HWK5 (P=0.308, GLM). However, the slope of the relationship was significantly steeper when in competition with Phanerochaete sp. (Fig. 3,

P<0.0001, GLM).

Table 1. Longitudinal two-level mixed model with repeated measures of individual Petri dishes to test for differences in the relationship between dispersal allocation and RGR. Reference level for the model was the control treatment with the focal species alone.

Main Effects Coefficient SE P DF Alone (Reference) 3.694 0.228 <0.0001 393 Self -0.33 0.116 0.007 42 Phanerochaete -0.359 0.121 0.005 42 Omphalotus -0.721 0.13 <0.0001 42 HWK 5 -0.512 0.124 0.0002 42

Interactions Coefficient SE P DF Alone (Reference) -0.151 0.063 0.018 393 Self -0.019 0.105 0.852 393 Phanerochaete -0.559 0.132 <0.0001 393 Omphalotus 0.234 0.123 0.058 393 HWK 5 0.119 0.116 0.308 393

85

Figure 3. Comparison of relationship between pycnidia development per 12 hours for P. lacerum against relative growth rate. Note that time runs opposite to relative growth rate (moving left). There were no significant differences in the slope of this relationship between when P. lacerum was grown alone versus when P. lacerum was in intraspecific competition, or in competition with

Omphalotus sp., or in competition with Isolate HWK5. The slope was significantly steeper when P. lacerum was in competition with Phanerochaete sp.

86

Figure 4. Phacidium lacerum interactions at 180 hours: A) Alone, B) Intraspecific competition (note the lack of colony fusion), C) Phanerochaete sp., D) Omphalotus sp., E) Isolate

HWK5. Note the melanisation of the hyphal front of P. lacerum (left) in D, E.

87

3.5 Discussion

Allocation to reproductive dispersal was not observed to increase under competition. Daily censuses also did not detect an earlier shift to allocation in response to interspecific competition.

This is contrary to how annual plants respond to competition (Martorell and Martínez-López 2014,

Fazlioglu et al. 2016). The production of pycnidia began across treatments prior to colony contact, with no significant difference in production even when P. lacerum was grown alone (Fig. 1). This suggests that the timing of allocation to dispersal begins prior to interaction and that allocation to dispersal occurred even in the absence of competition (Fig. 1). We also did not find support for our predictions that fungi will allocate relatively more to reproductive dispersal in response to competition. In contrast, we found support for the opposite effect, where interspecific competition drove down total allocation to reproductive dispersal (Fig. 2). This relationship is contrary to our hypothesis that competition will trigger an allocation to dispersal as an escape from adverse conditions (Clobert et al. 2009). There was no change to allocation to reproductive dispersal when P. lacerum was experiencing intraspecific competition with a genetic clone of itself (Fig. 2). The difference in response to intraspecific and interspecific interaction suggests that P. lacerum is able to integrate information about its neighbours and respond accordingly (Boddy 2000, Heilmann-Clausen and Boddy 2005). Competitor-dependent differences in the expression of reproductive strategies is consistent with predictions from informed dispersal theory where an organism is able to assess its local environment (Clobert et al. 2009). However, we did not observe the increased allocation to reproductive dispersal as a response to competition that is expected under this theory, likely due to the trade-off of the cost of competition against the allocation to reproductive dispersal.

88

3.5.1 Trade-offs in allocation

We observed a trade-off between growth and pycnidia production in P. lacerum (Fig. 3), where increasing pycnidia production came at the cost of colony growth. Trade-offs occur when it is not possible for evolution to optimise two functions simultaneously (Stearns 1992). In this system, P. lacerum has an initial high allocation to growth rate prior to a switch to allocation to reproductive dispersal (Moving left on Fig. 3). Gilchrist et al. (2006) present an evolutionary optimality model of this behaviour. In the Gilchrist model, increasing mycelial density as a result of fungal growth results in a decrease in local resources. When growth reaches a point where resources drop below a certain level, a switch to allocation to dispersal maximises spore production for a given mycelial density

(Gilchrist et al. 2006). The fact that we observed this in our study system suggests that P. lacerum may have a high initial allocation to growth to maximise the capture of territory, and thus maximise resource uptake before switching to a reproductive dispersal strategy to locate more resource patches (Heaton et al. 2016). However, when faced with competition, allocation purely to dispersal may leave a fungus with insufficient resources to defend itself against the attack from a neighbouring fungus.

In our study, P. lacerum had similar slopes for the relationships for trade-offs in allocation to dispersal and growth (Fig. 3) when grown alone and when experiencing both intraspecific and interspecific competition apart from the Phanerochaete sp. treatment. However, P. lacerum responded to interaction with Omphalotus sp. and Isolate HWK5 with a reduced allocation to dispersal (Fig. 2), suggesting that total allocation to dispersal shifts depending on the identity of the competitor. When P. lacerum was in competition with Phanerochaete sp., it had a higher slope for the trade-off between pycnidia and growth (Fig. 3) but had fewer total pycnidia (Fig. 2), likely owing to the reduced size of the Phanerochaete sp. colony (Fig. 4c). In line with the Gilchrist model, P. lacerum colonies could continue growing before occupying all available space and switching to dispersal to maximise the production of spores. On the other hand, when experiencing competition

89 with Omphalotus sp. and Isolate HWK5, P. lacerum colonies responded with prominent pigmentation of the leading hyphal front in contact with the neighbouring colony (Fig. 4d, e).

Pigmentation of hyphae (predominantly with melanin) is a response commonly associated with defence against oxidative attack (Butler and Day 1998, Butler et al. 2009, Hiscox et al. 2010).

The absence of this defensive response in the focal species when interacting with Phanerochaete sp. and a genetic clone of itself suggests that this defensive response is triggered only in the presence of certain competitors. The expression of defences in response to external stimuli is a phenomenon seen in plants known as induced defence (Agrawal 1999). The expression of these defences in the absence of attack is costly but prove to be beneficial for the persistence of an individual experiencing herbivory or parasitism (Agrawal 1999, 2001). In a fungal context, the expression of defence in a colony is in response to the presence of another fungal colony rather than direct mycophagy or parasitism. But the melanisation of hyphae against fungal competition achieves a similar goal.

Without the pressure of combative antagonism by a neighbouring colony, a fungus is unlikely to melanise their hyphae as a defensive response. Allocation purely to reproductive dispersal in the face of competition will likely result in colony demise prior to successful dispersal if a colony has not adequately defended itself from attack by a neighbour. Conversely, expending resources to upregulate defence in the absence of a competitor will reduce resources available for allocation to reproduction (Agrawal 1999, Heaton et al. 2016). Thus, allocation to defence against attack by neighbouring fungal colonies may trade off with allocation to dispersal or growth (Siletti et al. 2017).

Our research tends to support this alternative explanation of our observations, but further experimentation is needed to ascertain the true effects of competition on the observed trade-off between growth and dispersal.

The exclusion of competitors by allocation to defence may allow a fungus to exhaust present resources confined within a territory and persist within the local patch, before switching strategies and dispersing to find new accessible resource patches. Allocation to growth or defence to the

90 deficit of dispersal is not a viable strategy due to the dynamic nature of resource patches, and eventually most fungi must allocate to dispersal or risk local extinction (Thomas 1994). Thus, the allocation in reproduction and dispersal in fungi will likely be context dependent and may be optimised when considered under a game theoretic framework (Kozłowski 1992), with shifts in allocation depending on the quality of a present patch. Due to the limitations of the present study, how allocation to defence affects the allocation to dispersal should be examined in greater detail in future studies.

3.6 Conclusion

Microbial metacommunities are receiving a large amount of recent attention (Peay et al.

2007, Prosser et al. 2007, Fuhrman 2009, Fukami et al. 2010, Nemergut et al. 2013, van der Wal et al. 2015, Hiscox et al. 2015b, Maynard et al. 2018), and dispersal is increasingly identified as a key unknown with respect to the maintenance and continuity of these communities in a dynamic landscape (Kneitel and Miller 2003, Lancaster and Downes 2017, Smith et al. 2018, Calhim et al.

2018, Davison et al. 2018). We have demonstrated that fungi can acquire information from the environment, subsequently affecting changes to their allocation strategies to dispersal when faced with competition within the confines of the trade-off between growth and dispersal. Understanding how decisions to disperse are weighed against (meta-) community interactions will be a major step forward in predicting the development of these communities over time. The interactions within a community bound within an ephemeral resource island will shape the progression of species

(Maynard et al. 2018), and dispersal will be the bridge that connects the stochastically formed islands in the environment. With allocation to dispersal being such a key component of fungal life history, bridging our understanding from life history theory (e.g. Gilchrist et al. 2006, Heaton et al.

2016) to microbial communities will help us begin to understand the complex patterns in community assembly we are observing.

91

3.7 References

Abrams, P. A. 2000. The impact of habitat selection on the spatial heterogeneity of resources in varying environments. - Ecology 81: 2902–2913.

Agrawal, A. A. 1999. Induced plant defense: evolution of induction and adaptive phenotypic plasticity. - Inducible plant defenses against pathogens and herbivores: biochemistry, ecology, and agriculture. American Phytopathological Society Press, St. Paul, MN: 251–268.

Agrawal, A. A. 2001. Phenotypic plasticity in the interactions and evolution of species. - Science 294: 321– 326.

Boddy, L. 2000. Interspecific combative interactions between wood-decaying basidiomycetes. - FEMS Microbiol. Ecol. 31: 185–194.

Bonser, S. P. 2013. High reproductive efficiency as an adaptive strategy in competitive environments. - Funct. Ecol. 27: 876–885.

Butler, M. J. and Day, A. W. 1998. Fungal melanins: a review. - Can. J. Microbiol. 44: 1115–1136.

Butler, M. J. et al. 2009. Melanin synthesis by Sclerotinia sclerotiorum. - Mycologia 101: 296–304.

Cadotte, M. W. et al. 2006. The effects of resource enrichment, dispersal, and predation on local and metacommunity structure. - Oecologia 149: 150–157.

Calhim, S. et al. 2018. Fungal spore diversity reflects substrate-specific deposition challenges. - Sci. Rep. 8: 5356.

Clobert, J. et al. 2009. Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations. - Ecol. Lett. 12: 197–209.

Crous, P. W. et al. 2014. Phacidium and Ceuthospora (Phacidiaceae) are congeneric: taxonomic and nomenclatural implications. - IMA Fungus 5: 173–193.

Crowther, T. W. et al. 2011. Species-specific effects of grazing invertebrates on mycelial emergence and growth from woody resources into soil. - Fungal Ecol. 4: 333–341.

Davison, J. et al. 2018. Microbial island biogeography: isolation shapes the life history characteristics but not diversity of root-symbiotic fungal communities. - ISME J. 12: 2211–2224.

92

Diggle, P. et al. 2002. Analysis of Longitudinal Data. - OUP Oxford.

Edman, M. et al. 2004. Spore deposition of wood-decaying fungi: importance of landscape composition. - Ecography 27: 103–111.

Fazlioglu, F. et al. 2016. Reproductive efficiency and shade avoidance plasticity under simulated competition. - Ecol. Evol. 6: 4947–4957.

French, K. et al. 2017. Facilitation, competition and parasitic facilitation amongst invasive and native liana seedlings and a native tree seedling.: 17.

Fuhrman, J. A. 2009. Microbial community structure and its functional implications. - Nature 459: 193– 199.

Fukami, T. et al. 2010. Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. - Ecol. Lett. 13: 675–684.

Gilchrist, M. A. et al. 2006. Identifying fitness and optimal life-history strategies for an asexual filamentous fungus. - Evolution 60: 970–979.

Haňáčková, Z. et al. 2015. Fungal succession in the needle litter of a montane Picea abies forest investigated through strain isolation and molecular fingerprinting. - Fungal Ecol. 13: 157–166.

Hanski, I. 1998. Metapopulation dynamics. - Nature 396: 41–49.

Heaton, L. et al. 2012. Analysis of fungal networks. - Fungal Biol. Rev. 26: 12–29.

Heaton, L. L. M. et al. 2016. Energetic Constraints on Fungal Growth. - Am. Nat. 187: E27–40.

Heilmann-Clausen, J. and Boddy, L. 2005. Inhibition and stimulation effects in communities of wood decay fungi: exudates from colonized wood influence growth by other species. - Microb. Ecol. 49: 399–406.

Hiscox, J. et al. 2010. Changes in oxidative enzyme activity during interspecific mycelial interactions involving the white-rot fungus Trametes versicolor. - Fungal Genet. Biol. 47: 562–571.

Hiscox, J. et al. 2015a. Antagonistic fungal interactions influence carbon dioxide evolution from decomposing wood. - Fungal Ecol. 14: 24–32.

Hiscox, J. et al. 2015b. Priority effects during fungal community establishment in beech wood. - ISME J. 9: 2246–2260.

93

Holt, R. D. 2008. Theoretical perspectives on resource pulses. - Ecology 89: 671–681.

Johst, K. et al. 2002. Metapopulation persistence in dynamic landscapes: the role of dispersal distance. - Oikos 98: 263–270.

Kinlan, B. P. and Gaines, S. D. 2003. Propagule dispersal in marine and terrestrial environments: A community perspective. - Ecology 84: 2007–2020.

Kneitel, J. M. and Miller, T. E. 2003. Dispersal rates affect species composition in metacommunities of Sarracenia purpurea inquilines. - Am. Nat. 162: 165–171.

Kozłowski, J. 1992. Optimal allocation of resources to growth and reproduction: Implications for age and size at maturity. - Trends Ecol. Evol. 7: 15–19.

Lambers, H. and Poorter, H. 1992. Inherent Variation in Growth Rate Between Higher Plants: A Search for Physiological Causes and Ecological Consequences. - In: Begon, M. and Fitter, A. H. (eds), Advances in Ecological Research. Academic Press, pp. 187–261.

Lancaster, J. and Downes, B. J. 2017. A landscape-scale field experiment reveals the importance of dispersal in a resource-limited metacommunity. - Ecology 98: 565–575.

Lee, S. H. et al. 2016. Mesoscale analyses of fungal networks as an approach for quantifying phenotypic traits. - J Complex Netw 5: 145–159.

Leibold, M. A. et al. 2004. The metacommunity concept: a framework for multi-scale community ecology. - Ecol. Lett. 7: 601–613.

Martorell, C. and Martínez-López, M. 2014. Informed dispersal in plants: Heterosperma pinnatum (Asteraceae) adjusts its dispersal mode to escape from competition and water stress. - Oikos 123: 225–231.

Maynard, D. S. et al. 2018. Species associations overwhelm abiotic conditions to dictate the structure and function of wood-decay fungal communities. - Ecology 99: 801–811.

Nawrot-Chorabik, K. et al. 2016. Interactions between callus cultures of Pinus sylvestris and pine fungi with different trophic properties. - For. Pathol. 46: 179–186.

Nemergut, D. R. et al. 2013. Patterns and processes of microbial community assembly. - Microbiol. Mol. Biol. Rev. 77: 342–356.

94

Peay, K. G. et al. 2007. A strong species--area relationship for eukaryotic soil microbes: island size matters for ectomycorrhizal fungi. - Ecol. Lett. 10: 470–480.

Philpott, T. J. et al. 2014. Nitrogen translocation and accumulation by a cord-forming fungus (Hypholoma fasciculare) into simulated woody debris. - For. Ecol. Manage. 315: 121–128.

Prosser, J. I. et al. 2007. The role of ecological theory in microbial ecology. - Nat. Rev. Microbiol. 5: 384– 392.

Rotheray, T. D. et al. 2008. Grazing alters network architecture during interspecific mycelial interactions. - Fungal Ecol. 1: 124–132.

Siletti, C. E. et al. 2017. Distributions of fungal melanin across species and soils. - Soil Biol. Biochem. 113: 285–293.

Smith, G. R. et al. 2018. Competition–colonization tradeoffs structure fungal diversity. - ISME J. 12: 1758– 1767.

Spiegel, O. and Nathan, R. 2010. Incorporating density dependence into the directed-dispersal hypothesis. - Ecology 91: 1538–1548.

Stearns, S. C. 1992. The evolution of life histories. - sidalc.net.

Stenlid, J. and Gustafsson, M. 2001. Are Rare Wood Decay Fungi Threatened by Inability to Spread? - Ecol. Bull.: 85–91.

Tabassum, S. and Bonser, S. P. 2017. Allometry in the terminal velocity--dispersal architecture relationship explains variation in dispersal and offspring provisioning strategies in wind dispersed Asteraceae species. - Aust. J. Bot. 65: 149–156.

Therneau, T. 2015. A Package for Survival Analysis in S. version 2.38. in press.

Thomas, C. D. 1994. Extinction, colonization, and metapopulations: Environmental tracking by rare species. - Conserv. Biol. 8: 373–378.

Thompson, T. A. et al. 2012. Hypholoma lateritium isolated from coarse woody debris, the forest floor, and mineral soil in a deciduous forest in New Hampshire. - Botany 90: 457–464.

Tlalka, M. et al. 2008. Quantifying dynamic resource allocation illuminates foraging strategy in Phanerochaete velutina. - Fungal Genet. Biol. 45: 1111–1121.

95 van der Wal, A. et al. 2015. Neglected role of fungal community composition in explaining variation in wood decay rates.

Venables, W. N. and Ripley, B. D. 2013. Modern Applied Statistics with S-PLUS. - Springer Science & Business Media.

Wiseman, M. S. et al. 2016. A New Postharvest Fruit Rot in Apple and Pear Caused by Phacidium lacerum. - Plant Dis. 100: 32–39.

96

3.8 Supplementary Material

Supplementary Figure 1. A) Phacidium lacerum (Face008) pycnidial conidiomata, B) Spores extracted from pycnidia.

97

CHAPTER 4

Complex environments alter competitive dynamics in fungi

Justin Y. Chan, Stephen P. Bonser, Michael M. Kasumovic, Jeff R. Powell,

William K. Cornwell

98

4.1 Abstract

Competition is a key biotic factor that often structures natural communities. Many attempts to disentangle how competition shapes natural communities have relied on experiments on simplified systems or through simple mathematical models. But these simplified approaches are limited in their ability to represent the complexity seen in more natural settings. Here, we considered the competitive pairwise dynamics between four saprotrophic fungal species. We tested whether the contextual environment changed these dynamics, repeating competitive experiments in a simple agar media and a more ecologically realistic wood block setting. We found that the competitive outcomes on agar media differed from those within the wood blocks. While superior competitors were identified across all pairwise interactions on agar, within the wood blocks, two of six interactions resulted in deadlock, where neither competitor could breach territory of the other, and one interaction resulted in a reversed competitive outcome. These results suggest that the complexity within natural substrates can alter the strength of interspecific interactions and may contribute to coexistence and the resulting high diversity of fungi often observed within wood.

99

4.2 Introduction

“...Not that under nature the relations will ever be as simple as this. Battle within battle must be continually recurring with varying success; and yet in the long-run the forces are so nicely balanced that the face of nature remains for long periods of time uniform, though assuredly the merest trifle would give the victory to one organic being over another.” - Charles Darwin, The Origin of Species (1859)

The struggle for existence among co-occurring species is a question central to our fundamental understanding of life. What conditions allow one species to outcompete another? This question has spawned multiple domains of research in a bid to dissect the underlying elements that contribute to winning the struggle for existence. Theory developed from simplified systems have provided valuable insight into how competitive dynamics result in coexistence (Hutchinson 1961,

Leibold 1995, Sommer 1999). One theory is that the ability to harness and incorporate resources will affect competitive outcomes between species, where different rates of resource intake of a limiting resource can successfully predict the outcome of competition (Tilman et al. 1982, Wilson and Tilman

1993). Many mathematical models have explored this idea and have been used to estimate competitive dynamics to predict which species will emerge victorious during interspecies interactions (Levins and Culver 1971, Slatkin 1974). This approach has provided us insight to the mechanisms that may shape the outcome of the struggle for existence (Roughgarden 1983), and thus seek to inform the key question of why so many species are observed in close proximity despite predictions that competitive dominance drive down diversity when species compete for the same resources.

Many of these models and ideas stem from earlier papers by Gause in which he conducted several seminal experiments exploring the nature of competition between microbes (Gause 1932,

1934, Gause et al. 1934). With these microbial systems, Gause laid the foundation for mathematical

100 models of population growth (Lotka 1926, Volterra 1926) as well as the principle of competitive exclusion (Hardin 1960). But equally interesting research by Gause exploring the role of habitat complexity have been largely ignored. Habitat complexity has the potential to have important implications to our understanding of inter-species interactions because different species may have traits that provide them competitive benefits in more complex environments (Huffaker 1958). For example, research exploring predator-prey interactions between Paramecium caudatum and

Didinium nasutum demonstrate that, in homogeneous environments, D. nasutum was able to completely exhaust the population of P. caudatum. However, heterogeneous environments promoted the persistence of P. caudatum as they hid in refugia, resulting in D. nasutum perishing due to the lack of access to food, an observation echoed in another study using a microbial system

(Kerr et al. 2002). These results strongly suggest that environmental complexity plays a role in mediating biotic interactions that manifest in experimental systems and may play a part in attenuating the competitive hierarchies found in other simple systems.

In mathematical models, it is assumed that the progression of populations follows a predictable pattern as a result of the interaction between species (Wangersky 1978). But as complexity is introduced to experimental systems to more closely mimic “field” situations, the influence of the environment can shape the trajectory of population growth in unpredictable ways

(Gause 1934, Levin 1976). The difficulty in accounting for heterogeneous elements limits the ability of models to predict the variable outcomes of competition in complex environments (Simberloff

1982). But to understand whether Gause’s results on the role of complexity for species coexistence have further implications than simply predator-prey interactions, and broader implications on species-coexistence theories, we have to explore whether habitat complexity can also affect interactions between competitors. To further develop our understanding of species interactions in complex environments, we use an empirical approach to test how changes in environmental complexity can affect the strength of competitive interactions. Here, we utilise a simple fungal experimental system to test how habitat complexity can affect the outcome of competition between

101 wood decay fungi. To explore this idea, we used four different species of wood decay fungi to explore the relative competitive ability of each fungus against different competitors in complex and simple environments.

Wood decay fungi are potent saprotrophs that derive their nutrition from dead organic matter (Blanchette 1991, Boddy 2001). As a hyperdiverse group of organisms, there is potentially an extremely high diversity of interacting species within any given substrate (Mäkipää et al. 2017). The theory of competitive exclusion predicts that under these conditions, a superior competitor should dominate and exclude inferior competitors over a common resource (Hardin 1960, Holmer and

Stenlid 1997). With saprotrophic fungi as a model system, we can examine the differences in ranking in competitive ability (or competitive hierarchies) between simple and complex substrates. We established pairwise interspecific interactions between four wood decay fungal species on nutrient media as a simple resource, and then repeated this on paired wood blocks as a complex, more ecologically realistic setting. From the outcomes of the pairwise interactions, we compared the competitive hierarchy between fungal interactions on nutrient media and wood blocks following

(Hiscox et al. 2015). We used these two extreme competitive environments as our goal was to identify whether habitat complexity is an important determinant of competitive outcomes between species. Together with lignin degradation as a functional trait, we measured growth rate on nutrient media and metabolic rate on cellulose to predict competitive outcomes during interactions on agar and wood. We determined the competitive outcomes between fungi by measuring relative colony size divided by total occupied space within a substrate, with the superior competitor having control over a larger area within the substrate. We predicted that: 1) species with a high metabolic rate will extract resources more rapidly, and will be competitively superior across substrates, and 2) the competitive outcomes established on simple resources will not translate to complex resources as the complex environment slows the process of competitive exclusion.

102

4.3 Methods

4.3.1 Species selection

We extracted our study species from rotting Eucalyptus tereticornis logs from remnant

Cumberland Plain woodland in Richmond, NSW (33°37'04.0"S 150°44'25.3"E) in February 2016. We collected the felled logs and brought them back to the lab. Under sterile conditions, we split the logs with a hammer and a flame-sterilised chisel to expose the centre of the log (Thompson et al. 2012).

Using flame sterilised forceps, we aseptically extracted wood chips from the exposed internal surface of the logs and transferred them to Petri dishes with 2% malt extract agar (MEA). The isolates were allowed to grow out from the wood chips and the growing edge of the emerging hyphae of morphologically distinct colonies were subcultured onto new MEA to isolate a pure culture. From the pure culture, we extracted DNA from the edge of the growing colonies with the

DNeasy Plant Mini Kit (Qiagen, Chadstone, Victoria, Australia) as per the manufacturer’s instruction.

We amplified the ITS (ITS1F & ITS4) region of rDNA (White et al. 1990, Gardes and Bruns 1993,

Thompson et al. 2012) through PCR amplification and analysed the amplicons using an ABI3500

Genetic Analyser (Applied Biosystems, Life Technologies, Mulgrave, Victoria, Australia). We assigned taxonomic identities to the fungal isolates by conducting a BLAST search against the NCBI Nucleotide database and comparing the amplicons against verified nucleotide sequences in the database.

Fungal isolates were kept on 1% MEA at 4°C.

To test whether the identified fungal isolates were able to degrade lignin, we took a 5mm plug of the colony and inoculated it onto media containing lignin-guaiacol-benomyl agar (Thorn et al.

1996). Fungal isolates that produced laccase or peroxidase in the presence of lignin (indicating lignolytic activity noted with a distinct bright red ring surrounding the edge of the colony) were categorised as a lignin degrader, while fungal isolates that did not produce laccase or peroxidase

103 were categorised as a non-lignin degrader. Of the species extracted from the wood logs, four were selected, spanning across two lignin degrading basidiomycetes: Omphalotus sp. (Om), Peniophora sp. (Pe); and two non-lignin degrading ascomycetes: Phacidium lacerum (Pl), and an unidentified

Helotiales isolate (He).

4.3.2 Microbial respiration assay

We utilised a modified MicroResp™ method (Campbell et al. 2003) to measure respiration of our study species on cellulose. The MicroResp™ technique is a colorimetric method in detecting carbon evolution of microbes on a specific carbon source (Campbell et al. 2003). We filled 1.2 mL deep-well plates (Thermofisher, Australia) with H2O agar to reduce atmospheric headspace in the deep-well plate. Eight replicates of single 4 mm plugs from a fungal colony were placed in individual wells, and 8 μl of a suspension of 5% milled cellulose in dH2O was added as the carbon source. Two sets of 4 replicates received dH2O and no added material respectively to act as basal respiration rates. Blank wells were included as a measure of base atmospheric change. Our study species were inoculated onto H2O agar to prevent any spurious measures of respiration on nutrient media.

We made an initial absorbance reading of the colorimetric detection plates at 570 nm with a

CLARIOstar plate reader (BMG LABTECH GmbH, Germany). The deep-well plates were sealed and clamped with the colorimetric detection plates following the MicroResp™ method (Campbell et al.

2003). The deep-well plates were incubated in the dark at 20°C. Following 48 hours, the colorimetric plate was re-read with a plate reader at 570 nm and resealed with a new colorimetric plate and incubated for a further 48 hours. After a full 96 hours, the colorimetric plate was re-read at 570 nm for a final absorbance reading. To calculate the hyphal mass, we followed the grid intersect method outlined in (Giovannetti and Mosse 1980) to determine total hyphal length in an inoculum. We measured average hyphal diameter per inoculum to define hyphal volume. We converted this

104 measure into mass following the average biovolume mass measurement in (van Veen and Paul

1979). The rate of respiration (μg CO2/g/h) on cellulose was calculated from the adsorption data from the colorimetric plates, minus the measures from the basal respiration rate of the fungal inoculum.

4.3.3 Experimental design

To compare competitive hierarchies across simple and complex substrates, we set up pairwise interactions between our study species across two substrate treatments: agar and wood.

Pairwise interactions were set up across all species for a total of six pairwise competition treatments

(OmHe, PeHe, PeOm, PePl, PlHe, PlOm).

4.3.4 Establishment of simple competitive environments

Prior to the experiment, the selected species were inoculated onto H2O agar media (10 mg/L) for 7 days to normalise hyphal density between species. 5-mm diameter plugs from the growing edge of the colony were taken and inoculated onto 1% MEA. In the competition plates, inoculum from the study species were placed 2.25 cm away from the edge of the Petri dish. The competitor in the pair was placed 2.25 cm from the opposite edge of the Petri dish such that each colony had equal area to develop. We also inoculated each species onto the centre of their own

Petri-dish to track colony growth rate in a non-competitive environment. All Petri dishes were incubated in the dark at 25°C except during measurements of colony area.

All paired interactions were allowed to interact over 27 days. We scored competition outcomes by calculating area occupancy within the Petri-dish. We assessed colony growth rate by measuring colony size every day on an acetate sheet with a black permanent marker. Acetate sheets

105 were scanned at the end of the experiment using a Canoscan LiDE 210 scanner (300 dpi resolution).

Colony area was measured in ImageJ (NIH, USA) using the polygon tool. We created a total number of 60 simple competitive interactions, 10 replicates for each species pair.

4.3.5 Establishment of wood block interactions

We soaked 8 cm3 blocks of Eucalyptus diversicolor wood (2 cm x 2 cm x 2 cm) overnight in

MilliQ water and sterilised in an autoclave three times over three days to remove any existing contaminants. The uninoculated wood blocks were maintained at -20°C until the start of the experiment. We colonised 2% MEA plates with our study species. We then placed 240 uninoculated wood blocks onto those agar plates for three months (90 days) in the dark at 20°C to allow full colonisation of the wood blocks. After the initial inoculation period, we tested wood blocks inoculated by our study species to ensure that the wood blocks were fully colonised.

To establish competitive interactions, inoculated wood blocks were removed from the agar plates and aseptically scraped free of any surface mycelium. Under strict aseptic conditions, we arranged the inoculated wood blocks into their interactions and held them together with a sterilised rubber band so the wood grain on each block were face to face. These interaction blocks were then placed in a sterilised 100 mL plastic sauce pot (WF Plastic Pty Ltd, AU) containing 30 ml moistened sterile perlite and incubated at 20°C in the dark. A hole in the side of the pot was covered with micropore tape (3M, UK) to allow gas exchange. Wood block interactions were kept in the dark at

20°C and each paired interaction was harvested at 7, 21, 42, 70, and 105 days. We created a total of

120 complex competitive environments, 20 for each species pair, which allowed us to sample them at different time frames.

To determine the competitive outcome of each pair, we measured area occupancy in each wood block by sampling a transect through each two-block combination. To perform this sampling,

106 each wood block was split in half along the grain using a flame sterilised chisel. We then aseptically excised a small sliver of wood (~2 mm3) at 5 mm, 10 mm, and 15 mm from the point of contact between two blocks. Each point of isolation represented 25% occupancy. Although this does not allow us to calculate exactly how much space much of the initial occupancy was lost/gained during the competition in the 3D environment (as it is possible that individuals did not grow across as a single plane through the block), it allows us to estimate how far a competitor was able to intrude into an owner’s territory. We did not isolate from the far face of the block to avoid any surface hyphae that may have crept across the surface of the wood block. Each sliver of wood was reisolated on 1% MEA and incubated at 20°C until hyphae emerged and could be identified morphologically.

4.3.6 Statistical analysis

To test for differences in respiration rate between study species, we fitted a linear model

(ANOVA) with study species as an independent variable and carbon dioxide production as the dependent variable. We used a Tukey’s HSD to compare the respiration rates between species. We used a linear model (ANOVA) to compare the growth rates between study species when they were all reared in isolation on a single petri dish. We also used a linear model to explore the overall relationship between respiration rate and growth rate of the four species.

We quantified competitive outcomes between pairs of species as rate of proportional change of colony area over time on agar and across a linear transect within the three-dimensional wood substrate (see Hiscox et al. 2015). Interaction time varied between competitive treatments on agar as we measured proportional change in colony area over time after colonies met. To compare how the proportional change in colony area over time varied between substrates, we standardised the colony area measurements over time for both substrates and conducted the analysis on the combined data from both wood and agar substrates. To model the rate of change in area/volume

107 occupied, we fitted a mixed model (lme4; Bates et al. 2015), with the rate of proportional change of colony occupation after colony meeting as the response variable. Proportion of colony occupied was measured as a change in percentage occupation within the substrate. Time and substrate type were modeled as fixed factors along with a potential interaction between the two, and to account for the repeated measurements on the same petri dish, an experimental unit (i.e., petri dish ID) was included as a random factor. Each species pairing was analysed separately. The proportional area occupied by one species in the combined area available was measured as a percentage of total area for each species pairing treatment (as a percentage out of 100). The identity of the species measured did not affect the analysis as we focused on the overall rate of total proportional change in area occupied over time between substrates. While the difference in time points between agar and wood reduces the overall power of the model, the mixed model is still sufficient to address the difference in rate of proportional change in colony area over time between substrates. In this model the interaction term represents a shift in the slope of the occupied area with time across substrates.

All analyses were run in R (v.4.0.2).

Competitive hierarchies were categorically established by assessing and scoring competitive outcomes of interactions between pairwise interaction treatments of our study species at the end of the experiment for both agar and wood substrates. In both interactions on agar and wood, we scored species with a greater total proportional area colonised (>50%) as the superior competitor.

4.4 Results

4.4.1 Respiration rate and growth rate

The rate of cellulose respiration was significantly different across species (ANOVA,

F3,28=21.87, p<0.0001). With post hoc comparisons using the Tukey’s HSD test, we found significant

108 differences (P < 0.05) between all pairs of species with the exception of Omphalotus sp. and

Peniophora sp. (Supp. Table 1). The rate of respiration was lower in both the lignin degrading species; Omphalotus sp. and Peniophora sp. had a mean rate of respiration of 69.1 (μg CO2/g/h) and

63.31 (μg CO2/g/h), respectively. The two non-lignin degrading species had higher rates of respiration; the unidentified Helotiales isolate had a rate of respiration of 97.16 (μg CO2/g/h) and

Phacidium lacerum had the highest rate of respiration at 130.9 (μg CO2/g/h).

There were significant differences in growth rate on agar between all study species (ANOVA,

F3,116=673.8, p<0.0001). We found significant differences in growth rate between all species (Tukey’s

HSD, P < 0.05) (Supp. Table 1). Phacidium lacerum had the highest growth rate at 20.07 mm2/h, followed by Peniophora sp. at 13.21 mm2/h. Omphalotus sp. and the Helotiales isolate had the lowest growth rate at 8.24 mm2/h and 6.35 mm2/h respectively. We did not observe a relationship between growth rate on agar and the rate of respiration on cellulose.

4.4.2 Patterns of competition between fungi on agar and wood

We observed a strong effect of the competitive environment on treatments (mixed model,

Table 2). The competitive environment had a significant effect on the pattern of competition between study species over time in treatments: PeHe (P=0.003), PeOm (P<0.001), PlHe (P<0.001), and PlOm (P<0.001) (Table 1). The competitive environment did not affect competition between study species in the OmHe and PePl treatments (P=0.318 and P=0.116 respectively, Table 1). The competitive environment interacted with time in all treatments except OmHe (P=0.466, Table. 1) to affect the rate of proportional change in colony occupied between interacting species.

109

Table 1. Mixed model with time and competitive environment as fixed factors, with the rate

of proportional change of colony occupation as the response variable. To account for repeated

measurements of the same Petri-dish, replicate ID was included as a random factor. Here the

intercept represents the difference in slopes between agar and wood, with time being the average

slope of both treatments (or the speed by which proportioned occupied changes). In this model, the

substrate term represents the difference in competitive performance between agar and wood (see

Supplementary Fig. 1). Degrees of freedom differed between individual models because we

measured interaction times after colony meeting in the agar substrate treatment.

OmHe

Parameter Coefficient SE 95% CI t df p

(Intercept) 0.57 0.02 [0.53, 0.62] 23.59 54 <0.001

Time (in weeks) -0.02 0.01 [-0.03, 0.00] -2.72 54 0.007

Substrate (wood) -0.03 0.03 [-0.10, 0.03] -1 54 0.318

Time * Substrate 0 0.01 [-0.01, 0.02] 0.73 54 0.466

PeHe

Parameter Coefficient SE 95% CI t df p

(Intercept) 0.7 0.04 [0.62, 0.78] 17.94 64 <0.001

Time (in weeks) 0.03 0 [0.02, 0.04] 6.25 64 <0.001

Substrate (wood) -0.18 0.06 [-0.29, -0.06] -2.93 64 0.003

Time * Substrate -0.04 0.01 [-0.05, -0.02] -5.04 64 <0.001

110

PeOm

Parameter Coefficient SE 95% CI t df p

(Intercept) 0.6 0.01 [0.58, 0.62] 56.86 64 <0.001

Time (in weeks) 0.01 0 [0.01, 0.02] 7.89 64 <0.001

Substrate (wood) -0.1 0.02 [-0.14, -0.07] -6.47 64 <0.001

Time * Substrate -0.02 0 [-0.02, -0.01] -6.66 64 <0.001

PePl

Parameter Coefficient SE 95% CI t df p

(Intercept) 0.38 0.03 [0.32, 0.43] 13.41 84 <0.001

Time (in weeks) 0.02 0 [0.01, 0.02] 5.83 84 <0.001

Substrate (wood) 0.07 0.04 [-0.02, 0.16] 1.57 84 0.116

Time * Substrate 0.01 0 [0.00, 0.02] 2.06 84 0.04

PlHe

Parameter Coefficient SE 95% CI t df p

(Intercept) 0.22 0.01 [0.20, 0.23] 21.57 64 <0.001

Time (in weeks) 0.03 0 [0.03, 0.03] 12.81 64 <0.001

Substrate (wood) 0.27 0.01 [0.25, 0.30] 18.48 64 <0.001

Time * Substrate -0.03 0 [-0.03, -0.02] -10.28 64 <0.001

PlOm

Parameter Coefficient SE 95% CI t df P

(Intercept) 0.88 0.03 [0.81, 0.94] 25.93 74 <0.001

Time (in weeks) -0.09 0 [-0.09, -0.08] -21.18 74 <0.001

Substrate (wood) -0.38 0.05 [-0.48, -0.27] -7.11 74 <0.001

Time * Substrate 0.07 0.01 [0.06, 0.09] 11.98 74 <0.001

111

4.4.3 Competitive hierarchies on agar and wood

A distinct hierarchy emerged in competition on simple substrates that was not mirrored in the hierarchy established on complex substrates (Table 1). In the pairwise interactions on agar, respiration on cellulose and initial growth rate prior to colony meeting was a poor predictor for competitive success (Fig. 1). Peniophora sp. Was the strongest competitor among all interactions on agar, outgrowing and reducing the size of its opponent. On the other hand, Phacidium lacerum was the weakest competitor, consistently losing territory against its opponents despite having the highest growth rate prior to colony meeting. We observed no instances of deadlock between fungal colonies on agar, where neither competitor could capture territory from the other (Fig. 2). The competitive hierarchy in wood was less clear (Table 2). Where Peniophora sp. dominated in competition against its opponents on agar, Peniophora sp. was only competitively superior to

Phacidium lacerum, while reaching a deadlock with Omphalotus sp., and losing territory to the unidentified Helotiales isolate. The unidentified Helotiales isolate emerged as the superior competitor in pairwise interactions in wood and became competitively superior to Peniophora sp. in stark contrast to interactions on agar, where the unidentified Helotiales isolate was the weaker competitor against Peniophora sp. Across all interactions on agar and wood, Peniophora sp. consistently outcompeted Phacidium lacerum, and the Helotiales isolate consistently outcompeted

Omphalotus sp. (Table 2).

112

Table 2. Competitive hierarchy established between pairwise interactions on Wood and Agar between Peniophora sp. (Pe), Phacidium lacerum (Pl), Omphalotus sp. (Om), and the Helotiales isolate (He). Interactions where a species gained territory is denoted by (+), interactions where species lost territory is denoted by (-), and interactions where neither species gained territory is denoted by (=).

Wood + = - Agar + = - Pe Pl Om He Pe Pl, Om, He

Pl He Om, Pl Pl Om, He, Pe

He Pe, Om Pl He Om, Pl Pe

Om Pl Pe He Om Pl Pe, He

113

Figure 1. Comparison of total average growth rate ±SE (mm2 per hour) before and after colony meeting on Agar. Colonies with negative growth rate post interaction lost territory to the superior competitor. Peniophora sp. (Pe) consistently outcompeted its neighbour, and Phacidium lacerum (Pl) consistently lost against its neighbour. Size of the circles denote individual respiration rate on cellulose (μg CO2/g/h).

114

Figure 2. Average proportion of territory (% occupied) captured for the Helotiales isolate (He),

Omphalotus sp. (Om), Peniophora sp. (Pe), and

Phacidium lacerum (Pl) during interactions on agar

(right) and in wood (left) over time. In interactions on wood, both competitors started with equal territory (one wood block), while competitors on agar started without any territory (inoculum on agar).

115

4.5 Discussion

4.5.1 Patterns of competition across simple and complex substrates

The pattern of competition between species was not consistent between agar and wood. In four out of the six species pairs, the rate of competitive displacement differed between agar and wood. In competitive interactions on agar media, all species pairs resulted in a superior competitor

(Fig. 1). This was in contrast with the competitive outcomes seen in wood, where competitive deadlock between two species pairs were observed, along with a complete competitive reversal seen in one species pair. This was in line with our prediction that the competitive hierarchy established on simple substrates would not translate to complex substrates.

The complexity of resources found within substrates may be a key factor in altering the competitive dynamics between study species. Habitat complexity is not categorical, but rather a continuous variable that is made up of several different factors (e.g., resource availability, temperature, space) that can alter the competitive landscape. For saprotrophic fungi, resource supply is finite and regardless of substrate type, competition between species would progress until all available resources are exhausted through active mycelial growth or directly through combat between fungi (Owens et al. 1994, Hiscox et al. 2018). After colony contact, the competitively superior species would use whatever limited resources were available to replace the territory held by its opponent (Hiscox et al. 2015). In the agar substrate, the nutrients were highly labile in the form of glucose with added malt extract and peptone. The soft structure and nutrient availability would provide an unimpeded opportunity for hyphal extension during periods of active growth in a homogenous environment, thus influencing how competition progresses between species

(Amarasekare 2003). In contrast, the resources available in wood are heterogeneously arranged within the lignocellulose complex (Otjen and Blanchette 1986, Blanchette 1991). Not all fungi have the ability to break down lignocellulose (Swift 1977, Sánchez 2009, Nagy et al. 2016), and the

116 process is enzymatically costly (Martinez et al. 2009, Sánchez 2009). In competitive interactions on simple substrates, the rapid conversion of readily available nutrients into growth would alter the allocation of resources in the absence of the energetically expensive degradation of lignocellulose in more complex substrates like wood (Woods et al. 2005). Interestingly, the ability of species to break down lignocellulose did not predict contest outcome in the more complex wood block environment.

This supports the fact that it is not resource ability itself that altered competitive outcomes, but the complexity of the competitive environment.

Environmental complexity, and not access to resources in the environment, may explain the difference in competitive dynamics seen across simple and complex substrates. Competition on an agar plate is largely two-dimensional where the colony edge of both interacting species will only meet along a single front. As such, asymmetric competition will always favour the superior competitor, resulting in competitive exclusion of the inferior competitor (Amarasekare 2003). This is seen under single resource model predictions (Volterra 1926, Macarthur and Levins 1964, Armstrong and McGehee 1980), and on experiments using model systems (Grover 1988, Passarge et al. 2006).

However, under heterogeneous conditions, spatial variation can promote persistence in the inferior competitor through either life-history trade-offs or altered use of the local environment

(Amarasekare 2003). Microstructural variation within the wood forms cell walls and plant vasculature that can influence how fungi move within the substrate (Boddy 2001, Boddy and

Heilmann-Clausen 2008). The added spatial component introduces new elements that can influence the outcome of interspecific interactions (Biondini 2001, Hesterberg et al. 2017). Environmental heterogeneity is known to affect competition dynamics at multiple spatial scales (Chesson and

Rosenzweig 1991, Amarasekare and Nisbet 2001), and the microstructural heterogeneity found in wood could be a key factor in attenuating the competitive superiority seen in pairwise interactions on agar media. In our experiment, we see competition between species reach a deadlock on a complex substrate, while the same species pair will result in competitive dominance of a single species on a simple substrate. We also see no change in competitive outcome in a species pair

117 between complex and simple substrates. In this case, the degree of competitive asymmetry between interacting species may override the influence of spatial heterogeneity within the environment.

In models of competition, the nature of interaction between species is often scaled down to its bare elements, such as competition over a single or scant resource, or species represented by singular key traits (Abrams 1988, Sommer 1989, Chesson 2000). Models of competition are by necessity simple, and are often matched to equally simple experiments (Simberloff 1982), with few modeling frameworks in fungal competition explicitly considering space (Gilchrist et al. 2006) and priority effects (Abrams 1988, Ke and Letten 2018). Space occupancy is a fundamental part of fungal ecology, with simplified models and controlled experiments to examine the mechanistic processes of competition (Kolesidis et al. 2019). It is important to note that our measures of space occupancy within agar and wood differed in granularity, but here we show that competitive hierarchies and dynamics from complex, more ecologically relevant environments are not maintained in simple ones, and as such the addition of complexity is vital in understanding how the patterns we observe in experimental systems may scale back out to natural settings.

4.5.2 Metabolic rate and competitive ability

Drawing on plant ecology theory, we predicted that a rapid metabolic rate and growth rate would confer a competitive advantage against neighbours during pairwise interactions (Tilman 1987,

Abrams 1988, Aerts 1999, Craine and Dybzinski 2013). However, metabolic rate on cellulose appeared to be a poor predictor for superior competitive ability within the pairwise interactions between our four study species. But within our limited species pool, the true relationship may be difficult to detect. The fundamental differences in competition between heterotrophs and autotrophs may also limit the application of plant ecology theory to fungal ecology, with continuous vs. finite resource supply being a key difference. While rapid growth in plants leads to a competitive

118 advantage in denying competitors access to sunlight (Grime 1977), active antagonism and resistance to competition is key in competition over space between fungi (Boddy and Hiscox 2016). In our metabolic assay, we restricted our choice of substrate to reflect resources found in wood. All our study species had demonstrated the ability to metabolise cellulose, but only Peniophora sp. and

Omphalotus sp. were able to degrade lignin. Despite this, the ability to degrade lignin was not associated with a competitive advantage over the non-lignin degrading Helotiales isolate when fungi were competing in wood.

Attempting to predict competitive ability from a single trait is limited when scaling up to a natural system (Gaudet and Keddy 1988, Pascual and Kareiva 1996, Tilman 2007). In simple models of competition and empirical tests in simple environments, a single trait can often predict competitive superiority (Tilman 1994), but as complexity is introduced this predictive power is reduced. While many saprotrophic fungi can share a similar environment, no one trait can confer a competitive advantage across all substrates. As an environment changes, the competitive advantage of certain traits may wax or wane (Grime 1977, Funk and Wolf 2016). Utilising multiple key functional traits may provide a better framework in predicting competitive outcomes between fungi

(Crowther et al. 2014, Aguilar-Trigueros et al. 2015, Treseder and Lennon 2015, Cline and Zak 2015,

Dawson et al. 2019, Lehmann et al. 2019, Lustenhouwer et al. 2020). Further understanding how fungal functional traits may interact with substrate complexity may provide us insight in how fungal competitive hierarchies and communities are shaped by their environments and how the diverse array of fungal species co-occur in wood.

119

4.5.3 Conclusion

The struggle for existence between organisms will continually shift depending on changes in the environment (Amarasekare 2003). As species interactions on simple substrates scale to more complex substrates, the existing competitive hierarchy may not scale in the same way (Woods et al.

2005). The use of simple models of competition can allow us to observe the mechanistic underpinnings of interactions between species. But the addition of heterogeneity and complex elements into models of competition may begin to reconcile some of the patterns of species coexistence in natural environments. Here, we demonstrated the change in competitive dynamics in saprotrophic fungi from simple agar media to competition within wood blocks. The sheer complexity within wood, both in structure and resource, may be a driver in the immense fungal diversity that we see in nature by preventing any single species from becoming competitively dominant. Further focus on the interactions between traits within the substrates and the functional traits within fungi may lead the way in demystifying the processes that allow for the very diverse fungal communities that we observe in nature.

120

4.6 References

Abrams, P. A. 1988. Resource Productivity-Consumer Species Diversity: Simple Models of Competition in Spatially Heterogeneous Environments. - Ecology 69: 1418–1433.

Aerts, R. 1999. Interspecific competition in natural plant communities: mechanisms, trade-offs and plant– soil feedbacks. - J. Exp. Bot. 50: 29–37.

Aguilar-Trigueros, C. A. et al. 2015. Branching out: Towards a trait-based understanding of fungal ecology. - Fungal Biol. Rev. 29: 34–41.

Amarasekare, P. 2003. Competitive coexistence in spatially structured environments: a synthesis: Spatial coexistence mechanisms. - Ecol. Lett. 6: 1109–1122.

Amarasekare, P. and Nisbet, R. M. 2001. Spatial heterogeneity, source-sink dynamics, and the local coexistence of competing species. - Am. Nat. 158: 572–584.

Armstrong, R. A. and McGehee, R. 1980. Competitive Exclusion. - Am. Nat. 115: 151–170.

Bates D, Mächler M, Bolker B, Walker S (2015). “Fitting Linear Mixed-Effects Models Using lme4.” Journal of Statistical Software, 67(1), 1–48. doi: 10.18637/jss.v067.i01.

Biondini, M. 2001. A three-dimensional spatial model for plant competition in an heterogeneous soil environment. - Ecol. Modell. 142: 189–225.

Blanchette, R. A. 1991. Delignification by wood-decay fungi. - Annu. Rev. Phytopathol. 29: 381–403.

Boddy, L. 2001. Fungal Community Ecology and Wood Decomposition Processes in Angiosperms: From Standing Tree to Complete Decay of Coarse Woody Debris. - Ecol. Bull.: 43–56.

Boddy, L. and Heilmann-Clausen, J. 2008. Chapter 12 Basidiomycete community development in temperate angiosperm wood. - In: Boddy, L. et al. (eds), British Mycological Society Symposia Series. Academic Press, pp. 211–237.

Boddy, L. and Hiscox, J. 2016. Fungal Ecology: Principles and Mechanisms of Colonization and Competition by Saprotrophic Fungi. - Microbiol Spectr in press.

121

Campbell, C. D. et al. 2003. A rapid microtiter plate method to measure carbon dioxide evolved from carbon substrate amendments so as to determine the physiological profiles of soil microbial …. - Applied and in press.

Chesson, P. 2000. General theory of competitive coexistence in spatially-varying environments. - Theor. Popul. Biol. 58: 211–237.

Chesson, P. and Rosenzweig, M. 1991. Behavior, Heterogeneity, and the Dynamics of Interacting Species. - Ecology 72: 1187–1195.

Cline, L. C. and Zak, D. R. 2015. Initial colonization, community assembly and ecosystem function: fungal colonist traits and litter biochemistry mediate decay rate. - Mol. Ecol. 24: 5045–5058.

Craine, J. M. and Dybzinski, R. 2013. Mechanisms of plant competition for nutrients, water and light. - Funct. Ecol. 27: 833–840.

Crowther, T. W. et al. 2014. Untangling the fungal niche: the trait-based approach. - Front. Microbiol. 5: 579.

Dawson, S. K. et al. 2019. Handbook for the measurement of macrofungal functional traits: A start with basidiomycete wood fungi (M Hart, Ed.). - Funct. Ecol. 33: 372–387.

Funk, J. L. and Wolf, A. A. 2016. Testing the trait-based community framework: Do functional traits predict competitive outcomes? - Ecology 97: 2206–2211.

Gardes, M. and Bruns, T. D. 1993. ITS primers with enhanced specificity for basidiomycetes--application to the identification of mycorrhizae and rusts. - Mol. Ecol. 2: 113–118.

Gaudet, C. L. and Keddy, P. A. 1988. A comparative approach to predicting competitive ability from plant traits. - Nature 334: 242–243.

Gause, G. F. 1932. Experimental studies on the struggle for existence: I. Mixed population of two species of yeast. - J. Exp. Biol. 9: 389–402.

Gause, G. F. 1934. The struggle for existence. Reprint; 1971. in press.

Gause, G. F. et al. 1934. The Influence of Biologically Conditioned Media on the Growth of a Mixed Population of Paramecium caudatum and P. aureliax. - J. Anim. Ecol. 3: 222–230.

122

Gilchrist, M. A. et al. 2006. Identifying fitness and optimal life-history strategies for an asexual filamentous fungus. - Evolution 60: 970–979.

Giovannetti, M. and Mosse, B. 1980. An Evaluation of Techniques for Measuring Vesicular Arbuscular Mycorrhizal Infection in Roots. - New Phytol. 84: 489–500.

Grime, J. P. 1977. Evidence for the Existence of Three Primary Strategies in Plants and Its Relevance to Ecological and Evolutionary Theory. - Am. Nat. 111: 1169–1194.

Grover, J. P. 1988. Dynamics of Competition in a Variable Environment: Experiments with Two Diatom Species. - Ecology 69: 408–417.

Hardin, G. 1960. The competitive exclusion principle. - Science 131: 1292–1297.

Hesterberg, S. G. et al. 2017. Three-dimensional interstitial space mediates predator foraging success in different spatial arrangements. - Ecology 98: 1153–1162.

Hiscox, J. et al. 2015. Antagonistic fungal interactions influence carbon dioxide evolution from decomposing wood. - Fungal Ecol. 14: 24–32.

Hiscox, J. et al. 2018. Fungus wars: basidiomycete battles in wood decay. - Stud. Mycol. 89: 117–124.

Holmer, L. and Stenlid, J. 1997. Competitive Hierarchies of Wood Decomposing Basidiomycetes in Artificial Systems Based on Variable Inoculum Sizes. - Oikos 79: 77–84.

Huffaker, C. (1958). Experimental studies on predation: dispersion factors and predator-prey oscillations. Hilgardia, 27(14), 343-383.

Hutchinson, G. E. 1961. The Paradox of the Plankton. - Am. Nat. 95: 137–145.

Ke, P.-J. and Letten, A. D. 2018. Coexistence theory and the frequency-dependence of priority effects. - Nat Ecol Evol 2: 1691–1695.

Kerr, B., M. A. Riley, M. W. Feldman, and B. J. M. Bohannan. 2002. Local dispersal promotes biodiversity in a real-life game of rock-paper-scissors. Nature 418:171-174.

Kolesidis, D. A. et al. 2019. Predicting fungal community dynamics driven by competition for space. - Fungal Ecol. 41: 13–22.

Lehmann, A. et al. 2019. Tradeoffs in hyphal traits determine mycelium architecture in saprobic fungi. - Sci. Rep. 9: 14152.

123

Leibold, M. A. 1995. The niche concept revisited: Mechanistic models and community context. - Ecology 76: 1371–1382.

Levin, S. A. 1976. Population dynamic models in heterogeneous environments. - Annu. Rev. Ecol. Syst. 7: 287–310.

Levins, R. and Culver, D. 1971. Regional Coexistence of Species and Competition between Rare Species. - Proc. Natl. Acad. Sci. U. S. A. 68: 1246–1248.

Lotka, A. J. 1926. Elements of physical biology. - Science Progress in the Twentieth Century (1919-1933) 21: 341–343.

Lustenhouwer, N. et al. 2020. A trait-based understanding of wood decomposition by fungi. - Proc. Natl. Acad. Sci. U. S. A. 117: 11551–11558.

Macarthur, R. and Levins, R. 1964. Competition, Habitat Selection, and Character Displacement in a Patchy Environment. - Proc. Natl. Acad. Sci. U. S. A. 51: 1207–1210.

Mäkipää, R. et al. 2017. Interactions between soil- and dead wood-inhabiting fungal communities during the decay of Norway spruce logs. - ISME J. 11: 1964–1974.

Martinez, D. et al. 2009. Genome, transcriptome, and secretome analysis of wood decay fungus Postia placenta supports unique mechanisms of lignocellulose conversion. - Proc. Natl. Acad. Sci. U. S. A. 106: 1954–1959.

Nagy, L. G. et al. 2016. Comparative Genomics of Early-Diverging Mushroom-Forming Fungi Provides Insights into the Origins of Lignocellulose Decay Capabilities. - Mol. Biol. Evol. 33: 959–970.

Otjen, L. and Blanchette, R. A. 1986. A discussion of microstructural changes in wood during decomposition by white rot basidiomycetes. - Can. J. Bot. 64: 905–911.

Owens, E. M. et al. 1994. Outcome of interspecific interactions among brown-rot and white-rot wood decay fungi. - FEMS Microbiol. Ecol. 14: 19–24.

Pascual, M. A. and Kareiva, P. 1996. Predicting the Outcome of Competition Using Experimental Data: Maximum Likelihood and Bayesian Approaches. - Ecology 77: 337–349.

Passarge, J. et al. 2006. Competition for nutrient and light: stable coexistence, alternative stable states, or competitive exclusion? - Ecol. Monogr. 76: 57–72.

124

Roughgarden, J. 1983. Competition and Theory in Community Ecology. - Am. Nat. 122: 583–601.

Sánchez, C. 2009. Lignocellulosic residues: biodegradation and bioconversion by fungi. - Biotechnol. Adv. 27: 185–194.

Simberloff, D. 1982. The status of competition theory in ecology. - Ann. Zool. Fennici 19: 241–253.

Slatkin, M. 1974. Competition and Regional Coexistence. - Ecology 55: 128–134.

Sommer, U. 1989. The Role of Competition for Resources in Phytoplankton Succession. - In: Sommer, U. (ed), Plankton Ecology: Succession in Plankton Communities. Springer Berlin Heidelberg, pp. 57–106.

Sommer, U. 1999. Competition and coexistence. - Nature 402: 366–367.

Swift, M. J. 1977. The Roles of Fungi and Animals in the Immobilisation and Release of Nutrient Elements from Decomposing Branch-Wood. - Ecol. Bull.: 193–202.

Thompson, T. A. et al. 2012. Hypholoma lateritium isolated from coarse woody debris, the forest floor, and mineral soil in a deciduous forest in New Hampshire. - Botany 90: 457–464.

Thorn, R. G. et al. 1996. Isolation of saprophytic basidiomycetes from soil. - Appl. Environ. Microbiol. 62: 4288–4292.

Tilman, D. 1987. On the Meaning of Competition and the Mechanisms of Competitive Superiority. - Funct. Ecol. 1: 304–315.

Tilman, D. 1994. Competition and Biodiversity in Spatially Structured Habitats. - Ecology 75: 2–16.

Tilman, D. 2007. Resource Competition and Plant Traits: A Response to Craine et al. 2005. - J. Ecol. 95: 231–234.

Tilman, D. et al. 1982. Phytoplankton Community Ecology: The Role of Limiting Nutrients. - Annu. Rev. Ecol. Syst. 13: 349–372.

Treseder, K. K. and Lennon, J. T. 2015. Fungal traits that drive ecosystem dynamics on land. - Microbiol. Mol. Biol. Rev. 79: 243–262. van Veen, J. A. and Paul, E. A. 1979. Conversion of biovolume measurements of soil organisms, grown under various moisture tensions, to biomass and their nutrient content. - Appl. Environ. Microbiol. 37: 686–692.

125

Volterra, V. 1926. Fluctuations in the Abundance of a Species considered Mathematically 1. - Nature 118: 558–560.

Wangersky, P. J. 1978. Lotka-Volterra Population Models. - Annu. Rev. Ecol. Syst. 9: 189–218.

White et al. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. PCR protocols: a guide to methods and applications. - unknown.

Wilson, S. D. and Tilman, D. 1993. Plant Competition and Resource Availability in Response to Disturbance and Fertilization. - Ecology 74: 599–611.

Woods, C. M. et al. 2005. In vitro interactions in artificial and wood-based media between fungi colonizing stumps of Sitka spruce. - For. Pathol. 35: 213–229.

126

4.7 Supplementary Material

Supplementary Figure 1. Theoretical mixed model results detailing difference in slope of proportion occupied between wood and agar as substrates. Black circle defined as 50% proportion occupancy.

127

Supplementary Table 1. Tukey’s HSD test comparisons for respiration rate between species. With the exception of Peniophora sp. and Omphalotus sp., all test species had significantly different rates of respiration on cellulose.

Tukey’s HSD

p

Om - He 0.006

Pe - He 0.027

Pl - He 0.006

Pe - Om 0.925

Pl - Om <0.0005

Pl - Pe <0.0005

Supplementary Table 2. Tukey’s HSD test comparisons for growth rate between species. All test species had significantly different growth rates on agar.

Tukey’s HSD

p

Om - He <0.0005

Pe - He <0.0005

Pl - He <0.0005

Pe - Om <0.0005

Pl - Om <0.0005

Pl - Pe <0.0005

128

CHAPTER 5

The effect of fungal dynamics on early wood decomposition: insights from a laboratory experiment

Justin Y. Chan, Stephen P. Bonser, Jeff R. Powell, William K. Cornwell

129

5.1 Abstract

Dead wood on forest floors is a globally important pool of carbon and understanding the decomposition dynamics of wood is an important goal. Field measurements of the decay of this wood often reveal a very slow start to decay followed by a later speed up in the rate of decay. The mechanism for this early lag is largely unknown with hypotheses in the literature to explain it focusing on the dynamics of the wood-decay fungi, through some combination of decomposer community interactions, priority effects, and competition over space within wood. To better understand the contribution of species interactions to early-phase decomposition, we used a paired wood-block decomposition experiment in which colonisation was carefully controlled to include only a single wood-degrading fungal species or a pair of species each with the opportunity to colonise their own block before interactions began. We found that for both single species and pairs of species, decay was negligible: no notable mass loss in wood blocks were detected in the first three months following a three-month initial period of colonisation, in line with lag phases commonly observed in field experiments. In the wood blocks, the estimated amount of biomass imported by fungi differed between lignin degrading Basidiomycetes and non-lignin degrading Ascomycetes species despite the same opportunity for import, with wood blocks colonised by non-lignin degrading species increasing in mass. Interspecific competition was found to have a context dependent effect on the mass of wood blocks, with the superior competitor in paired wood blocks having a significant effect on the combined mass of wood blocks in interspecific pairings, while competition without any clear winner resulted in little effect of competition on the mass of wood blocks. Our results highlight that import and competition in fungi may be key components that shape the lag phase in early wood decomposition.

130

5.2 Introduction

Wood comprises a large fraction of the carbon pool in terrestrial ecosystems (Cornwell et al.

2009). After tree death, the wood is sequentially colonised and degraded by microbes (chiefly by saprotrophic fungi) and the stored carbon is released back into the atmosphere through microbial respiration (Schimel 1995). As such, saprotrophic fungi are responsible for a major flux of carbon in terrestrial ecosystems in the global carbon cycle (Cornwell et al. 2009, Lustenhouwer et al. 2020).

Wood decay occurs over relatively long time scales, and the process from standing tree to complete decomposition can take many decades (Rock et al. 2008, Weedon et al. 2009). While much of the focus in wood decay research has been on the overall rate of decomposition, the process of decay is complex, and how the mechanisms of competitive interactions between saprotrophs shape the rate and trajectory of decay is poorly understood (Hiscox et al. 2016).

While the simplest and most common models of decomposition assume steady rates of decay per mass (ie. exponential decline), empirical evidence suggests that decomposition trajectories are typically much more complex (Cornwell and Weedon 2014, Oberle et al. 2020). Field studies have shown contrasting effects of both environmental factors, and individual wood traits on the rate of decomposition (Cornwell et al. 2009). Precise tracking of wood decay in field conditions is logistically difficult (Franklin et al. 1987, Harmon et al. 2000), so models based off existing decay data are often used (Freschet et al. 2012). The most commonly used exponential decay model predicts that wood has a constant relative decomposition rate per mass through time (Freschet et al. 2012), but this model cannot adequately represent the decomposition trajectory for all tree species

(Harmon et al. 2000, Cornwell and Weedon 2014). Many tree species have an initial slow rate of decay followed by more rapid mass loss (Freschet et al. 2012). In these species, sigmoidal decay dynamics predict an initial lag phase occurs before mass loss begins (Freschet et al. 2012). Specific plant traits have been hypothesised to drive this lag (Hale and Pastor 1998, Cornwell et al. 2008,

131

2009, Radtke et al. 2009, Zell et al. 2009), but how fungi contribute to this lag in wood decay is uncertain.

Fungal interactions jointly diminish the resources found in wood, with the outcome of interactions determining the progression of decay. In living trees, endophytic fungi are present in the living sapwood, typically without causing any decay (Arnold et al. 2003, Sieber 2007, Rodriguez et al.

2009, Lee et al. 2019), but once a tree has died, many of these species may then act as saprotrophs

(Parfitt et al. 2010). After tree death, later arriving fungi advance the process of wood decay through sequential competitive displacement (Boddy 2001). Different species of fungi metabolise available resources at different rates (Worrall et al. 1997, van der Wal et al. 2015), and over time, the combative processes within an interacting fungal community in wood will influence how the resources present are metabolised, thus altering the overall rate of wood decay (Wells and Boddy

2002).

The stochastic nature of species arrival can influence how wood decomposes. Chiefly, the assembly history of saprotrophic fungi may affect subsequent species establishment, with consequences for wood decay (Fukami et al. 2010). Known as priority effects, the establishment of species during the earliest stages of fungal community assembly can affect the rate of decay over time (Dickie et al. 2012). Where species are able to both successfully establish territory and have the enzymatic capacity to extract resources from wood, it can persist until it is competitively replaced by another species, it has used up all available resources, is killed by external abiotic conditions, or produces a fruiting body for dispersal (Boddy and Hiscox 2016). If a species is able to arrive early and persist for long periods of time, it is possible that it can inhibit the establishment of late arriving species (Heilmann-Clausen and Boddy 2005, Fukami 2015). In such a scenario, the early arriving species can greatly influence the trajectory of decay, with slow decayers contributing to the initial lag in wood decomposition.

132

The mode of entry into wood may also influence how wood decay may progress. Fungi arriving by mycelia can import nutrients through hyphae (Boswell et al. 2002, Tlalka et al. 2008,

Philpott et al. 2014), and can preferentially translocate nutrients to different organic resources

(Wells and Boddy 1995). The increase of nutrients can provide an advantage in the competitive environment in wood (Tlalka et al. 2008, Boddy and Hiscox 2016), and preconditioning the wood for subsequent decay (Philpott et al. 2014). The nutrient import in certain species in the early stage of colonisation aids in the establishment of territory within the wood (Boddy 1999, 2001), and may be a contributing factor to the lag phase observed in wood decay. After the initial phase of colonisation of dead wood, secondary colonisers arrive and existing endophytes with limited decay capacity can become competitively displaced as more aggressive wood decay species begin to colonise the substrate (Holmer and Stenlid 1997).

Competition and territory capture within wood is a fundamental aspect of saprotrophic fungi, but the influence of these two factors on the early stage of wood decay is poorly understood. Here we explore this idea with a paired wood block experiment, testing how territory capture and competition between saprotrophic fungi may affect wood in the earliest stages of decomposition. Using a wood block combination experiment in which wood blocks are colonised by a single fungal species and then are placed in close proximity (Hiscox et al. 2016), we tested how fungi shape wood decay following initial colonisation under controlled lab conditions with four study species over 6 months in total. We examined mass change in wood blocks colonised with a single species, and separately established 6 interspecific pairwise interactions with intraspecific interactions included as a procedural control. Previous work has established a competitive hierarchy among the species presented in this study (Chan et al. 2021), and with this hierarchy we can test how interaction strengths may shape the outcome of wood decay. We compared outcomes in wood decomposition between individual wood blocks and interspecific paired wood blocks. With this experiment, we investigate how colonisation and competition between fungi may impact early wood decay dynamics. We predicted that: 1) There will be limited mass loss of the wood blocks by a single

133 species during early wood decomposition as the fungus establishes territory within the blocks, 2)

Early-stage decomposition is reduced by interspecific competition between interacting fungal pairs as fungi are unable to simultaneously optimise both resource extraction and combat, and 3) Biomass import will differ among fungal species.

5.3 Methods

5.3.1 Species selection

Our study species were aseptically cultured from rotting logs of Eucalyptus tereticornis taken from remnant Cumberland Plain woodland in Richmond, NSW (33°37'04.0"S 150°44'25.3"E) in

February 2016 (see Chan et al. 2021). With a flame sterilised chisel, we split the felled logs in the lab to expose the centre. With flame sterilised forceps, we extracted some slivers of wood and transferred these slivers of wood to 2% malt extract agar (MEA) and incubated the Petri dishes at

20°C. We reisolated from the emerging hyphal front from the wood onto new MEA to obtain a single fungal isolate. From pure cultures, we extracted DNA from the growing tips of the fungal colony with the DNeasy Plant Mini Kit (Qiagen, Chadstone, Victoria, Australia) as per the manufacturer’s instruction. We amplified the ITS (ITS1F & ITS4) region of rDNA (Gardes and Bruns 1993, Thompson et al. 2012) through PCR amplification. We analysed the amplicons using a ABI3500 Genetic Analyser

(Applied Biosystems, Life Technologies, Mulgrave, Victoria, Australia). Each isolated species was assigned a taxonomic identity by comparing their sequence against a BLAST search against the NCBI

Nucleotide database. We kept the isolates on 1% MEA at 4°C.

134

5.3.2 Ability to respire cellulose and lignin

We conducted a lignin degradation assay using a lignin-guaiacol-benomyl agar media (LGBA,

(Thorn et al. 1996)). 5 mm plugs of the fungal colonies were re-cultured onto Petri-dishes onto LGBA media. Colonies that produced laccase or peroxidase in the presence of lignin would produce a bright red ring due to the lignolytic activity (Thorn et al. 1996). Colonies able to degrade lignin were categorised as a lignin degrader, while colonies unable to degrade lignin were classified as a non- lignin degrader. From the pool of species tested, four were selected: two lignin degrading basidiomycetes: Omphalotus sp. (Om), Peniophora sp. (Pe), and two non-lignin degrading ascomycetes: Phacidium lacerum (Pl), and an unidentified Helotiales isolate (He).

We tested the respiration potential of our study species on cellulose with a modified

Microresp™ method ((Campbell et al. 2003), see Chapter 3). We added eight replicates of our study

species into 1.2 mL deep-well plates (Thermofisher, Australia) prefilled with H2O agar. We added 8 μl

of a 5% milled cellulose in dH2O as the carbon source. We included two sets of 4 replicates with

added dH2O and no added material to act as controls. We also included blank wells to measure the basal atmospheric change over the course of the assay. The deep-well plates were clamped together and sealed with a Microresp™ seal and a colorimetric detection plate. We made two absorbance readings at 48 and 96 hours of the colorimetric detection plates at 570 nm with a CLARIOstar plate reader (BMG LABTECH GmbH, Germany). We calculated biomass of the fungal inoculum using the grid intersect method (Giovannetti and Mosse 1980)) and measured the average hyphal diameter per inoculum to define hyphal volume. We converted this measure to mass by following the average biovolume mass measurement in (van Veen and Paul 1979). We measured the rate of respiration (μg

CO2/g/h) on cellulose by calculating the colour change in the colorimetric plates on cellulose minus the basal respiration rate of the fungal inoculum.

135

5.3.3 Establishing wood block interactions

We inoculated 8 cm3 blocks of Karri (Eucalyptus diversicolor) wood (2 cm x 2 cm x 2 cm) with our study species. Eucalyptus diversicolor is found in south-western Australia and is commonly used commercially for its wood (O’Connell 1988). Using the Leco C/N analyser (LECO TruMac Analyser), we determined the nitrogen percentage to be 0.106%. Potassium and phosphorus percentage was measured with the nitric acid digest method (Mills and Bryson 2015), but both were below the detectable threshold (<0.01). E. diversicolor has approximately 18% lignin by dry weight (Bland et al.

1950).

We gave each wood block a unique identity code and the blocks were oven dried at 100°C for 24 hours before each wood block was weighed and tracked for the duration of the experiment.

Wood blocks were hydrated overnight in MilliQ water and autoclaved three times over three days to sterilise the blocks. Prior to the start of the experiment, we kept the wood blocks at -20°C. To inoculate the wood blocks, 120 wood blocks were placed onto separate 2% MEA Petri dishes each colonised with each of our four study species (n=480) for three months (90 days) in the dark at 20°C.

An additional 15 uninoculated wood blocks were also kept in 100 mL plastic sauce pots with moistened perlite as a control.

To establish interspecific wood block pairs, all inoculated wood blocks were taken off from the agar plates and scraped clean of any surface mycelia with a sterilised scalpel. Under strict aseptic conditions, we arranged inoculated wood block interactions wood grain to wood grain and held it loosely with a sterile rubber band. There were a total of six pairwise interactions (OmHe, PeHe,

PeOm, PePl, PlHe, PlOm). The interacting wood blocks were then placed in a sterilised 100 mL plastic sauce pot (WF Plastic Pty Ltd, AU) containing 30 ml moistened sterile perlite and incubated at 20°C in the dark (Fig. 1). We drilled a small hole on the side of the pot and covered it with micropore tape

(3M, UK) for continued gas exchange. This was repeated for the intraspecific wood block pairs.

Individual wood blocks were placed alone in the sterilised pots. All wood block interactions were

136 kept in the dark at 20°C. Following the establishment of the interactions, wood block pairs were harvested at 7, 21, 42, 70, and 105 days. All harvested wood blocks had any adhering mycelium scraped free, unique code verified, and then weighed. Harvested wood blocks at different time points were immediately dried at 100°C for 2 days to prevent further decay. After all wood blocks were harvested, we placed all blocks in a drying oven at 100°C for 2 days and weighed to determine mass loss of each wood block. Previous work has established a competitive hierarchy of interaction strengths between the study species by placing the species in competition on wood (Chan et al.

2021). Interaction outcomes were determined by territory gain and loss over 3 months of

(A)

interaction. (B)

Figure 1. (A) Experimental set up

of wood block interactions. (B) Wood

block interaction (PeHe)

137

5.3.4 Statistical analysis

To test for change in wood block mass through time, we fitted a linear model (ANOVA) with study species (including the non-inoculated control as a “species”) and time as the independent variables and mass change in wood blocks as a dependent variable. We used Tukey's HSD post-hoc tests. From the results of the analysis, we were able to compare mass change in treatment wood blocks against the control wood blocks. To test how early-stage wood decomposition is affected by interspecific competition, we fitted separate ANOVAs for each interspecific species treatment. In this model, we had species treatment combination and time as the independent variables, with mass change in wood as the dependent variable. We used Tukey’s HSD post-hoc tests to compare mass change in interspecific treatment blocks against individually colonised wood blocks to see how competition affects mass change in wood. To rule out any effects of the addition of wood blocks on total mass loss, we fitted an ANOVA comparing mass change in individual wood blocks and wood blocks in the intraspecific interaction treatment. We included treatment and time as independent variables, and mass change in wood as the dependent variable. To estimate mass import in wood blocks at the start of the experiment following a three-month inoculation period, we fitted a generalised linear model (GLM) using the gaussian error family with species and time as the independent variables and mass change in wood as the dependent variable. We estimated mass import by calculating the intercept of the slopes at day 0 of the experiment, giving us an approximation of the mass of the wood blocks after the initial colonisation period. All analyses were run in R (v.4.0.2).

138

5.4 Results

5.4.1 Lag in early wood decay

We did not observe consistent mass loss following three months after an initial three-month inoculation period (total six months). Overall, there was a significant difference in mass in wood

blocks colonised by our different study species (ANOVA, F4,85=10.978, p<0.0001). Post hoc comparisons using the Tukey’s HSD test reveal that wood blocks colonised with Pl and He had significantly increased in mass by 2.27% (p<0.0001) and 2.6% (p=0.0003) on average respectively, while Pe and Om colonised wood blocks were not significantly different from the control blocks

(p=0.5785 and p=0.9152 respectively, Supplementary Material).

5.4.2 The effect of competition on mass change in wood blocks during early wood decay

We found that where there was competitive asymmetry, competition between species had a significant effect on the combined mass of wood blocks. Across all competitive treatments, time did not have a significant effect on wood block mass, and species pairing did not significantly interact with time (Supplementary Material). In the PeOm and PlHe treatments, the mass of wood blocks were not significantly different between individual or interspecific interaction treatments (ANOVA,

F2,54=0.421, p=0.772 and ANOVA, F2,54=0.852, p=0.6 respectively). When scoring competitive outcomes, the PeOm and PlHe treatments were competitively symmetric (Chan et al. 2021, Fig 2). In the interspecific competition treatments: PePl and PlOm, Pe and Om were competitively superior against their paired species (Chan et al. 2021, Fig 2). In these treatments, the mass of the wood blocks colonised by the competitively superior species was not significantly different when compared to the combined mass of wood blocks in the combined species treatment. The mass of the wood blocks of the competitively weaker species was significantly higher when compared to the combined mass of wood blocks in the combined species treatment. In the PePl treatment, there was

139 no significant difference in mass in wood blocks between the interspecific treatment and individual

Pe treatment blocks. But the individual Pl treatment blocks were significantly higher in mass than the interspecific PePl treatment blocks (Tukey’s HSD, p=0.0098). In the PlOm treatment, there was no significant difference in mass between the interspecific treatment wood blocks and the individual

Om blocks. The individual Pl blocks were also similarly higher in mass than the interspecific PlOm treatment blocks (Tukey’s HSD, p=0.0418). In the OmHe treatment, the Helotiales isolate was competitively superior (Chan et al. 2021, Fig 2), but competition did not significantly affect mass in the wood blocks between Om (Tukey’s HSD, p=0.0591) and He (Tukey’s HSD, p=0.0723). In the PeHe treatment, the Helotiales isolate was competitively superior (Chan et al. 2021, Fig 2), but the He- colonised wood block had significantly higher mass than the interspecific PeHe treatment blocks

(Tukey’s HSD, p=0.0224). There was no significant difference in mass between the individual Pe wood blocks and the interspecific PeHe treatment blocks. Intraspecific interactions in wood blocks had no significant effect on mass change in wood compared to individually colonised wood blocks alone (Supplementary Material).

5.4.3 Biomass import in early wood decay

We found that the estimated biomass import into the wood blocks at day 0 of the experiment following a three-month inoculation period significantly differed between species overall

(ANOVA, F3,155=19.609, p<0.0001, Fig. 3). Post hoc comparisons with the Tukey’s HSD test showed that wood blocks colonised with Pe and wood blocks colonised with Om had significantly lower wood mass than wood blocks colonised with He (p<0.0001 & p<0.0001 respectively, Table 1) and wood blocks colonised with Pl (p<0.0001 & p=0.0038 respectively, Table 1). The mass of wood blocks colonised by Pe and Om did not significantly differ from each other (p=0.2076, Table 1), and the mass of wood blocks colonised by He and Pl were also not significantly different from each other (p=

0.4968, Table 1).

140

Figure 2. Corrected mass change in wood (against control). Time points (7, 21, 42, 70, 105 days) pooled as there was no significant effect of time on mass change. Axes standardised across graphs. Letters denote significant differences between treatments. Asterisks denote superior competitors from Chan et al. 2021.

141

Table 1. Tukey’s HSD post-hoc test comparing estimated biomass at time point 0 relative to control blocks. Omphalotus sp. (Om) and Peniophora sp. (Pe) were not significantly different. Phacidium lacerum (Pl) and the Helotiales isolate (He) were also not significantly different.

Tukey’s HSD

p Om – Pe 0.2076 Pl – Pe <0.0001 He – Pe <0.0001 Pl – Om 0.0038 He – Om <0.0001 He – Pl 0.4968

Figure 3. Estimated biomass at time point 0 relative to control blocks. Time point 0 was estimated by calculating the intercept of the slopes of the GLM at day 0 and is defined as when the colonisation from the agar plate finished and the experimental decomposition began. Data from all blocks colonized with each species (regardless of which treatment they ended up in) were normalized relative to the control blocks then fit with a linear model, with the intercept at time 0; the plots show the estimated intercept and standard error of that estimate. Phacidium lacerum (Pl) and the Helotiales isolate (He) had significantly higher estimated net mass import relative to Omphalotus sp. (Om) and Peniophora sp. (Pe) and the control wood blocks. Neither Pl and He have the capacity to degrade lignin, unlike Om and Pe. The estimated mass at time 0 of Om and Pe colonised wood blocks were not significantly different from the control blocks.

142

5.5 Discussion

5.5.1 Lag in early wood decay

The ecology of saprotrophic fungi is a vital component in the decay of wood. As fungi gain territory within wood, the extraction of resources may slow in the presence of many interacting elements. In line with the lag phase observed in the field, we did not observe any appreciable level of mass loss across individual or interspecific treatment combinations compared to uncolonised wood blocks after a combined 6-month period (Fig. 2). A 6-month residence time for saprotrophic fungi in wood is comparatively short and represents an early phase in wood decay (Kahl et al. 2017).

Here, we find that even in a controlled experimental setting, wood decay was still remarkably slow.

Lag phases in wood decay are commonly observed in field experiments (Harmon et al. 2000), and microbial colonisation time has been used as a possible explanation for this delay (Freschet et al.

2012). Previous experiments have shown that successful early establishment within wood by fungi can have long-lasting effects on wood decay (Fukami et al. 2010, Dickie et al. 2012). When an obligate saprotrophic fungus reaches a new substrate, colonisation of new territory within the substrate is a proxy for isolating the resources therein from eventual competitors (Boddy and Hiscox

2016). In this arena, we would predict that resource allocation strategies employed by fungi should favour rapid colonisation upon initial entry (Cadotte et al. 2006, Chan et al. 2020). Increased rates of colonisation within wood would benefit a fungus by increasing territory within wood, thus increasing the amount of available resources, and denying it from other fungi (Ghoul and Mitri 2016).

Biomass import may also play a significant role in this lag in decay. In our experiment, we found that after the initial colonisation phase, both non-lignin degrading species had higher estimated biomass import compared to the lignin degrading basidiomycetes (Fig. 3). In the case of the He and Pl colonised wood blocks, there was a 2% increase in wood block mass following initial colonisation. While we observed this difference in biomass import, it is important to note that there

143 may have been comparable levels of biomass import, but the lignin degrading species may have begun utilising available resources, thus driving down mass. Fungi have the ability to translocate nutrients across their mycelia, allowing them to import vital nutrients into the relatively resource- poor environment within wood (Tlalka et al. 2008). The injection of nutrients into the wood may precondition the wood for rapid decay if the fungus is enzymatically capable of breaking down the recalcitrant lignocellulose that forms much of the available resource (Swift 1977, Philpott et al.

2014). In fungi without the ability to break down the lignocellulose complex into more labile nutrients, imported nutrients may aid in the development of defensive structures in the highly combative environment within wood while opportunistic fungi absorb resources freed by other wood decay fungi (Boddy 2000).

5.5.2 The effect of fungal competition on wood decay

Interspecific competition had a context specific effect on mass loss in wood, determined by the outcome of competition. We found that in two of the interspecific treatments (PePl, PlOm), the superior competitor had a greater influence on the mass change of wood blocks (Fig. 2). In fungal competition within wood, a superior competitor can competitively replace a weaker neighbour and gain access to the previously defended resources (Boddy and Hiscox 2016). In our experiment, we observed the superior competitor in an interspecific pair replacing the previously colonised wood block of the weaker species (Chan et al. 2021). It stands to reason that the species occupying the highest proportion of territory will have a greater effect on the resource extraction and rate of decay of the paired wood blocks. In symmetric competition where neither species had a competitive advantage, we found no strong signal of competition on change in mass in wood blocks (PeOm, PlHe,

Fig. 2). Unlike asymmetric competition where one species is able to replace the other, the deadlock between interacting species where neither can gain headway against the other prevents any one species from exerting more influence on the trajectory of decay. In two of the interspecific

144 treatments (PeHe, OmHe), the Helotiales isolate failed to influence mass change in the paired wood blocks despite marginally replacing both of the weaker competitors in the interspecific pairs (Fig. 2).

This may be due to the inability of the Helotiales isolate to degrade lignin, reducing its capacity to decay wood.

In wood, both competition and resource extraction are energetically costly (Hiscox et al.

2015a). While initial entry into wood would favour rapid colonisation, competition is inevitable. To persist within wood, a fungus needs to utilise available resources to detoxify and break down the antagonistic enzymes used in fungal competition (Arfi et al. 2013). As such, competition incurs a metabolic cost and influences resource use in fungi. In early wood decay, where rapid colonisation is so important (Holmer and Stenlid 1993), the cost of competition may alter fungal metabolism and change the rate of decomposition (Arfi et al. 2013, Moeller and Peay 2016). Failure of a fungus to adequately defend against competitors will inevitably result in a loss of held territory and resources

(Hiscox et al. 2015a), and superior competitors can greatly influence the rate of decay by taking over a higher proportion of territory within wood. Ultimately, wood represents a combined resource pool, and all interactions between saprotrophic fungi collectively deplete the available resources.

But as different fungal species have different metabolic rates and rates of decay, fungi that can persist after initial entry and colonisation can have a greater impact on the trajectory of decay.

5.5.3 Priority effects and implications on early wood decay

The struggle for persistence in the earliest stages of wood decay involves the early establishment of pioneer species, but the assembly sequence of species can have far reaching implications for later decomposition (Fukami et al. 2010, Dickie et al. 2012). Critically, latent or early arriving species able to establish a territory in newly dead wood can grow quickly to exhaust easily accessible nutrients before slowly degrading more recalcitrant substances (van der Wal et al. 2015).

145

Rapid growth during initial colonisation may be bolstered by higher rates of nutrient translocation which may allow a species to attain more territory and resources (Hiscox et al. 2015b). Here, the combination of high growth rate and nutrient translocation rate may be important fungal traits that affect the trajectory of wood decay via priority effects, where the early arrival of competitively superior species can inhibit the establishment of later arrivals (Heilmann-Clausen and Boddy 2005, van der Wal et al. 2015, Hiscox et al. 2015b).

There has been strong experimental evidence of priority effects on the rate of wood decay

(Fukami et al. 2010, Dickie et al. 2012, van der Wal et al. 2015), but how the mechanisms of competition shape priority effects are still poorly understood (Hiscox et al. 2015b). While our experiment did not set out to explicitly study priority effects, we began to look at how certain fungal traits interact with competition during early wood decay. Fungal traits that are associated with higher competitive ability along with traits associated with colonisation and the ability to degrade lignocellulose may allow a species to persist during the strong interspecific interactions that dominate the early stage of wood decay (Lustenhouwer et al. 2020). Together with established wood traits such as density (Radtke et al. 2009), nitrogen concentration (Hale and Pastor 1998), and anatomical traits (Cornwell et al. 2009) that predict longer residence times, we can begin to move towards a trait-based understanding of wood decay by considering all the components present in wood decomposition by saprotrophic fungi.

5.5.4 Conclusion

Understanding the fluxes in carbon at a global scale is becoming increasingly important

(Cornwell et al. 2009, Rinne-Garmston et al. 2019, Oberle et al. 2020), with the interactions between saprotrophs and dead plant material still representing a dearth in our knowledge. While models of carbon movement in the carbon cycle is our best approach to tackling this issue, there is an urgent

146 need to improve these models with experimental data on the effect of microbial interactions on wood decay (Venugopal et al. 2017). In our experiment, we explored how the ecology of saprotrophic fungi may affect early wood decay. We highlighted how observed lags in wood decay may be attributable to the ecological needs of saprotrophic fungi. But further experimental approaches will be necessary to improve our understanding of the mechanistic underpinnings of fungal dynamics in wood. Uncovering how the ecology of individual fungi within a larger interacting community shapes the trajectory of wood decay may help us understand how these mechanisms scale up to larger ecosystem processes.

147

5.6 References

Arfi, Y. et al. 2013. Differential gene expression in Pycnoporus coccineus during interspecific mycelial interactions with different competitors. - Appl. Environ. Microbiol. 79: 6626–6636.

Arnold, A. E. et al. 2003. Fungal endophytes limit pathogen damage in a tropical tree. - Proc. Natl. Acad. Sci. U. S. A. 100: 15649–15654.

Bland, D. E. et al. 1950. Aromatic Aldehydes from the Oxidation of Some Australian Woods and their Chromatographic Separation. - Aust. J. Chem. 3: 642–648.

Boddy, L. 1999. Saprotrophic Cord-Forming Fungi: Meeting the Challenge of Heterogeneous Environments. - Mycologia 91: 13–32.

Boddy, L. 2000. Interspecific combative interactions between wood-decaying basidiomycetes. - FEMS Microbiol. Ecol. 31: 185–194.

Boddy, L. 2001. Fungal Community Ecology and Wood Decomposition Processes in Angiosperms: From Standing Tree to Complete Decay of Coarse Woody Debris. - Ecol. Bull.: 43–56.

Boddy, L. and Hiscox, J. 2016. Fungal Ecology: Principles and Mechanisms of Colonization and Competition by Saprotrophic Fungi. - Microbiol Spectr in press.

Boswell, G. P. et al. 2002. Functional consequences of nutrient translocation in mycelial fungi. - J. Theor. Biol. 217: 459–477.

Cadotte, M. W. et al. 2006. On testing the competition-colonization trade-off in a multispecies assemblage. - Am. Nat. 168: 704–709.

Campbell, C. D. et al. 2003. A rapid microtiter plate method to measure carbon dioxide evolved from carbon substrate amendments so as to determine the physiological profiles of soil microbial …. - Applied and in press.

Chan, J. Y. et al. 2020. Environmental cues for dispersal in a filamentous fungus in simulated islands. - Oikos 129: 1084–1092.

Chan, J. et al. 2021. Complex environments alter competitive dynamics in fungi. in press.

148

Cornwell, W. K. and Weedon, J. T. 2014. Decomposition trajectories of diverse litter types: a model selection analysis (J Oksanen, Ed.). - Methods Ecol. Evol. 5: 173–182.

Cornwell, W. K. et al. 2008. Plant species traits are the predominant control on litter decomposition rates within biomes worldwide. - Ecol. Lett. 11: 1065–1071.

Cornwell, W. K. et al. 2009. Plant traits and wood fates across the globe: rotted, burned, or consumed? - Glob. Chang. Biol. 15: 2431–2449.

Dickie, I. A. et al. 2012. Do assembly history effects attenuate from species to ecosystem properties? A field test with wood-inhabiting fungi. - Ecol. Lett. 15: 133–141.

Franklin, J. F. et al. 1987. Tree Death as an Ecological Process. - Bioscience 37: 550–556.

Freschet, G. T. et al. 2012. Interspecific differences in wood decay rates: insights from a new short-term method to study long-term wood decomposition: New method to assess wood decay dynamics and rates. - J. Ecol. 100: 161–170.

Fukami, T. 2015. Historical Contingency in Community Assembly: Integrating Niches, Species Pools, and Priority Effects. - Annu. Rev. Ecol. Evol. Syst. 46: 1–23.

Fukami, T. et al. 2010. Assembly history dictates ecosystem functioning: evidence from wood decomposer communities. - Ecol. Lett. 13: 675–684.

Gardes, M. and Bruns, T. D. 1993. ITS primers with enhanced specificity for basidiomycetes--application to the identification of mycorrhizae and rusts. - Mol. Ecol. 2: 113–118.

Ghoul, M. and Mitri, S. 2016. The Ecology and Evolution of Microbial Competition. - Trends Microbiol. 24: 833–845.

Giovannetti, M. and Mosse, B. 1980. An Evaluation of Techniques for Measuring Vesicular Arbuscular Mycorrhizal Infection in Roots. - New Phytol. 84: 489–500.

Hale, C. M. and Pastor, J. 1998. Nitrogen content, decay rates, and decompositional dynamics of hollow versus solid hardwood logs in hardwood forests of Minnesota, USA. - Can. J. For. Res. in press.

Harmon, M. E. et al. 2000. Decomposition vectors: a new approach to estimating woody detritus decomposition dynamics. - Can. J. For. Res. 30: 76–84.

149

Heilmann-Clausen, J. and Boddy, L. 2005. Inhibition and stimulation effects in communities of wood decay fungi: exudates from colonized wood influence growth by other species. - Microb. Ecol. 49: 399–406.

Hiscox, J. et al. 2015a. Antagonistic fungal interactions influence carbon dioxide evolution from decomposing wood. - Fungal Ecol. 14: 24–32.

Hiscox, J. et al. 2015b. Priority effects during fungal community establishment in beech wood. - ISME J. 9: 2246–2260.

Hiscox, J. et al. 2016. Effects of pre-colonisation and temperature on interspecific fungal interactions in wood. - Fungal Ecol. 21: 32–42.

Holmer, L. and Stenlid, J. 1993. The importance of inoculum size for the competitive ability of wood decomposing fungi. - FEMS Microbiol. Ecol. 12: 169–176.

Holmer, L. and Stenlid, J. 1997. Competitive Hierarchies of Wood Decomposing Basidiomycetes in Artificial Systems Based on Variable Inoculum Sizes. - Oikos 79: 77–84.

Kahl, T. et al. 2017. Wood decay rates of 13 temperate tree species in relation to wood properties, enzyme activities and organismic diversities. - For. Ecol. Manage. 391: 86–95.

Lee, M. R. et al. 2019. Good neighbors aplenty: fungal endophytes rarely exhibit competitive exclusion patterns across a span of woody habitats. - Ecology 100: e02790.

Lustenhouwer, N. et al. 2020. A trait-based understanding of wood decomposition by fungi. - Proc. Natl. Acad. Sci. U. S. A. 117: 11551–11558.

Mills, H. A. and Bryson, G. 2015. Plant Analysis Handbook IV. in press.

Moeller, H. V. and Peay, K. G. 2016. Competition-function tradeoffs in ectomycorrhizal fungi. - PeerJ 4: e2270.

Oberle, B. et al. 2020. Accurate forest projections require long-term wood decay experiments because plant trait effects change through time. - Glob. Chang. Biol. 26: 864–875.

O’Connell, A. M. 1988. Nutrient Dynamics in Decomposing Litter in Karri (Eucalyptus Diversicolor F. Muell.) Forests of South-Western Australia. - J. Ecol. 76: 1186–1203.

150

Parfitt, D. et al. 2010. Do all trees carry the seeds of their own destruction? PCR reveals numerous wood decay fungi latently present in sapwood of a wide range of angiosperm trees. - Fungal Ecol. 3: 338– 346.

Philpott, T. J. et al. 2014. Nitrogen translocation and accumulation by a cord-forming fungus (Hypholoma fasciculare) into simulated woody debris. - For. Ecol. Manage. 315: 121–128.

Radtke, P. J. et al. 2009. Modeling production and decay of coarse woody debris in loblolly pine plantations. - For. Ecol. Manage. 257: 790–799.

Rinne-Garmston, K. T. et al. 2019. Carbon flux from decomposing wood and its dependency on temperature, wood N2 fixation rate, moisture and fungal composition in a Norway spruce forest. - Glob. Chang. Biol. 25: 1852–1867.

Rock, J. et al. 2008. Estimating decomposition rate constants for European tree species from literature sources. - Eur. J. For. Res. 127: 301–313.

Rodriguez, R. J. et al. 2009. Fungal endophytes: diversity and functional roles. - New Phytol. 182: 314– 330.

Schimel, D. S. 1995. Terrestrial ecosystems and the carbon cycle. - Glob. Chang. Biol. 1: 77–91.

Sieber, T. N. 2007. Endophytic fungi in forest trees: are they mutualists? - Fungal Biol. Rev. 21: 75–89.

Swift, M. J. 1977. The Roles of Fungi and Animals in the Immobilisation and Release of Nutrient Elements from Decomposing Branch-Wood. - Ecol. Bull.: 193–202.

Thompson, T. A. et al. 2012. Hypholoma lateritium isolated from coarse woody debris, the forest floor, and mineral soil in a deciduous forest in New Hampshire. - Botany 90: 457–464.

Thorn, R. G. et al. 1996. Isolation of saprophytic basidiomycetes from soil. - Appl. Environ. Microbiol. 62: 4288–4292.

Tlalka, M. et al. 2008. Quantifying dynamic resource allocation illuminates foraging strategy in Phanerochaete velutina. - Fungal Genet. Biol. 45: 1111–1121. van der Wal, A. et al. 2015. Neglected role of fungal community composition in explaining variation in wood decay rates.

151 van Veen, J. A. and Paul, E. A. 1979. Conversion of biovolume measurements of soil organisms, grown under various moisture tensions, to biomass and their nutrient content. - Appl. Environ. Microbiol. 37: 686–692.

Venugopal, P. et al. 2017. Assemblage composition of fungal wood-decay species has a major influence on how climate and wood quality modify decomposition. - FEMS Microbiol. Ecol. in press.

Weedon, J. T. et al. 2009. Global meta-analysis of wood decomposition rates: a role for trait variation among tree species? - Ecol. Lett. 12: 45–56.

Wells, J. M. and Boddy, L. 1995. Phosphorus translocation by saprotrophic basidiomycete mycelial cord systems on the floor of a mixed deciduous woodland. - Mycol. Res. 99: 977–980.

Wells, J. M. and Boddy, L. 2002. Interspecific carbon exchange and cost of interactions between basidiomycete mycelia in soil and wood: Carbon cost of fungal interactions. - Funct. Ecol. 16: 153– 161.

Worrall, J. J. et al. 1997. Comparison of wood decay among diverse lignicolous fungi. - Mycologia 89: 199– 219.

Zell, J. et al. 2009. Predicting constant decay rates of coarse woody debris—A meta-analysis approach with a mixed model. - Ecol. Modell. 220: 904–912.

152

5.7 Supplementary Material

Supplementary Table 1. Analysis of variance (ANOVA) with species (including control) with time in days as independent variables, with wood block mass as the dependent variable. Time did not significantly affect wood mass. Species was found to have a significant effect on wood mass overall.

ANOVA

DF SS Mean Sq F P Combination 4 98.81 24.702 10.978 <0.0001* Day 1 1.56 1.559 0.693 0.408 Combination:Day 4 4.86 1.214 0.540 0.707 Residuals 85 191.27 2.250

Supplementary Table 2. Tukey’s HSD post-hoc test comparisons of species (including control) effect on mass of wood blocks. Pl and He were significantly different from the control, Om, and Pe. Om and Pe were not significantly different from the control wood blocks.

Tukey’s HSD

p Om – Control 0.9152 Pe – Control 0.5785 Pl – Control 0.0003 He – Control <0.0001 Pe – Om 0.9595 Pl – Om 0.0019 He – Om <0.0001 Pl – Pe 0.017 He – Pe 0.0018 He – Pl 0.9541

153

Supplementary Table 3. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for PePl. Time did not significantly affect wood block mass. Treatment had an overall significant effect on wood block mass.

ANOVA PePl

DF SS Mean Sq F P Day 1 0.08 0.083 0.028 0.8675 Combination 2 33.76 16.866 5.680 0.0058 Combination:Day 2 1.47 0.733 0.247 0.7821 Residuals 54 16.034 2.969

Supplementary Table 4. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for PePl. Pl was significantly different than Pe alone and the combined PePl treatment.

Tukey’s HSD PePl

p PePl – Pe 0.9577 Pl – Pe 0.02 Pl – PePl 0.0098

Supplementary Table 5. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for PlOm. Time did not significantly affect wood block mass. Treatment had an overall significant effect on wood block mass.

ANOVA PlOm

DF SS Mean Sq F P Day 1 6.96 6.956 2.413 0.1261 Combination 2 35.93 17.967 6.233 0.0037 Day:Combination 2 7.77 3.886 1.348 0.2683 Residuals 54 155.66 2.883

154

Supplementary Table 6. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for PlOm. Pl was significantly different than Om alone and the combined PlOm treatment.

Tukey’s HSD PlOm

p Pl – Om 0.0034 PlOm – Om 0.625 PlOm – Pl 0.0418

Supplementary Table 7. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for PlHe. Time and treatment did not significantly affect wood block mass.

ANOVA PlHe

DF SS Mean Sq F P Day 1 0.51 0.5113 0.379 0.541 Combination 2 1.39 0.6966 0.516 0.600 Day:Combination 2 2.30 1.1496 0.852 0.432 Residuals 54 72.84 1.3489

Supplementary Table 8. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for PlHe. There were no significant differences in wood block mass between treatments.

Tukey’s HSD PlHe

p Pl – He 0.6336 PlHe – He 0.6793 PlHe – Pl 0.9970

155

Supplementary Table 9. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for OmHe. Time did not significantly affect wood block mass. Treatment had an overall significant effect on wood block mass.

ANOVA OmHe

DF SS Mean Sq F P Day 1 0.12 0.124 0.056 0.8146 Combination 2 47.05 23.525 10.512 <0.0001 Day:Combination 2 11.71 5.855 2.616 0.0823 Residuals 54 120.84 2.238

Supplementary Table 10. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for OmHe. Om was significantly different from He. The combined OmHe was not significantly different from Om or He.

Tukey’s HSD OmHe

p Om – He <0.0001 OmHe – He 0.0723 OmHe – Om 0.0591

Supplementary Table 11. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for PeOm. Time and treatment did not significantly affect wood block mass.

ANOVA PeOm

DF SS Mean Sq F P Day 1 2.12 2.117 0.566 0.455 Combination 2 1.95 0.973 0.260 0.772 Day:Combination 2 3.15 1.575 0.421 0.659 Residuals 54 202.14 3.743

156

Supplementary Table 12. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for PeOm. There were no significant differences in wood block mass between treatments.

Tukey’s HSD PeOm

p Pe – Om 0.8569 PeOm – Om 0.9862 PeOm – Pe 0.7712

Supplementary Table 13. Analysis of variance (ANOVA) with treatment and time as the independent variables and wood block mass as the dependent variable for PeHe. Time did not significantly affect wood block mass. Treatment had an overall significant effect on wood block mass.

ANOVA PeHe

DF SS Mean Sq F P Day 1 0.46 0.456 0.165 0.6863 Combination 2 37.59 18.795 6.798 0.0023 Day:Combination 2 0.23 0.114 0.041 0.9597 Residuals 54 149.30 2.765

Supplementary Table 14. Tukey’s HSD post-hoc test comparing treatment effect on wood block mass over time for PeHe. He was significantly different than Pe alone and the combined PeHe treatment.

Tukey’s HSD PeHe

p Pe – He 0.0026 PeHe – He 0.0224 PeHe – Pe 0.7225

157

Supplementary Table 15. Analysis of variance (ANOVA) with time and species as independent variables and estimated nutrient import as the dependent variable. Day did not significantly affect estimated import. Species significantly affect estimated nutrient import.

ANOVA

DF SS Mean Sq F P Day 1 5.9 5.87 1.607 0.207 Species 3 214.7 71.57 19.609 <0.0001 Residuals 155 565.7 3.65

Supplementary Table 16. Analysis of variance (ANOVA) comparing wood block mass between intraspecific treatment and individual treatment for Peniophora sp.. Treatment and time as independent variables and wood block mass as dependent variables. There were no significant differences in wood block mass between intraspecific and individual treatments.

ANOVA Pe

DF SS Mean Sq F P Day 1 2.4 2.39 0.243 0.6249 Combination 1 37.1 37.14 3.781 0.0597 Day:Combination 1 1.5 1.51 0.154 0.6969 Residuals 36 353.6 9.82

Supplementary Table 17. Analysis of variance (ANOVA) comparing wood block mass between intraspecific treatment and individual treatment for Helotiales isolate. Treatment and time as independent variables and wood block mass as dependent variables. There were no significant differences in wood block mass between intraspecific and individual treatments.

ANOVA He

DF SS Mean Sq F P Day 1 2.571 2.5706 3.490 0.0699 Combination 1 0.486 0.4855 0.659 0.4222 Day:Combination 1 0.253 0.2526 0.343 0.5617 Residuals 36 26.513 0.7365

158

Supplementary Table 18. Analysis of variance (ANOVA) comparing wood block mass between intraspecific treatment and individual treatment for Phacidium lacerum. Treatment and time as independent variables and wood block mass as dependent variables. There were no significant differences in wood block mass between intraspecific and individual treatments.

ANOVA Pl

DF SS Mean Sq F P Day 1 0.02 0.0235 0.018 0.894 Combination 1 0.96 0.9619 0.731 0.398 Day:Combination 1 0.74 0.7381 0.561 0.459 Residuals 36 47.36 1.3156

Supplementary Table 19. Analysis of variance (ANOVA) comparing wood block mass between intraspecific treatment and individual treatment for Omphalotus sp.. Treatment and time as independent variables and wood block mass as dependent variables. There were no significant differences in wood block mass between intraspecific and individual treatments.

ANOVA Om

DF SS Mean Sq F P Day 1 3.42 3.416 1.351 0.253 Combination 1 1.57 1.572 0.621 0.436 Day:Combination 1 1.91 1.910 0.755 0.391 Residuals 36 91.07 2.530

159

CHAPTER 6

General Conclusions

Justin Y. Chan

160

6.1 Fungal life history in the context of declining resource patches

Fungi are ubiquitous and a vital component of the ecosystems they inhabit, but our understanding of the ecological needs and processes of these organisms are limited. Unfortunately, the dearth of fungal ecology theory has limited our collective attempts at making predictions about the basic life history of fungi (Gilchrist et al. 2006). The limitations set by the lack of theory have in turn stymied our understanding of other key ecological processes such as fungal community development (Bruns 2019). Other elements of fungal ecology such as the effects of competition on ecosystem functions has received recent attention (Hiscox et al. 2015, Maynard et al. 2017,

Lustenhouwer et al. 2020), but the role of dispersal in the life history of fungi remains understudied

(Cadotte et al. 2006, Fuhrman 2009, Nemergut et al. 2013, Lancaster and Downes 2017).

In this thesis, I explore how the life history strategies for saprotrophic fungi are influenced by the episodic nature of the declining resource patches that they inhabit. Here, I found that competition over available and remaining resources, and dispersal in search for new resource patches before patch collapse are important parts of the life history of saprotrophic fungi. For saprotrophic fungi, resources patches form randomly across space, but predictably through time, within the environment (Jonsson et al. 2005). As the available resources in these patches inevitably draw down as a result of fungal metabolism, competition is inevitable, and dispersal is necessary for the continued persistence of a fungus at both a local patch scale and greater landscape scale (Norros et al. 2012). Understanding how allocation to dispersal and competition shift in response to environmental changes may allow us to predict how saprotrophic fungi will respond to an uncertain future.

161

6.2 The need to disperse

As largely sessile organisms, most saprotrophic fungal species have an airborne dispersal phase to search for new resources in the environment (Edman et al. 2004). Saprotrophic fungi inhabit finite resource patches, and dispersal is a vital ecological process to escape a resource patch before complete resource drawdown. Having evolved under these variable environmental conditions, fungi would need to respond appropriately to cues in the environment in anticipation for dispersal. In the studies presented in Chapters 2 and 3, I explored how allocation to dispersal in

Phacidium lacerum was affected by changes in the environment, chiefly by examining how dispersal allocation changed in response to both the size and nutrient concentration of a resource patch, and to the presence of competitors. The results from these studies highlighted that resource allocation to dispersal changes in response to the local environment. Central to the findings from both of these studies is the concept of informed dispersal. Informed dispersal theory posits that organisms can take information about their external environment and internal condition to disperse if the benefits of dispersal outweigh the costs of remaining stationary (Clobert et al. 2009). In experiments with P. lacerum, this manifested as an upregulation or downregulation in the production of pycnidia in response to cues from the environment.

In Chapter 2, I focused on how the quality of a patch altered both the timing and amount of pycnidia (asexual fruiting bodies) produced. I found that P. lacerum responded to both absolute nutrient level and patch size to affect its allocation to dispersal. This is in line with our original hypothesis that P. lacerum can detect cues in the environment to alter their pattern of resource allocation to dispersal. In our experiment, P. lacerum maximised growth prior to an upregulation of pyncidia production following detection of the patch edge on smaller Petri-dishes. This response was expected as the absence of room for continued growth allowed a total allocation switch to the production of pycnidia. When the optimal dispersal response is modelled, this “bang-bang” pattern

162 of dispersal allocation is predicted (Gilchrist et al. 2006). However, on larger Petr-dishes, P. lacerum would begin the production of pycnidia prior to the cessation of growth. Here, the early allocation to dispersal may be triggered if the P. lacerum colonies reach a minimum resource threshold for dispersal (Kozłowski 1992). From the results of this experiment, it would suggest that both the resource concentration and size of a patch may act as cues and are integrated to trigger a dispersal allocation response in P. lacerum.

Where intense competition was included, I did not observe the model-congruent patterns of optimal dispersal allocation. In Chapter 3, I tested how P. lacerum altered its dispersal allocation in response to competition. While I initially hypothesised that P. lacerum would increase dispersal allocation to avoid competition, I found the opposite. Where there was no change to the pattern of allocation to dispersal when P. lacerum was faced with a genetic clone of itself, interspecific competition saw a decrease in total dispersal allocation when facing Omphalotus sp. and Isolate

HWK5. While the production of pycnidia was reduced in response to competition in this experiment, the results are still in line with informed dispersal theory, with interspecific competitors triggering a reduced allocation to dispersal (Clobert et al. 2009). Where model predictions would suggest that maximising growth followed by a total resource allocation switch to dispersal would maximise spore production (Gilchrist et al. 2006), allocation only to dispersal or growth in the face of competition may limit a fungus from adequately defending itself from attack. The reduction of allocation to dispersal in P. lacerum in response to interspecific competition was likely due to the induced defence response observed (Agrawal 1999). Critically, the allocation towards defence reduced resources available for dispersal. The upregulation of defence against competitors may be a strategy to maximise dispersal by utilising the available resources in a patch before a fungus is competitively displaced (Agrawal 1999, Heaton et al. 2016).

163

The results from Chapters 2 and 3 suggest that trade-offs between growth, dispersal, and defence may underpin life history strategies when optimising finite resources available within patches. While the experiments here take advantage of a model species able to produce asexual fruiting bodies in-vitro, it is important to note that fungal life cycles are incredibly variable and there are limitations to the broader inferences made from these two studies. Nonetheless, the conceptual model of the trade-offs between growth, dispersal and defence presented in this thesis are transferable to other saprotrophic fungi who inhabit finite resource patches. Trade-offs are central to the ecology of all organisms when maximising fitness under limiting resources (Stearns 1992,

2000). Specifically with saprotrophic fungi, dispersal represents a vital ecological process to persist amongst resource patches that inevitably collapse as a result of saprotrophic metabolism (Boddy and Hiscox 2016). For saprotrophic fungi, the optimal allocation of resources must be informed by changes in the environment, with trade-offs striking a delicate balance dependent on the current resource level of the patch and presence of neighbouring competitors.

6.3 Fungus eat fungus world

Due to the nature of saprotrophic fungi, competition for resources is simultaneously competition for space (Hiscox et al. 2015). Thus, competition between fungi primarily manifests as competition over territory within resource patches, where the superior competitor can gain access to all available resources within the space colonised by the fungus (Boddy and Hiscox 2016). Within the context of finite resource patches that saprotrophic fungi reside, failure to adequately defend against attack would result in localised extinction prior to successful dispersal. But the nature of competition between species is not constant, and heterogeneity within the environment can affect the outcome of competition (Gause 1934). In Chapters 4 and 5, I investigated how the environment

164 can affect competition between fungi, and how competition between fungi can affect the process of decomposition during early wood decay. I found that competitive dynamics between fungi are affected by the substrate where competition takes place, and that the ecological needs for saprotrophic fungi may influence the progression of wood decay shortly after initial fungal colonisation. While the use of simple substrates in lab settings can provide a wealth of information about the interactions between saprotrophic fungi, testing competitive dynamics and examining fungal life history on ecologically relevant substrates provide an opportunity to understand how patterns observed in controlled experiments may scale out to more natural settings. Here, I took advantage of a paired wood block experiment to conduct two studies to observe how the environment can affect competition between saprotrophic fungi, and to observe the reciprocal impact of competition on the early stage of wood decay of the wood blocks.

To explore how the environment can influence competition between saprotrophic fungi, I tested how competitive dynamics shifted between a simple substrate and a complex substrate. In

Chapter 4, I compared the pairwise competitive outcomes of four saprotrophic fungi on both malt extract agar and wood blocks. I predicted that there would be a difference in the competitive dynamics between the two substrates owing to the increase in habitat complexity in wood compared to agar. In line with my predictions, the competitive dynamics established on agar was different to the competitive dynamics established in wood. This disparity was likely due to the many differences between the two substrates. Where labile nutrients were homogeneously distributed within agar across a largely two-dimensional surface, nutrients within wood were locked within the recalcitrant lignocellulose complex, with microstructural heterogeneity spread through the three- dimensional wood block. When comparing competitive outcomes in wood, the differences between simple and complex substrates may have overwhelmed any competitive ability between species, attenuating the competitive asymmetry that may be more pronounced on agar. The results from this

165 study show that nutritive and structural complexity within a resource can alter the competitive dynamics established on simpler substrates.

In Chapter 5, I explored how competition between saprotrophic fungi may affect resource use. From the same experiment established in Chapter 4, I compared mass loss in wood blocks established from the paired competitive interactions. Due to the metabolic cost of both competition and resource extraction for saprotrophic fungi, I predicted there would be reduced mass loss where there was interspecific competition compared to wood blocks in the absence of interaction. Contrary to my predictions, I did not observe any notable mass loss across interspecific competition treatments, and paradoxically, observed a gain in mass instead. This was likely due to biomass import during the initial colonisation of wood blocks prior to the start of the experiment. Here, the ecology of saprotrophic fungi may have influenced the progression of wood decay. Upon initial entry into a virgin substrate, saprotrophic fungi need to establish a territory within the resource patch, thereby isolating available nutrients from competitors (Boddy and Hiscox 2016). While fungi arriving via spore deposition have limited nutrient reserves from the spores themselves, fungi arriving via mycelium are able to translocate external nutrients via the hyphae, gaining access to a greater pool of nutrients within the substrate (Fricker et al. 2008). Comparing both modes of colonisation, entry via mycelium provides a competitive advantage in the successful establishment of territory. In this experiment, the wood blocks were inoculated via fungal culture on agar, allowing the fungi to colonise via mycelia, and consequently allowing nutrient import into the wood blocks. While I initially predicted that competition would reduce decay rate as a fungus would trade-off allocation to defence vs. allocation to resource extraction, the results from this study point to this critical colonisation phase alongside competition that may alter how resources are used within a resource patch during early decay in wood.

166

Saprotrophic fungal communities are structured by both positive and negative interactions

(Niemelä et al. 1995, Boddy 2001, Lee et al. 2019), but how the environment affects these interactions remains poorly understood. In Chapter 4, I found that the complexity of a substrate can alter competitive dynamics between interacting fungi. This result suggests that the environment may have a role in preventing dominant competitors from excluding weaker competitors within saprotrophic fungal communities, manifesting in the rich diversity observed with resource patches

(Mäkipää et al. 2017). But any single species is unable to persist indefinitely regardless of competitive ability (Boddy 2001, van der Wal et al. 2013). As a saprotrophic fungus extracts resources from a patch, the quality of the environment will change, applying a filter on the species best fit for the changed environment (Boddy 2001). Here, the interactions between fungi and their environment can shape the trajectory of community development as a resource patch declines. But due to the stochastic nature of species arrival with saprotrophic fungi, the order of establishment may also affect community development through priority effects (Fukami 2015). In Chapter 5, I found that the initial colonisation of territory within a new resource patch may be an important component of the life history of saprotrophic fungi. With the importance of successfully establishing and defending a territory within a resource patch, the identity of an earlier coloniser may shape how the community assembles within the finite lifetime of the resource patches that saprotrophic fungi inhabit.

6.4 An archipelago of ephemeral resource islands?

Viewed at a coarse scale, the organic resources that saprotrophic fungi inhabit can appear as a continuous landscape. However, both temporally and spatially, these resource patches appear to saprotrophic fungi as discrete resource islands. The way resource patches are haphazardly arrayed within the environment can be thought of as an archipelago of resource islands connected through

167 the spread of dispersal propagules by saprotrophic fungi (Jonsson et al. 2005, Norros et al. 2012). By the joint interactions of fungi, these resource islands inevitably decline into non-existence. As a result, a fungus must be ready to disperse and move on to the next ephemeral island before the resources on the current island fall below a level where dispersal is possible (Chan et al. 2020).

Viewing the existence of saprotrophic fungi through this lens provides us with valuable insight into the life history trade-offs we may expect to maximise fitness. Resource allocation purely to dispersal in the face of competition will likely result in competitive exclusion, and allocation purely to defence in the absence of competitors will result in reduced fitness (Heaton et al. 2016). To persist within this

“archipelago of resource islands'', saprotrophic fungi would need to continually detect cues within the environment, and make the correct allocation decisions at every step in an environment in constant flux (Clobert et al. 2009, Sæther and Engen 2015).

Beyond considering the life history strategies of singular fungi, viewing the resource patches that fungi inhabit as islands allows us to incorporate elements of metacommunity and island biogeography - two classical areas of ecological theory. While direct application of these theories are limited, resource islands for saprotrophic fungi do have distinct patterns of immigration (newly arrived species) and extinction (species excluded by change in environment or new competitors)

(Andrews et al. 1987). But climax states and equilibrium cannot realistically be reached before island collapse due to the finite resources of the island. But prior to the collapse of an island, resource islands will continually form via the episodic and stochastic creation of detritus. These islands represent multiple communities of different ages connected via the dispersal of propagules at a greater spatial scale. The connectivity of these individual islands, each with their own distinct communities, represents the movement and spatial dynamics of saprotrophic fungi at a landscape scale (Hanski and Gilpin 1991, Leibold et al. 2004). While very far beyond the purview of this thesis, viewing saprotrophic fungi at a temporal and spatial scale beyond the interactions within the

168 lifetime of a single resource patch will provide context for much of the life history strategies and trade-offs that have been a focus for the studies presented here in this thesis.

6.5 A need for theory

There is a lack of theory for the basic ecology of saprotrophic fungi. To progress our understanding of how fungi interact, and further, how the environment plays a role in mediating these interactions, we need to return to a first-principles approach to understand the critical ecological processes of these organisms. Current approaches have focused on measurable traits, with particular focus on traits linked to ecosystem functioning (Crowther et al. 2014, Aguilar-

Trigueros et al. 2015, Treseder and Lennon 2015, Cline and Zak 2015, Dawson et al. 2019, Lehmann et al. 2019, Lustenhouwer et al. 2020). In doing so, we have gained numerous insights into how certain traits structure fungal communities at various spatial scales (Maynard et al. 2019). Trait- based analysis of fungal communities is an important step in understanding broader spatial patterns in fungal diversity and distribution, but I believe we need to take steps to link functional traits with life history strategies in fungi. Traits in saprotrophic fungi that promote ecosystem functioning are a by-product of fungi simply trying to complete their life cycle.

We are in need for new study systems to develop theory for saprotrophic fungi. In combination with empirical data and trait-based analysis, developing fungal life history theory and meta-population theory will allow us to make sense of the patterns in coexistence in fungi that we observe. In combination of the granular insights derived by first-principles research into life-history trade-offs within a greater understanding of functional traits, we can begin to make strides towards a more holistic view of fungal life cycles and ecology. One of the biggest hurdles in demystifying the life history strategies of saprotrophic fungi (and broadly, fungi in general) is the difficulty in

169 quantifying fitness (Pringle and Taylor 2002, Gilchrist et al. 2006, Heaton et al. 2016). In Chapters 2 and 3, I was fortunate to find a species able to produce countable fruiting bodies in-vitro, but for many other species, quantification of reproduction or dispersal remains untenable.

In this thesis, I explored elements of life history in saprotrophic fungi I serendipitously extracted from rotting logs of wood. I focused mainly on dispersal allocation and competition within finite resource patches. With my first-principles approach to fungal life history in the studies presented here, I contribute empirical data from my experiments informed by existing theories.

There were many more components of fungal biology and ecology that simply could not be included here. Mycelial recycling, where fungi are able to recover redundant biomass through autophagy, is a common strategy for the reuptake of vital nutrients (Josefsen et al. 2012, Heaton et al. 2016), but is not often considered within the life history strategies of higher eukaryotes. The dearth in understanding represents a gap in our knowledge of how the process of mycelial recycling interacts with fungal life history, and there remain many such gaps with regards to the ecology of saprotrophic fungi. Future studies to extend the body of work presented in this thesis would include experiments under field conditions for longer durations of time using natural substrates, and further experimental work to tease apart the different dimensions of resource and structural complexity within wood. Consideration of how life-history trade-offs in fungi impact the ecological functioning of fungal species is of great importance, along with the Inclusion of more complex competitor/parasitic relationships to explore life history allocation strategies. Finally, additional study species able to produce fruiting bodies to quantify dispersal would be the single biggest step forward in understanding allocation to dispersal within an archipelago of resource islands under natural settings.

Overall, the aim of this thesis was to explore how dispersal and competition fit into the greater scope of life history in saprotrophic fungi, with consideration for how the environment may influence how fungi live. Here, I took an experimental approach to extend our understanding of

170 dispersal allocation and competition in saprotrophic fungi. I highlight the importance of viewing saprotrophic fungi within the greater context of detritus as finite, ephemeral resource islands. While my earliest stated question of: “what do saprotrophic fungi do?” may have been far too broad for the purposes of this thesis, I considered how fungal life history strategies may provide us insight into what saprotrophic fungi might do when living in a variable environment. The next steps moving forward will be to understand how fungal life history and functional traits intersect to manifest the diverse communities observed in nature and to predict how these communities assemble (Crowther et al. 2014, van der Wal et al. 2015, Song et al. 2017). Developing our ideas on a strong foundational bedrock of understanding of the life history of individual fungi will be key to developing theory that is unique to the ecology of saprotrophic fungi.

171

6.6 References

Agrawal, A. A. 1999. Induced plant defense: evolution of induction and adaptive phenotypic plasticity. - Inducible plant defenses against pathogens and herbivores: biochemistry, ecology, and agriculture. American Phytopathological Society Press, St. Paul, MN: 251–268.

Aguilar-Trigueros, C. A. et al. 2015. Branching out: Towards a trait-based understanding of fungal ecology. - Fungal Biol. Rev. 29: 34–41.

Andrews, J. H. et al. 1987. Fungi, leaves, and the theory of island biogeography. - Microb. Ecol. 14: 277– 290.

Boddy, L. 2001. Fungal Community Ecology and Wood Decomposition Processes in Angiosperms: From Standing Tree to Complete Decay of Coarse Woody Debris. - Ecol. Bull.: 43–56.

Boddy, L. and Hiscox, J. 2016. Fungal Ecology: Principles and Mechanisms of Colonization and Competition by Saprotrophic Fungi. - Microbiol Spectr in press.

Bruns, T. D. 2019. The developing relationship between the study of fungal communities and community ecology theory. - Fungal Ecol. 39: 393–402.

Cadotte, M. W. et al. 2006. The effects of resource enrichment, dispersal, and predation on local and metacommunity structure. - Oecologia 149: 150–157.

Chan, J. Y. et al. 2020. Environmental cues for dispersal in a filamentous fungus in simulated islands. - Oikos 129: 1084–1092.

Cline, L. C. and Zak, D. R. 2015. Initial colonization, community assembly and ecosystem function: fungal colonist traits and litter biochemistry mediate decay rate. - Mol. Ecol. 24: 5045–5058.

Clobert, J. et al. 2009. Informed dispersal, heterogeneity in animal dispersal syndromes and the dynamics of spatially structured populations. - Ecol. Lett. 12: 197–209.

Crowther, T. W. et al. 2014. Untangling the fungal niche: the trait-based approach. - Front. Microbiol. 5: 579.

Dawson, S. K. et al. 2019. Handbook for the measurement of macrofungal functional traits: A start with basidiomycete wood fungi (M Hart, Ed.). - Funct. Ecol. 33: 372–387.

172

Edman, M. et al. 2004. Spore deposition of wood-decaying fungi: importance of landscape composition. - Ecography 27: 103–111.

Fricker, M. D. et al. 2008. Chapter 1 Mycelial networks: Structure and dynamics. - In: Boddy, L. et al. (eds), British Mycological Society Symposia Series. Academic Press, pp. 3–18.

Fuhrman, J. A. 2009. Microbial community structure and its functional implications. - Nature 459: 193– 199.

Fukami, T. 2015. Historical Contingency in Community Assembly: Integrating Niches, Species Pools, and Priority Effects. - Annu. Rev. Ecol. Evol. Syst. 46: 1–23.

Gause, G. F. 1934. The struggle for existence. Reprint; 1971. in press.

Gilchrist, M. A. et al. 2006. Identifying fitness and optimal life-history strategies for an asexual filamentous fungus. - Evolution 60: 970–979.

Hanski, I. and Gilpin, M. 1991. Metapopulation dynamics: brief history and conceptual domain. - Biol. J. Linn. Soc. Lond. 42: 3–16.

Heaton, L. L. M. et al. 2016. Energetic Constraints on Fungal Growth. - Am. Nat. 187: E27–40.

Hiscox, J. et al. 2015. Antagonistic fungal interactions influence carbon dioxide evolution from decomposing wood. - Fungal Ecol. 14: 24–32.

Jonsson, B. et al. 2005. Ecology of species living on dead wood – lessons for dead wood management. - Silva Fenn. in press.

Josefsen, L. et al. 2012. Autophagy provides nutrients for nonassimilating fungal structures and is necessary for plant colonization but not for infection in the necrotrophic plant pathogen Fusarium graminearum. - Autophagy 8: 326–337.

Kozłowski, J. 1992. Optimal allocation of resources to growth and reproduction: Implications for age and size at maturity. - Trends Ecol. Evol. 7: 15–19.

Lancaster, J. and Downes, B. J. 2017. A landscape-scale field experiment reveals the importance of dispersal in a resource-limited metacommunity. - Ecology 98: 565–575.

Lee, M. R. et al. 2019. Good neighbors aplenty: fungal endophytes rarely exhibit competitive exclusion patterns across a span of woody habitats. - Ecology 100: e02790.

173

Lehmann, A. et al. 2019. Tradeoffs in hyphal traits determine mycelium architecture in saprobic fungi. - Sci. Rep. 9: 14152.

Leibold, M. A. et al. 2004. The metacommunity concept: a framework for multi-scale community ecology. - Ecol. Lett. 7: 601–613.

Lustenhouwer, N. et al. 2020. A trait-based understanding of wood decomposition by fungi. - Proc. Natl. Acad. Sci. U. S. A. 117: 11551–11558.

Mäkipää, R. et al. 2017. Interactions between soil- and dead wood-inhabiting fungal communities during the decay of Norway spruce logs. - ISME J. 11: 1964–1974.

Maynard, D. S. et al. 2017. Diversity begets diversity in competition for space. - Nature ecology & in press.

Maynard, D. S. et al. 2019. Consistent trade-offs in fungal trait expression across broad spatial scales. - Nat Microbiol 4: 846–853.

Nemergut, D. R. et al. 2013. Patterns and processes of microbial community assembly. - Microbiol. Mol. Biol. Rev. 77: 342–356.

Niemelä, T. et al. 1995. Interactions of fungi at late stages of wood decomposition. - Ann. Bot. Fenn. 32: 141–152.

Norros, V. et al. 2012. Dispersal may limit the occurrence of specialist wood decay fungi already at small spatial scales. - Oikos 121: 961–974.

Pringle, A. and Taylor, J. 2002. The fitness of filamentous fungi. - Trends Microbiol. 10: 474–481.

Sæther, B.-E. and Engen, S. 2015. The concept of fitness in fluctuating environments. - Trends Ecol. Evol. 30: 273–281.

Song, Z. et al. 2017. Fungal endophytes as priority colonizers initiating wood decomposition. - Funct. Ecol. 31: 407–418.

Stearns, S. C. 1992. The evolution of life histories. - sidalc.net.

Stearns, S. C. 2000. Life history evolution: successes, limitations, and prospects. - Naturwissenschaften 87: 476–486.

Treseder, K. K. and Lennon, J. T. 2015. Fungal traits that drive ecosystem dynamics on land. - Microbiol. Mol. Biol. Rev. 79: 243–262.

174 van der Wal, A. et al. 2013. A thready affair: linking fungal diversity and community dynamics to terrestrial decomposition processes. - FEMS Microbiol. Rev. 37: 477–494. van der Wal, A. et al. 2015. Neglected role of fungal community composition in explaining variation in wood decay rates.

175