The Pennsylvania State University

The Graduate School

College of Agricultural Sciences

BEHAVIORALLY ACTIVE SUBSTANCES AFFECTING REPRODUCTIVE SUCCESS OF THE FUNGUS GNAT,

LYCORIELLA INGENUA, A PEST OF WHITE BUTTON MUSHROOMS, AGARICUS BISPORUS

A Dissertation in

Entomology

Kevin R. Cloonan

 2017 Kevin R. Cloonan

Submitted in Partial Fulfillment of the Requirements for the Degree of

Doctor of Philosophy

May 2017

The dissertation of Kevin R. Cloonan was reviewed and approved* by the following:

Thomas C. Baker Distinguished Professor of Entomology and Chemical Ecology Dissertation Advisor Chair of Committee

Nina E. Jenkins Professor of Entomology

Shelby Fleischer Professor of Entomology

John A. Pecchia Professor of Mushroom Science and Technology

Gary Felton Professor of Entomology Department of Entomology, Chair

*Signatures are on file in the Graduate School

iii

ABSTRACT

Little is known about the host seeking and courtship behaviors of the fungus gnat,

Lycoriella ingenua (Diptera: Sciaridae), one of the most severe insect pests of cultivated white button mushrooms, Agaricus bisporus. Published works concerning L. ingenua focus primarily on aspects of control including: insecticide susceptibility, insecticide resistance, viability of biological control agents, and determining damage thresholds. The work presented in this dissertation examines the host seeking behavior of female flies and the copulatory behavior of male flies. Host seeking and copulatory behaviors in insects involve a complex series of events beginning with signal acquisition, signal transduction, and signal processing. Signal molecules provide the searching insect with information about the quality and location of the host material or potential mate.

In this dissertation I present work showing that gravid L. ingenua female flies are attracted the composted substrate that mushrooms are grown on, not the mushrooms themselves, and to the mycoparasitic green mold, Trichoderma aggressivum. Fungi including Penicillium citrinum, Mycothermus thermophilium, and several Aspergillus sp. present in mushroom compost, may be responsible for the initial attraction to female L. ingenua flies. We also isolated a single female produced pheromone component, a germacradienol, that attracts male L. ingenua flies and elicits copulatory behavior including wing fanning and abdomen curling.

Ultimately the work presented in this dissertation may provide growers with new tools to manage and control L. ingenua populations in mushroom houses.

TABLE OF CONTENTS

LIST OF FIGURES ...... vii ACKNOWLEDGEMENTS ...... ix DEDICATION ...... x Chapter 1 - Introduction ...... 1

Overview of chapters ...... 1 Overview of study organisms ...... 3 White button mushroom, Agaricus bisporus ...... 3 Fungus gnats in the family Sciaridae ...... 6 Pest ecology: Lycoriella...... 7 Pest ecology: Bradysia...... 13 Entomophilous fungi...... 15 Lycoriella sp. transporting fungal spores...... 15 Bradysia sp. transporting fungal spores...... 17 Vectoring of fungal spores by L. ingenua: mutualism or coincidence? ...... 18 Chapter 2 – Attraction of female fungus gnats, Lycoriella ingenua, to mushroom-growing substrates and the green mold, Trichoderma aggressivum ...... 21

Abstract ...... 21 Introduction ...... 22 Materials and Methods ...... 24 Composting phases: preparation of mushroom growing-test substrates ...... 24 Agaricus bisporus mycelia...... 26 Trichoderma aggressivum (green mold) ...... 26 Flies ...... 26 Two-choice, static-flow olfactometer ...... 28 Olfactometer experiments ...... 30 Statistical analysis...... 31 Results ...... 31 Discussion ...... 33 Acknowledgements ...... 35 Chapter 3 – Attraction, oviposition, and survival of the fungus gnat, Lycoriella ingenua, on fungal species isolated from adults, larvae, and mushroom compost ...... 36

Abstract ...... 36 Introduction ...... 37 Materials and Methods ...... 41 Insects ...... 41 Fungal cultures ...... 42 Fungal species present in mushroom compost ...... 42 Fungal species isolated from adult flies ...... 42 Fungal species isolated from larvae ...... 44 DNA extraction, PCR, sequencing and identification of isolated fungi ...... 45 Two-choice, static flow olfactometer attraction assays ...... 46 Two-choice oviposition assays ...... 49 No-choice assays ...... 50 Larvae-to-adult survival assays ...... 52 Statistical analysis...... 52 Results ...... 53 Discussion ...... 59 Acknowledgements ...... 64 Chapter 4 – Isolation of a female-emitted sex pheromone component of the fungus gnat, Lycoriella ingenua, attractive to males ...... 65 Abstract ...... 65 Introduction ...... 66 Materials and Methods ...... 68 Insect rearing ...... 68 Olfactometer choice assays ...... 69 Extraction of sex pheromone component ...... 72 GC analysis ...... 73 Mass spectrometry analysis ...... 74 GC-EAD assays ...... 75 GC-BB experiments ...... 76 Statistical analyses ...... 77 Results ...... 78 Olfactometer choice assays ...... 78 GC-EAD assays of virgin female extract ...... 79 GC-BB ...... 82 GC-MS analyses ...... 84

Discussion ...... 85

Acknowledgements ...... 90

Chapter 5 – Conclusions and future directions ...... 91 Conclusion ...... 91 Implications for L. ingenua control in mushroom houses ...... 93 Possible mutualistic associations between L. ingenua and various fungal species ...... 94 Difficulties in proving mutualism ...... 97 Identification of volatiles responsible for the attraction of L. ingenua females ...... 101 Ecological contexts in which attraction and oviposition of L. ingenua females occur...... 103 Bibliography ...... 105

vii

LIST OF FIGURES

Figure 1-1: Magnified image of male and female Lycoriella ingenua adult flies in copula…….…7

Figure 2-1: Plastic static flow, two-choice olfactometer apparatus…………………………….…….….29

Figure 2-2: Percent of Lycoriella ingenua female flies attracted to unspawned compost, spawned compost, sterilized compost, and mushroom fruiting bodies in two-choice olfactometer assays…………………………………………………………………….………………………………….32

Figure 2-3: Percent of Lycoriella ingenua female flies attracted to Agaricus bisporus mycelia, spawned compost infested with Trichoderma aggressivum, and pure cultures of Trichoderma aggressivum in two-choice olfactometer assays…………………………….….………32

Figure 3-1: Pure fungal cultures, grown on potato dextrose agar, isolated from mushroom compost, adult flies, and larval frass……………………………………………………….………….…….……44

Figure 3-2: Glass static flow, two-choice olfactometer apparatus …………………………………….…47

Figure 3-3: Mean response of Lycoriella ingenua female flies attracted to pure fungal cultures isolated from mushroom compost, adult fly bodies, and larval frass in two- choice olfactometer assays………………………………………………………………..……………………….…54

Figure 3-4: Mean number of eggs lain by Lycoriella ingenua female flies on pure fungal cultures isolated from mushroom compost, adult fly bodies, and larval frass in two- choice oviposition assays………………………………………………………………..…………………………….55

Figure 3-5: Mean number of eggs lain by Lycoriella ingenua female flies on pure fungal cultures isolated from mushroom compost, adult fly bodies, and larval frass in no- choice oviposition assays………………………………………………………………..…………………………………………...…57

Figure 3-6: Mean survival of Lycoriella ingenua eggs to adulthood placed on pure fungal cultures isolated from mushroom compost, adult fly bodies, and larval frass ……………….59

Figure 4-1: (a) Schematic drawing of the constant-flow y-tube olfactometer used in assays monitoring the attraction of male Lycoriella ingenua flies to female flies and female fly extracts. (b) Mean percentages of male Lycoriella ingenua to adult female flies and female fly extracts in y-tube olfactometer experiments……………………………………….…..……71

Figure 4-2: Magnified image of the gas-chromatogram coupled electroantennogram (GC- EAD) assays with male Lycoriella ingenua flies……………………………………………….…….……..…76

viii

Figure 4-3: (a) Representative gas-chromatogram coupled electroantennogram (GC-EAD) traces from male Lycoriella ingenua flies in response to female fly extract on the polar EC-5 column. (b) Elution time of the EAD active component in female fly extracts on the polar EC-5 column………………………………………………………………..………………….…….……..…80

Figure 4-4: (a) Representative GC-EAD traces from male Lycoriella ingenua flies in response to female fly extract on the non-polar DB-255 column. (b) Elution time of the EAD active component in female fly extracts on the non-polar DB-255 column…………………………..…81

Figure 4-5: Analysis of the gas-chromatogram coupled behavioral bioassays (GC-BB) examining the response of adult male Lycoriella ingenua flies to female fly extract on the polar EC-5 column……………………………………………………………………………………………….……83

Figure 4-6: Electron-Impact mass spectrum from the Lycoriella ingenua female fly extract on an HP-5MS column……………………………………………………………………………….………………….…….85

ACKNOWLEDGEMENTS

Most importantly I would like to thank my advisor, Dr. Tom Baker, for all of his support.

Tom’s authenticity and focus on the present helped keep me grounded and sane during the hardest four years of my life. His guidance and encouragement helped me to become a better writer, scientist, and student of entomology. I would also like to thank my committee members, Dr. Nina Jenkins, Dr. Shelby Fleischer, and Dr. John Pecchia for providing guidance, insight, and support throughout my time at Penn State. Special thank you to Nina for her wit and encouragement during my times of anxiety.

Thanks to the members of my lab, Loyal Hall, Mike Domingue, Andy Myrick, Maria Mazin, and Stefanos Andreadis for all of their support and comradery. I would also like to thank Jason

Woolcott and Giovani Bellicanta for all of their technical assistance and insight into experimental designs.

I would also like to thank Loyal Hall, Arash Maleki, Kyle Burks, Joey Walls, Maridel

Fredericksen, Mehmet Ali Döke, and Dave Galbraith for their friendship. Without the support, laughs, and encouragement from these humans I wouldn’t have made it to the finish line.

Special thank you to Rebecca and Loyal Hall for feeding me an average of five times per month. Finally I would like to thank my family and friends back home for helping me become the person I am today, and for supporting me through all the times.

DEDICATION

This work is dedicated to my friend,

Luca Franzini (1982-2015)

One of the most creative, cunning, and intelligent people I have ever known. Luca was a great naturalist to boot.

Introduction

Overview of chapters

The purpose of the research presented within the pages of my thesis, including the results from three published papers, an introductory review chapter, and a concluding chapter, is similar to the goals of an early 20th Century sciarid fly researcher, H. B. Hungerford. In an early publication of his (Hungerford, 1916) he describes an infestation of flies in the genus Sciara, then in the family

Mycetophilidae, in potted plants around Manhattan, Kansas. He confesses in the introduction that, “…in the present state of our knowledge, we knew no satisfactory means of control”

(Hungerford, 1916). Thus, Hungerford set out to comb the literature for reports of flies in the genus Sciara. He also performed life history experiments with all stages of the fly. His stated purpose was to better understand the ecology of these flies in order to provide tools for greenhouse growers and homeowners to control Sciaria populations.

Now, 100 years later and following a path similar to Hungerford’s (1916), I have performed research on a different fungus gnat species, Lycoriella ingenua (Diptera: Sciaridae) to gain a better understanding of its life history, attraction to host volatiles, selection of oviposition substrates, and its sexual behavior. This fly is one of the primary insect pests of the white button mushroom, Agaricus bisporus. The ultimate goal of my research is to better understand the ecology of L. ingenua in order to provide mushroom growers with new tools for managing and controlling their populations in mushroom growing houses.

The research I conducted is laid out in the next three chapters. In the second chapter, published in Entomologia Experimentalis et Applicata (Cloonan et al., 2016a) I explored the attraction of female L. ingenua flies to various substrates found in mushroom cultivation. This work sought to identify which of the substrates used in different stages of the mushroom growing process that gravid female L. ingenua flies are attracted to. Here I demonstrated that females are preferentially attracted to, and lay eggs on, the composting material on which mushrooms develop, not to the developing mushrooms themselves. I also found that females are preferentially attracted to mushroom compost infested with the green mold, Trichoderma aggressivum. This mold is a parasite of A. bisporus and can decimate an entire mushroom crop.

In Chapter 3, published in PLoS ONE (Cloonan et al., 2016b), I expanded my studies to try to determine whether there were other fungal species that were more attractive to L. ingenua females than T. aggressivum green mold. I also went further and sought to determine the preference, if any, for females to oviposit on certain fungal species compared to others, regardless of their initial tendency to be attracted to the fungi. Since I had shown in Cloonan et al (2016a) that female flies are not attracted to the developing A. bisporus mycelia, and are attracted to the green mold T. aggressivum, I focused on finding other fungal species found in mushroom compost. I isolated and cultured pure cultures of six fungal species from mushroom compost and tested females’ attraction to them in a pitfall bioassay that I developed (Cloonan et al., 2016a; Cloonan et al., 2016b). I also isolated other species of fungi from adult fly bodies and from larval frass, that I then used to assess female attraction in the pitfall bioassay. I used both a choice and a no-choice oviposition assay to see the degree to which attraction of females corresponded to females’ tendency to invest eggs on these species. Surprisingly, I found some

3 non-correspondence between attraction and oviposition, and offer an interpretation of how this may work in this specialized micro-ecosystem of cultivated mushrooms. Finally, I found further non-correspondence between propensity of females to oviposit on certain fungal species and the survivorship of their larval offspring.

In Chapter 4, I report the significant advances that I helped produce in isolating and identifying the L. ingenua female-emitted sex pheromone ((Andreadis et al., 2015a); I was co- senior author). I played the major role in allowing our team to be able to isolate a single, behaviorally active component from extracts of adult female L. ingenua bodies. This single component elicited significant electroantennographic (EAG) signals in male antennae, and produced robust, stereotypical sexual behavior in males. This research corrected a 35-year-old misidentification and sets the stage for finally synthesizing a truly effective pheromone that can be used by growers for mushroom IPM. I felt that in performing this research, once the pheromone could be identified and synthesized, it could be developed for use in potentially controlling L. ingenua populations in mushroom growing houses via mass trapping or mating disruption. Its immediate use would be in monitoring traps to for assessing population levels to aid in IPM.

Overview of the study organisms

White button mushroom, Agaricus bisporus

According to the National Agricultural Statistics Service (NASS), there were 945,639,000 lbs of white button mushrooms produced in 2015 yielding a total sales of $1,190,672,000 (Agriculture,

2015). White button mushrooms are grown on a highly selective composted substrate that goes

4 through two phases of composting (De Groot et al., 1998). This composted substrate typically includes: wheat straw, straw bedded horse manure, chicken manure, and gypsum (Royse &

Beelman, 2013). These raw materials are either stacked outdoors in large windrows (“ricks”) or they are filled into a forced aerated “bunker” that then undergo the first step of composting referred to as “Phase I”.

The goals of Phase I composting are to create a selective substrate, increase its water holding capacity, and begin the process of breaking down cellulose in the straw (Savoie et al.,

1996). This process typically lasts 7-10 days and temperatures within the compost reach up to

80°C. Without Phase I composting the resulting mushroom yield is about 20% lower (Straatsma et al., 1994; Straatsma & Op Den Camp, 1994). After the 7-10 days Phase I composting regime the compost is either filled into a tunnel or into a growing room for the second stage referred to as “Phase II”.

The goal of Phase II composting is to pasteurize and condition the substrate. The compost temperatures are allowed to reach approximately 60°C for 2 to 8 hours during the pasteurization stage, followed by compost temperatures being brought down to 48°C for the condition stage.

The conditioning stage lasts between 4 to 10 days, depending on the amount of ammonia in the compost and the Phase II system being utilized. The goal of this conditioning period is to reduce the amount of ammonia in the substrate to around 10 ppm. After the ammonia level is reduced below 10ppm, the compost temperature is reduced to 27°C to prepare for spawning.

Lower ammonia during Phase II pasteurization is directly linked to the presence of the fungus, Mycothermus thermophilum, previously described as Scatyldium thermophilum (Straatsma et al., 1994; Straatsma & Op Den Camp, 1994). Without M. thermophilium in the final composted material the harvestable mushroom yield drops significantly (Coello-Castillo et al., 2009). Phase II pasteurization also makes the compost a selective growing substrate for microbes and fungi that are beneficial to A. bisporus development

(Straatsma et al., 2000). This process also decreases the incidence of other competitor- and pest- fungi in the compost.

This pasteurized compost is then mixed with a slow-release nitrogen supplement and grain spawn. Grain spawn are small pieces of millet or rye grain that have been sterilized and colonized by mycelia of one A. bisporus clonal variety. This pasteurized compost, grain spawn, and nitrogen supplement mixture is then moved into a growing room called a double. These growing rooms typically contain about 743 sq. m. of growing area. For the next two weeks the compost mixture undergoes spawn run, referring to A. bisporus mycelia colonizing the compost.

Temperatures inside the growing room during spawn run remain between 22°C-25°C for two weeks with a relative humidity of around 90%. After two weeks of colonization the heavily spawned compost is cased, or covered with a layer (ca. 4-5 cm thick) of a peat moss, limestone, and casing inoculum mixture. The casing inoculum are small fragments of already developed A. bisprous mycelia. This casing mixture encourages the formation of sporophores, which later become the fruiting bodies (i.e. mushrooms). Casing inoculum encourages mycelia in the compost to “knit” together with mycelia in the casing so they begin to form the upward-growing stalk of each mushroom fruiting body. The casing inoculum is a key feature that allows for rapid

6 sporophore formation. After casing the compost temperature is reduced to 20°C and the CO2 level is dropped from around as high as 8,000 ppm to between 300-1,000 ppm. This rapid reduction in temperature and CO2 levels encourages sporophore formation, although the precise mechanism has never been described.

As the small sporophores are forming and growing on the top of the casing layer mushroom growers are continuing to water on an as-needed basis. Approximately 15-17 days after the spawned compost is covered with the casing material the first batch of mushrooms is ready for harvest. This “first break”, or the first flush of harvestable mushrooms, is hand-picked.

The interval between casing and the first break is referred to as casehold. A second break is picked one week after first break. Between the first and second break growers are continuing to hand water on an as-needed basis. This process is repeated until the third and final break.

Although more mushrooms will continue to grow after the third break, the likelihood of pest and disease outbreaks increases significantly. After the third break the entire room, including the spawned compost material, is pasteurized in a process called steam-off. This ensures that pests and diseases are not transferred out of each growing room. Steamed compost, or mushroom compost, can be disposed of or sold as a soil amendment.

Fungus gnats in the family Sciaridae Flies in the family Sciaridae (Diptera), commonly referred to as the “dark-wing fungus gnats”, are

5mm-2cm long with a characteristic black humpbacked thorax and prognathous mouthparts

(Mohrig et al., 2012) (Figure 1-1). They live primarily in moist and shady environments containing relatively high amounts of organic matter. Feeding habits in the family vary widely from fungivory, detritivory, to herbivory (Mohrig et al., 2012). Two genera of Sciaridae, Lycoriella and

Bradysia, contain the majority of the described pest species. Flies in the genus Lycoriella are pests primarily of mushroom production systems, and flies in the genus Bradysia are pests primarily of fruit and vegetable production in greenhouse systems. Curiously, species in both genera are pests of the Montezuma pine, Pinus montezumae in Mexico (Marín-Cruz et al., 2015).

Figure 1-1: Magnified image of an adult male and an adult female L. ingenua fly.

Pest ecology: Lycoriella One of the two primary dipteran pests of A. bisporus in North America is Lycoriella ingenua

(Lewandowski et al., 2004). In European A. bisporus-growing regions the primary sciarid pests are

Lycoriella auripila (Farsani et al., 2013), Lycoriella castanescens (Tibbles et al., 2005), and

Lycoriella solani (Jess & Kilpatrick, 2000). The pest ecologies of all three species and their associations with A. bisporus are similar.

Gravid L. ingenua females are initially attracted to growing rooms filled with newly spawned compost (Cloonan et al., 2016a; O'Connor & Keil, 2005; Richardson, 1987). Adults emerge 21 days after a house is filled (Mehelis, 1995), and the L. ingenua generation time under these conditions from egg to adult averages 21 days (Lewandowski et al., 2012). Thus, L. ingenua females enter a growing house and lay their eggs the first day a house is filled with newly spawned compost. Until my recent work showing that L. ingenua adults are attracted to mushroom compost without any A. bisporus growing in it (Cloonan et al., 2016a), it was commonly believed that flies were attracted to the developing mushrooms (Kim et al., 1999).

Adult L. ingenua emerge outside of a mushroom house when the temperatures are between 9-20°C (Joshi et al., 2011). No research has explored other L. ingenua hosts or where the initial infesting populations originate in the environment. When gravid females enter a newly spawned growing room they oviposit their eggs 2-4 cm-deep into this freshly spawned compost

(Cantelo, 1988). The hatched larvae then begin to feed on the compost and the microorganisms that inhabit the substrate. As they molt, larval instars move deeper into the compost. The fourth larval instar ascend toward the top layer of the compost to pupate (Cantelo, 1988).

An economic threshold for L. ingenua was calculated to be 13 larvae per 125 g of casing

(Kielbasa & Snetsinger, 1980). It is important to note that this economic threshold was calculated using casing instead of compost. Early reports cite a 17% decrease (Cantelo, 1979b) and a 24%

9 decrease (Rinker et al., 1984) in overall A. bisporus yield, respectively, due to high L. ingenua infestations. However, the authors in both studies mention this yield reduction was also likely due to heavy infestations of the mushroom phorid fly, Megaselia halterata, observed in their experimental blocks. A negative linear relationship between A. bisporus yield and fly infestations has been described for L. auripila (Shirvani-Farsani et al., 2013).

The four L. ingenua larval instars cause the majority of damage done to the developing A. bisporus mycelia. No experimental data has described the mechanism of L. ingenua damage to developing A. bisporus mycelia in compost. Some authors suggest that larvae “nibble” on the developing mycelia (Tibbles et al. 2005). These cite early publications stating that L. ingenua larvae have strong mandibles, and thus they must be chewing on the A. bipsorus mycelia (Binns,

1980a; Lee et al., 1999; White, 1986). Anecdotal and conflicting claims like these in the literature make it difficult to describe the extent of L. ingenua damage to mushroom crops.

When L. ingenua and L. castanescens larvae have been placed directly on cultivated strains of A. bipsorus grown on potato dextrose agar, no larvae survived to adulthood (O'Connor

& Keil, 2005; Smith et al., 2006). However, L. ingenua can fully develop on mycelia of a cultivated strain grown on sterilized grain spawn (Tung & Snetsinger, 1973). Microbial fauna may have allowed L. ingenua to survive on the piece of grain spawn. These data suggest direct feeding damage to the developing mycelia might be accidental as the larvae move through and eat other microorganisms and nutrients in the compost. Other authors suggest that L. ingenua larvae directly compete with A. bisporus mycelia for nutrients in the compost (Grewal & Richardson,

1993; White, 1986), thereby reducing yield.

When a fourth larval instar moves upwards through the compost and casing to pupate, it can damage the sporophores that will later become the mushroom fruiting bodies. Larvae can tunnel directly into the stipe of these developing mushrooms to pupate (Hussey & Gurney, 1968).

This often severs mushrooms from mycelia, preventing water and nutrient exchange to the mushroom and it dies. If L. ingenua larvae are introduced into an otherwise healthy A. bisporus crop right at casing, about 50% of the developing sporophores will be damaged (Erler et al.,

2011).

Early research into the pest ecology of L. ingenua found that larvae survive better on unspawned compost versus spawned compost (Cantelo & Antonio, 1982). Similar results were seen with the closely related species L. auripila (Scheepmaker et al., 1996) and L. castanescens

(Smith et al., 2006; Tibbles et al., 2005). A significant reduction in L. ingenua larval survival was seen after just two (Hussey & Gurney, 1968) and seven days (Cantelo & Antonio, 1982) of A. bisporus colonization in compost. Conflicting results around the same time showed that A. bisporus in compost had no effect on L. ingenua development (Kielbasa & Snetsinger, 1981).

Reduced larval survival may be due to the production of mannitol and calcium oxalate by the developing A. bisporus. Both mannitol and calcium oxalate are toxic to L. auripila (Binns,

1980b; White, 1997). Proposed roles of calcium oxalate in fungi include increased structural support, detoxifying heavy metals, and aiding in the breakdown of cellulose (Dutton & Evans,

1996). Mannitol primarily acts as an oxidative and osmotic stress buffer in fungi (Patel &

Williamson, 2016). It also plays a role in the fungus’ susceptibility to microbial infection (Wang et al., 2012). These two toxic compounds may also serve as a protectant against insect feeding.

The survival of L. ingenua larvae is reduced to almost zero when compost is sterilized

(Kielbasa & Snetsinger, 1981). Similar results were seen with L. auripila (Binns, 1972). Both authors suggest that L. ingenua and L. auripila larvae are consuming other fungi and bacteria present in the compost instead of the developing A. bisporus mycelia. Unspawned mushroom compost left for 15 days and then exposed to L. ingenua adults resulted in significantly more adults emerging versus fresh unspawned compost (Cantelo & Antonio, 1982). These data suggest that the compost held for 15 days contained more bacteria and fungi for the L. ingenua larvae to consume. It is clear from my research in the following chapters that L. ingenua females are attracted to fungal and other microbial growth in mushroom compost.

In 16-choice static-flow olfactometer tests L. ingenua females were preferentially attracted to and oviposited eggs on species of Penicillium, Aspergillus, Fusarium, and

Trichoderma (Frouz & Novakova, 2001). Several species of Penicllium have been isolated from mushroom compost (Siyoum et al., 2016) and from L. ingenua female bodies (Cloonan et al.,

2016b). Both Aspergillus niger and Aspergillus fumigatus have been isolated from commercially produced compost (Buczynska et al., 2008; Straatsma & Op Den Camp, 1994; Zhang et al., 2014) and from adult L. ingenua bodies (Cloonan et al., 2016b). These fungi are the causal agent of

“mushroom workers lung”, an inflammatory disease caused by the repeated inhalation of

Aspergillus sp. spores (Bogart et al., 1993; Kwon-Chung & Sugui, 2013).

Chemical insecticides are the most common method used to control Lycoriella sp. The organophosphate diazinon (Shamshad, 2010) and the neonicotinoid Imidicloprid (Shirvani-

Farsani et al., 2013) significantly reduce L. ingenua populations when applied at casing. There is a significant risk of these flies developing insecticide resistance. Resistance to permethrin was

12 first described in the late 80’s (Brewer & Keil, 1989a), suggesting the resistance was due primarily to oxidative metabolism and target site mutations (Bartlett & Keil, 1997). Around the same time, reports of L. auripila resistant to diazinon emerged (White & Gribben, 1989b).

The insect growth regulators Teflubenzuron and Novaluron significantly reduced the emergence of L. ingenua adults in mushroom compost (Erler et al., 2011). The chitin synthase inhibitors Diflubenzuron (Ali et al., 1999; Cantelo, 1983; Jess & Kilpatrick, 2000) and Triflumuron

(Shamshad et al., 2008) significantly reduced the number of emerging flies compared to untreated controls. However Diflubenzuron significantly reduces A. bisporus yield in treated blocks (Ali et al., 1999; Grewal et al., 1992; White, 1999). This is likely due to the fact that

Diflubenzuron is a chitin synthase inhibitor and A. bisporus has chitin in the cell walls. There is a moderate but insignificant mortality of larvae and adults treated with high doses of several essential oils (Choi et al., 2006; Park et al., 2006; Park et al., 2008; Yi et al., 2015).

Regarding potential biocontrol agents for use in controlling L. ingenua, infective

Steinernema feltiae juveniles applied at casing have been shown to reduce L. ingenua populations by 95% (Nickle & Cantelo, 1991), and by 82% with regard to L. ingenua larval survival (Jess &

Kilpatrick, 2000). There was no effect on L. ingenua survival when S. feltiae were applied at filling

(Grewal & Richardson, 1993). Predatory mites, Hypoaspis miles, were shown to have no effect on L. ingenua populations when applied at filling (Jess & Schweizer, 2009), but did significantly reduce emerging populations of Lycoriella solani by 30% when applied throughout the A. bisporus growing cycle (Ali et al., 1999). Bacillus thuringiensis var. isrealiensis applied at both filling and casing reduced L. ingenua populations by 50-75% (Cantwell & Cantelo, 1984).

Pest ecology: Bradysia

Flies in the genus Bradysia are primarily pests of vegetable and fruit crops grown in greenhouses

(Cloyd, 2008; Cloyd, 2015). Similar to Lycoriella sp. on mushrooms, these flies enter a greenhouse when it is filled with potting soil and plant material (Harris et al., 1996). One report suggests that seedling-plugs of several vegetable species purchased from third-party growers are already infested with Bradysia larvae (Cloyd & Zaborski, 2004). The flies’ ability to enter small openings makes it almost impossible for growers to exclude them from entering a greenhouse.

Neonicotinoid insecticides, including Dinotefuran, Imidacloprid, Thiamethoxam and

Clothianidin, are the most widely used and effective tools for controlling Bradysia populations in greenhouse systems (Cloyd & Dickinson, 2006). More recently growers have adopted the use of less toxic insect growth regulators such as Pyriproxyfen, Fenoxycarb, and Azadirachtin for effectively controlling Bradysia populations (Cloyd & Dickinson, 2006; Ludwig & Oetting, 2001).

In conjunction with insecticides, some entompathogenic nematode species have been said to produce effective control for Bradysia populations (Jagdale et al., 2004). However, other reports suggest entomopathogenic nematodes offer no control for Bradysia sp. (Cloyd & Dickinson,

2006). Although it has proven very difficult to execute in practice, the most successful control tool is exclusion of flies through the use of potting media that is not attractive to host-seeking

Bradysia females (Cloyd, 2015; Evans et al., 1998; Olson et al., 2002).

Soilless media that is not attractive to invading Bradysia sp. populations may reduce the initial pest pressure in greenhouse systems. Under multi-choice experiments female Bradysia coprophila flies were equally un-attracted to three of the most commonly used soilless potting media: Metro-Mix 560, Sunshine LC1 Mix, and Universal SB 300 Mix (Meers & Cloyd, 2005). To date no soil-less media has been adopted by the greenhouse industry in an attempt to exclude

Bradysia sp..

Other authors suggest that by removing sources of fungi, host seeking females will be less attracted to a greenhouse (Harris et al., 1996). Two-choice attraction assays showed that

Bradysia impatiens females always chose to oviposit in composted substrates that were older and more humid versus those that were younger and drier (Cloyd et al., 2009). Female flies were likely more attracted to the moister compost because of increased fungal and microbial activity.

More large-scale experiments should examine if B. impatiens are less attracted to greenhouses filled with compost containing less fungal and microbial growth.

Bradysia coprophila females are more attracted to soils with an 80%-by-mass organic matter composition (Anas & Reeleder, 1988a) than to other compositions. Females cease all attraction to soils with an organic matter composition of 40% or lower (Anas & Reeleder, 1988a).

Soils with a lower composition of organic matter likely have less microbial and fungal activity.

Early work suggested that B. impatiens actually feed primarily on the microorganisms present in greenhouse soils, not on the plants themselves (Kennedy, 1974). In an extension publication,

Lindquist et al. observed that more Bradysia sp. adults emerged from potting soils having greater levels of microbial activity (Lindquist et al., 1985).

Entomophilous fungi

Entomophilious fungi, or fungi that rely on insects for spore dispersal, have been best described in ant (Abril & Bucher, 2002; 2004), beetle (De Fine Licht & Biedermann, 2012; Endoh et al., 2011), and termite (Aanen et al., 2002) systems. These mutualistic insect-fungal interactions benefit the fungi through increased spore dispersal, and benefit the insects through nutrient acquisition and recycling, substrate detoxification, and defense against potential agonists in the host substrate

(De Fine Licht & Biedermann, 2012). A body of applied experimental studies have shown that pestiferous sciarid flies can carry fungal spores of several species of fungi to host substrates. Here

I present the available literature showing that both Lycoriella sp. and Bradysia sp. are able to transport fungal spores of several fungi to artificial substrates and host material. I then attempt to integrate these interactions into the context of insect-fungi mutualisms.

Lycoriella sp. transporting fungal spores

Several references suggest L. ingenua carry fungal spores on their bodies, including the green mold T. aggressivum, to A. bisporus (Kim et al., 1999; Rinker et al., 1995; Smith et al., 2006; Yi et al., 2015). However little experimental work has described the mechanism of the flies’ possible vectoral capacity to produce new fungal infestations. Smith et al. claims that L. ingenua flies are a pest of A. bisporus primarily because they vector fungal pathogens into mushroom houses

(Smith et al., 2006). These authors cite a publication (Scheepmaker et al., 1997) that actually describes the mushroom phorid fly pest, Megaselia halterata, carrying fungal spores on their bodies, not L. ingenua. Another example that is often cited claims that L. ingenua adults vector

Trichoderma harzianum and Verticillium fungicola to A. bisporus (Rinker et al., 1995). These authors do not offer experimental evidence or citations to such evidence to support this claim.

Interestingly, Erler et al. had to throw out several replicates of their insecticide treatment trials on spawned mushroom compost because of fungal pathogen outbreaks. Those treatment blocks infested with L. ingenua had high incidences of the fungal pathogens Verticillium fungicola, the causal agent of dry bubble disease in A. bisporus, and Mycogone perniciosa, the causal agent of wet bubble disease in A. bisporus (Erler et al., 2011). Spores of V. fungicola are seen on adult

L. ingenua flies in the joints between their femur and tibia under scanning electron microscopy

(Shamshad et al., 2009). However L. ingenua flies exposed to sporulating V. fungicola were shown to be unable to vector the pathogen to A. bisporus growing in compost (Shamshad et al., 2009).

Infestations of V. fungicola seen by Erler et al. could have been due to adult L. ingenua vectoring the fungal pathogen. Alternatively, the methods used by Shamshad et al. were not sufficient for spreading V. fungicola to A. bisporus.

Other authors reported in their biological control trials similarly high infections of several

Fusarium sp., a genus of parasitic fungi that cause developing A. bisporus fruiting bodies to

“damp-off”. Only those blocks infested with large numbers of L. ingenua adults experienced

Fusarium sp. outbreaks (Jess & Kilpatrick, 2000). Adult L. ingenua flies could have spread the

Fusarium sp. spores throughout the experimental blocks. The same group later reported, in further biological control trials, that blocks infested with L. ingenua adults experienced an outbreak of the weed mold, Murcor sp. (Jess & Schweizer, 2009). They associated this outbreak with mycophagous mites. Adult L. ingenua (Binns, 1980a; Cantelo & Antonio, 1982) and L. auripila

(Binns, 1972) flies have been shown to vector several mite species, including some that are pests

17 of A. bisporus (Clift & Larsson, 1987). In the case of Jess and Schweizer (2009), L. ingenua adults could have vectored the mites onto the mushroom compost, and these mites could have then vectored the Mucor sp. between the treated blocks.

Bradysia sp. transporting fungal spores

More work has examined the ability of Bradysia sp. flies to carry fungal spores than flies in the genus Lycoriella. This is likely due to Bradysia sp.’s pest status on a wide variety of high value greenhouse crops (Cloyd, 2008; Cloyd, 2015). These flies have been associated with fungal plant diseases since the 1960’s. Outbreaks of Fusarium sp. in red clover and alfalfa seedlings are thought to be caused by feeding Bradysia sp. larvae (Leath & Newton, 1969). The causal agent of grey mold in many greenhouse fruit and vegetable crops, Botrytis cinerea, is generally associated with heavy Bradysia sp. infestations (Keates et al., 1989). Causal links between Bradysia sp. adults and these fungal infestations have never been shown. However there is experimental data describing the vectoral capacity of other Bradysia sp. flies and plant fungal pathogens.

Under laboratory conditions Bradysia difformis adults can vector spores of several

Fusarium sp. to nutrient agar (Hurley et al., 2007). Female B. difformis adults prefer to oviposit on plant roots infected with Fusarium culmorum (Kuhne & Heller, 2009). Several species of

Cladosporium, the causal agent of many greenhouse fruit leaf molds, was found on adult B. difformis bodies (Hurley et al., 2007). Other unidentified Bradysia sp. vector the fungal plant pathogen, Fusarium oxysporum f.sp. radicis-lvcopersici, to healthy lima bean and tomato plants

(Gillespie & Menzies, 1993), and Fusarium avenaceum to healthy lisianthus plants (El-Hamalawi

& Stanghellini, 2005).

Female Bradysia impatiens adults have been shown to be capable of mechanically vectoring spores of the fungal pathogens Verticillium dahliae, Fusarium acuminatum and

Thielaviopsis basicola to artificial media (El-Hamalawi, 2008). Scanning electron microcopy images revealed that Fusarium sp. spores are not present on adult B. impatiens fly bodies exposed to Fusarium sp. grown on artificial media (Jarvis et al., 1993), and adult flies are not viable vectors of several plant pathogenic Pythium sp. (Braun et al., 2010). Ingested oospores, however, do remain viable in the larval digestive tract for 48 hours and larvae can transmit

Pythium sp. infections to healthy plants (Braun et al., 2009; Gardiner et al., 1990). Alfalfa wilt, caused by several species of Verticillium, is also vectored by adult B. impatiens flies (Kalb & Millar,

1986). Larval B. impatiens can survive to adulthood on pure cultures of the fungus Altcrnaria tennis (Kennedy, 1974), the causal agent of leaf spot diseases in over 350 plant species. Adult B. impatiens have never been shown to carry A. tennis spores.

Vectoring of fungal spores by L. ingenua: mutualism or coincidence?

I show in Chapter 2 that spawned mushroom compost infested with T. aggressivum is more attractive to gravid L. ingenua females than uninfected spawned mushroom compost (Cloonan et al., 2016a). Unpublished data by a Penn State graduate student, Maria Mazin, provides evidence that female L. ingenua adults are able to carry spores of the green mold T. aggressivum onto artificial media. More recent unpublished data by Maria Mazin shows that L. ingenua survival is significantly increased on spawned mushroom compost infested with T. aggressivum.

Although not in the goals of my dissertation research, an interesting question is raised by M.

Mazin’s and my research findings that concerns the ecology and evolution of the fungus/fly

19 microecosystem in nature and in mushroom houses. The question is whether or not there has evolved a mutual benefit to both the vectored fungal species and to the fly with regard to the fly’s attraction to, and oviposition on, the fungus.

The T. aggressivum in spawned mushroom compost may provide several fitness benefits to developing L. ingenua larvae. First, T. aggressivum inhibits A. bisporus mycelial growth in the compost (Krupke et al., 2003). Developing A. bisporus mycelia directly compete with developing

L. ingenua larvae in the compost for nutrients (Grewal & Richardson, 1993; White, 1986).

Reducing A. bisporus colonization of the compost due to T. aggressivum infestation may liberate more nutrients available for developing L. ingenua larvae. Second, developing A. bisporus mycelia produce the toxic compounds mannitol and calcium oxalate (Binns, 1980b; White, 1997), which could be detoxified by T. aggressivum. Taken together, these data suggest this could be an example of a mutually beneficial insect-fungal interaction. The green mold would obtain the benefit of increased spore dispersal, and the ovipositing L. ingenua females would obtain the fitness benefit of increased larval survival. These outcomes deserve more in-depth experimentation however, to demonstrate definitively a true mutualistic relationship. Further discussion of the difficulties in proving the existence of mutualism in fly/fungal ecosystems can be found in the Conclusions section (Chapter 5) of this dissertation.

Although the inside of a mushroom growing house is an artificial setting, it may provide an arena to produce some insight into how T. aggressivum and L. ingenua interact in a more natural landscape. In order to better understand the ecology of L. ingenua as a pest of A. bisporus, more research needs to examine the flies’ natural ecology. By doing so we may gain better insight

20 into the pest ecology of these flies that leads to new ways to control their populations in mushroom houses.

Chapter 2 Attraction of female fungus gnats, Lycoriella ingenua, to mushroom-growing substrates and the green mold Trichoderma aggressivum

*Published June 2016: Cloonan KR, Andreadis SS, Baker TC. Attraction of female fungus gnats, Lycoriella ingenua, to mushroom-growing substrates and the green mold Trichoderma aggressivum. Entomologia Experimentalis et Applicata. 2016;159(3)

Abstract

To evaluate the attractiveness of several mushroom-growing substrates to the female mushroom fly Lycoriella ingenua (Dufour) (Diptera: Sciaridae), a pest of the cultivated white button mushroom, Agaricus bisporus (JE Lange) Emil J Imbach (Agaricales), we developed a two-choice, static-flow olfactometer. Behavioral assays using this olfactometer indicated that mushroom compost with A. bisporus mycelia growing in it was not more attractive than compost lacking growing mycelia. We also found that female flies were more attracted to compost lacking A. bisporus mycelia than to the actual commodity, the white button mushroom fruiting bodies. Flies were not, however, attracted to sterilized compost, suggesting the attraction is due to volatiles produced by microbial metabolism in the compost. We also found that female L. ingenua flies were attracted to the mycoparasitic green mold Trichoderma aggressivum Samuels & W Gams

(Hypocreales). Flies preferred mushroom compost that had T. aggressivum growing in it over compost lacking T. aggressivum, providing an experimental outcome consistent with the

22 anecdotal belief that L. ingenua flies are vectors of T. aggressivum spores that can infest mushroom growing houses.

Introduction

Fresh market production of the white button mushroom, Agaricus bisporus (JE Lange) Emil J

Imbach (Agaricales), in the USA reached 394 650 metric tons (877 million pounds) and grossed nearly US$ 1.4 billion between 2012 and 2013. Approximately 60% of the production occurred in

Pennsylvania (USDA, 2014), where this crop has two major insect pests: the fungus gnats

Megaselia halterata (Wood) (Diptera: Phoridae) and Lycoriella ingenua (Dufour) (Diptera:

Sciaridae), formerly Lycoriella mali (Fitch) (Wetzel et al., 1982). Although little definitive experimental evidence exists, it is generally thought that L. ingenua damages mushrooms through (1) direct larval feeding on developing A. bisporus mycelia in the growing compost media

(Cantelo, 1979b; Grewal et al., 1993; Kielbasa & Snetsinger, 1980), (2) larval competition with developing A. bisporus mycelia for nutrients in the compost (Binns, 1980b), thus reducing the water-holding capacity of the compost (Fletcher et al., 1989), (3) possible vectoring of several mycoparasitic Trichoderma spp. (Hussey, 1968), and (4) a negative impact of larval frass on mycelial growth (Hussey & Gurney, 1968). A significant reduction in crop yield is known to occur at a density of just 130 larvae per 30 g of spawned compost (Hussey, 1968), or one larva per 125 g of casing (White, 1986).

Females live on average 5 days (Wetzel et al., 1982) and deposit on average 200 eggs in their lifetime. Highly fecund females are able to produce eggs 24 h after emergence and mating

(Cantelo, 1988), with a generation time of approximately 20 days (egg to adult) at normal A.

23 bisporus growing temperatures (Lewandowski et al., 2004). Thus, for an economic problem to develop it only takes a few gravid female flies entering a growing house and laying eggs.

Several integrated approaches to the control of L. ingenua primarily target the developing larval stages of the fly. They include compost drenches with the insecticides diazinon or imadocloprid (Shirvani-Farsani et al., 2013), treatments using the juvenile hormone analog methoprene, or the application of the chitin-synthase inhibitor diflubenzuron (Cantelo, 1983).

Efforts have also been made to incorporate biocontrol agents such as predatory mites (Ali et al.,

1999; Jess & Schweizer, 2009) and entomopathogenic nematodes (Shamshad et al., 2008) into the cropping regime to control the fly. The most common control measures still involve long- term use of compost drenches with chemical insecticides. This practice has had some negative consequences. For example, resistance to permethrin was reported for sciarid pests including L. mali (subsequently named L. ingenua) in commercial mushroom growing houses after just 3 years of application (Bartlett & Keil, 1997; Brewer & Keil, 1989a). Also, possible resistance to diazinon was reported for the mushroom pest Lycoriella auripila (Winnertz) (Diptera: Sciaridae)

(White & Gribben, 1989b).

Because of such resistance development, the high fecundity of female L. ingenua, and their almost impossible exclusion from growing rooms, novel tools are necessary to control this significant pest. It is anecdotally believed that females enter a growing house directly after ‘phase

II’ composting, before the compost has A. bisporus added to it, i.e., is 'spawned' or 'filled' in a growing room (O'Connor & Keil, 2005; Shamshad, 2010). Although not experimentally proven, this belief exists because the first generation flies emerge as adults 15–20 days after a room is filled. If this is the case, female flies may not be attracted to the volatile emissions of developing

A. bisporus in the compost, but instead may be attracted to other volatiles released from microorganisms within the compost itself. We therefore developed a two-choice, static-flow olfactometer to conduct a series of bioassays that would allow us to find the natural substrates that are most attractive to female L. ingenua, so that we subsequently might begin to isolate and identify the volatile compounds that are responsible for this attraction. The bioassay apparatus proved to be discriminating in getting females to move toward some mushroom cultivation materials more than others at different stages of production and infestation with fungal mycelia or spores of different fungal species.

Materials and methods

Composting phases: preparation of mushroom-growing test substrates

Compost goes through various preparation phases before it can be used for mushroom production. These phases are outlined below as they are relevant to our experimental uses.

Phase I compost. The composting material was provided by Dr. John Pecchia (Mushroom

Research Center, Pennsylvania State University, University Park, PA, USA). The compost consisted of a combination of switchgrass (Panicum virgatum L.) straw, wheat (Triticum aestivum L.) straw- bedded horse manure, and dried and pelletized poultry manure. The percentages of raw materials in the compost, on a dry-weight basis, were 72% wheat straw-bedded horse manure,

10% switchgrass straw, 6.5% dried distiller’s grain, 6.5% dried, pelletized poultry manure, and 5% agricultural gypsum. This mixture underwent a 6-day aerated phase I composting outside made up of the previously mentioned materials plus additional wheat straw-bedded horse manure, additional dried and pelletized poultry manure, gypsum, and dried distillers grain. The primary

25 goals of phase I composting are to break down the physical structure of the compost to increase its water-holding capacity, and to allow microbial activity to degrade complex proteins and polysaccharides (Pudelko, 2014).

Phase II compost. After the 6-day phase I composting regime, the composting material was brought inside a growing room and the temperature was raised to 52 °C for 9 days. The goal of phase II compositing is to allow for the degradation of ammonia by microorganisms within the compost, so that the nitrogen concentration is maintained at approximately 1.5-2% within the compost (de Siqueira et al., 2011). Nitrogen concentrations higher than 2% are detrimental to A. bisporus mycelial cultivation of the substrate (de Siqueira et al., 2011).

Phase III compost. After a 9-day phase II composting regime, the material is considered to be ready for ‘spawning’. Mushroom grain spawn consists of pieces of sterilized millet grain that have been colonized by a specific clonal variety of A. bisporus mycelia. This spawn, plus a slow-release nitrogen supplement, are then added to the phase II compost at a 100:5:1 weight ratio of compost, spawn, and supplement. For our experiments, this phase III compost was mixed in a

10-l plastic autoclave bag (30.5 × 60 cm; VWR International, Atlanta, GA, USA) for 14 days at 21

°C and 70% r.h., after which the compost was thoroughly colonized by white A. bisporus mycelia

(Fletcher et al., 1989).

Spawned compost. For all experiments, spawned compost is synonymous with phase III compost, i.e., it is the 14-day-old phase III compost as described above.

Unspawned compost. For all experiments, unspawned compost refers to 14-day-old phase II compost that was never spawned with mushroom grain spawn, but was still provided with the nitrogen supplement.

Sterilized compost. For all experiments, sterilized unspawned compost refers to the unspawned compost that had been autoclaved at 250 °C for 20 min and then used in experiments immediately after it cooled to room temperature, i.e., approximately 5 h after autoclaving.

Agaricus bisporus mycelia

To determine the potential attractiveness of A. bisporus mycelia alone to L. ingenua females, mycelia from four grain-spawn particles were allowed to colonize a Petri plate of water agar (19 g of agar in 1 l of deionized water) for 12 days (21 °C, 70% r.h.).

Trichoderma aggressivum (green mold)

The green mold Trichoderma aggressivum Samuels & W Gams (Hypocreales) was obtained from the Department of Plant Pathology (Pennsylvania State University). To determine the potential attractiveness of T. aggressivum mycelia to L. ingenua females, T. aggressivum was grown on potato dextrose agar (PDA) and allowed to develop on this medium for 3 days in the dark (25 °C,

70% r.h.) in a growth chamber (Model 1-30 BL; Percival Scientific, Perry, IA, USA). For spore formation, T. aggressivum was allowed to develop for 3 days in constant light under the same temperature and humidity conditions. Trichoderma aggressivum was also grown on spawned compost by placing a 2-cm-diameter disc of 6-day-old T. aggressivum (grown on PDA) in a parafilm-covered PDA Petri plate and allowing the disc to colonize the spawned compost under constant light at 21 °C and 70% r.h. for 7 days.

Flies

The flies used in this study were from a 2-year-old laboratory colony maintained at the University

Park Campus of Pennsylvania State University, Department of Entomology. Flies were positively identified by Dr. Seunggwan Shin (North Carolina State University, Department of Entomology,

Raleigh, NC, USA). This colony was initiated in 2012 by using gravid adult female flies that had been aspirated from the beds of spawned A. bisporus compost in Berks County, PA, USA. The flies in the laboratory colony were reared on a mixture of phase II mushroom compost and nitrogen supplement (100:1) in an environmental growth chamber at 21 °C, 70% r.h., and L12:D12 photoperiod.

To initiate the L. ingenua colony, 355-ml (12-oz.) plastic cups (Solo, Riverview, MI, USA) were filled to the top with the above-described phase II compost–nitrogen supplement mixture.

We found that for rearing L. ingenua, it was not necessary to spawn this compost with A. bisporus mycelia. We could maintain a robust and healthy colony without any A. bisporus at all (unpubl. data). Nine of these compost-filled cups were then placed into mesh cages (BioQuip, Rancho

Dominguez, CA, USA; 30 × 30 × 30 cm), and approximately 50 male and 50 female L. ingenua flies were added to each cage. These cages were then left under colony conditions for 2 days to allow the females to oviposit in the compost mixture provided. After 2 days, the cages were covered with plastic autoclave bags to prevent the compost from drying out. The cages were left under colony conditions until future adult flies emerged approximately 21 days later. This process was repeated until we had a continuously emerging colony. For colony maintenance, cages with newly emerged flies had their cups of compost mixture moved to clean mesh cages. Nine cups of fresh compost mixture were added to the cage of newly emerged adults, and the cage was covered with an autoclave bag. The colonies were kept under the above-mentioned standard conditions until newly emerged adults again emerged approximately 21 days later, and then the process was repeated. Old colony cups (approximately 30 days old) were discarded.

For all experiments, 2-day-old gravid females were used. To obtain gravid females of this

28 age, cages of newly emerged adults had their emergence cups removed, and a cotton wick with a 10% table sugar solution was then provided at the bottom of the cage to allow adults to feed ad libitum. The cages were then kept under colony conditions for 2 days, thereby providing sufficient time to allow the adults to mate. Under such conditions, approximately 90% of females laid fertilized eggs (unpubl. data). For all experiments, it was thus assumed that all females in the cage after 2 days were gravid.

Two-choice, static-flow olfactometer

We used a modified version of the olfactometer design developed by Tibbles et al. (2005); instead of testing the movement of multiple gravid females from a release arena onto test substrates, we tested the movement of individual females in each olfactometer. We used 20 or more identically machined olfactometers that consisted of several sections (Figure 2-1). Each olfactometer had a release arena in which an individual fly was placed to start the test. This arena consisted of a 5-cm-diameter plastic Petri plate (Corning Incorporated, Corning, NY, USA) with two 5-mm-diameter holes drilled into the bottom of the plate, spaced 4 cm apart. Each hole was connected to a pitfall trap on each side of the release arena via 1.5-cm-long, 6-mm-diameter plastic straws. The straws extended from the floor of the release arena into the pitfall trap (Figure

2-1). The base of each straw was 1 mm wider in diameter than the hole drilled in the release arena, such that each straw laid flush with the floor of the release arena when the arena was set on top of it. The pitfall traps consisted of 30-ml (1-oz.) plastic portion-cups (Dart Solo, Harrisburg,

PA, USA) fixed with a snap-top plastic lid (4 cm diameter, 3 cm high). A 6-mm-diameter hole was drilled into the lid, and the plastic straw was snuggly inserted into this hole so that it extended

7.5 mm into the plastic-cup pitfall trap. We observed that once females entered a pitfall trap,

29 none of them were able to move back in the opposite direction into the release arena. Thus, the measurement of the attractiveness of the material in one pitfall trap compared to another was measured by each female’s irrevocable choice for that trial.

Figure 2-1 Two-choice, static-flow olfactometer. (A) The olfactometer consists of a 5-cm-diameter release arena (R) and two 4-cm-diameter, 3-cm-high pitfall traps (P). (B, C) A 5-mm-diameter, 1.5-cm-long plastic straw (S) extends from each of two holes (arrow in C) in the release arena floor to the interiors of the two pitfall traps. Note that the top of the straw lays flush with the release arena floor, this is critical in facilitating female Lycoriella ingenua entrance into the pitfall traps. (D) A single female fly (red arrow) is shown here as she enters a straw leading to a pitfall trap.

A layer of Parafilm was used to seal the release arena and thereby prevent flies from escaping and also to maintain sufficient humidity within the arena such that none of the flies would die during the 24 h of each replicate. Our previous experiments found that without this parafilm covering, ca. 10% of flies would die during the 24-h period. An individual 2-day-old gravid

30 female was then aspirated into the arena through one of the holes drilled in the bottom, and the arena was gently placed on top of the two pitfall traps so that it rested on top of the straws that were fixed into the lids of the pitfall traps. These traps were then placed in one of two growth chambers under colony conditions (21 °C, 70% r.h., L12:D12 photoperiod) for 24 h. After 24 h, the location of each female fly in the individual olfactometers was recorded (i.e., whether in either of the pitfall traps or in the release chamber). At the end of the experiments, all pitfall traps and plastic straws were discarded, and the release arenas were washed with hot water and detergent, soaked in a 70% ethanol solution, and allowed to air dry for 24 h before reuse.

Olfactometer experiments

Six sets of choice tests were performed over several weeks in the following combinations: (1) spawned vs. unspawned compost, (2) unspawned compost vs. A. bisporus white button mushrooms (fruiting bodies; i.e., the white button mushroom that is sold in stores) (ca. 3 cm diameter), (3) unspawned non-sterilized compost vs. unspawned sterilized compost, (4) A. bisporus growing on water agar vs. blank water agar, (5) T. aggressivum growing on PDA vs. blank

PDA, and (6) T. aggressivum growing in spawned compost vs. spawned compost alone.

For each set of choice tests, 20 2-day-old females were tested each night in individual static-flow olfactometers. The 20 olfactometers were divided into two identical groups of 10 that were placed in separate growth chambers. The positions of olfactometers within both growth chamber varied randomly. This procedure was carried out over three consecutive nights, resulting in 60 replicates for each experiment, e.g., three groups of 20 females on three consecutive nights were allowed to choose between two substrates whose attractiveness was being compared.

Statistical analysis

All analyses were conducted using PRISM v.5.0 (GraphPad Software, San Diego, CA, USA).

Differences among percentages of female flies found in individual pitfall traps for each choice- test experiment were assessed via conventional goodness of fit χ2 analysis, testing frequencies observed vs. expected (α = 0.05). Non-responders remaining in the release chamber were excluded from analysis.

Results

In the olfactometer choice tests, the presence of A. bisporus mycelia in compost did not increase attraction of gravid female L. ingenua (spawned vs. unspawned compost: χ2 = 0.019; d.f. = 1 P =

0.88; Figure 2-2A). Mature, unspawned compost is more attractive to L. ingenua females than sterilized unspawned compost (χ2 = 8.64, d.f. = 1, P = 0.003; Figure 2-2B). Lycoriella ingenua females were more attracted to unspawned compost (lacking mycelia or other life stages of A. bisporus) than to an A. bisporus fruiting body (χ2 = 4.92, d.f. = 1, P = 0.02; Figure 2-2C). Hence, volatiles released from actively metabolizing microorganisms within the compost may explain female attraction to compost, with or without A. bisporus.

More females responded to A. bisporus mycelia growing on a water agar medium than to water agar medium alone, but this trend was not significant (χ2 = 3.1, d.f. = 1, P = 0.07; Figure 2-

3A). Gravid female L. ingenua were attracted to T. aggressivum growing both on spawned compost compared to spawned compost alone (χ2 = 4.74, d.f. = 1, P = 0.03; Figure 2-3B) and on

PDA compared to PDA alone (χ2 = 27.9, d.f. = 1, P<0.0001; Figure 2-3C). Hence, something emitted

32 by this fungus, even without compost, appears attractive.

Figure 2-2: Percentage of female Lycoriella ingenua gravid females attracted to various mushroom compost– related substrates in two-choice, static-flow olfactometer assays: (A) unspawned vs. spawned compost, (B) unspawned vs. sterilized unspawned compost, and (C) unspawned compost vs. fruiting bodies of Agaricus bisporus. Numbers in the bars indicate the number of females choosing either chamber or staying in the release arena (not responding).

Figure 2-3: Percentage of female Lycoriella ingenua gravid females attracted to Agaricus bisporus or Trichoderma aggressivum mycelia in two-choice, static-flow olfactometer assays: (A) A. bisporus growing on water agar vs. water agar alone, (B) T. aggressivum mycelium growing in spawned compost vs. spawned compost alone, and (C) T. aggressivum mycelia growing in potato dextrose agar vs. potato dextrose agar alone. Numbers in the bars indicate the number of females choosing either chamber or staying in the release arena (not responding).

Discussion

This is the first evidence, to our knowledge, of attraction of sciarid flies to T. aggressivum actually growing in mushroom compost. Future experiments should build on this relationship to examine the spore- or mycelium-carrying vectorial capacity of L. ingenua adults that might visit this compost and be exposed to T. aggressivum life stages. Also the vectorial capacity of L. ingenua larvae should be examined once this mold has become established in mushroom growing beds.

The results of the first three choice-test experiments suggest that flies entering a mushroom house might not be attracted to developing A. bisporus mycelia or fruiting bodies, but rather might be attracted to the compost itself, as suggested previously (Anon, 1982). Thus, in our study with L. ingenua (Figures 2-2 A–C) and in two-choice assays with Lycoriella castanescens

(Lengersdorf) (Tibbles et al., 2005), flies of both species were found to be similarly attracted to mushroom compost with and without A. bisporus. However, other studies have shown that compost that is densely colonized by A. bisporus mycelia is actually repellent to ovipositing L. ingenua females (Kielbasa & Snetsinger, 1981). A possible ecological explanation may be that a negative relationship exists between larval survival and the amount of A. bisporus mycelia growing in compost, as suggested by (Tibbles et al., 2005).

The weak difference in attraction to A. bisporus mycelia growing on water agar and water agar alone may have been due to a relative paucity of volatiles from either chamber, as suggested by the unusually high number of non-responders in this experiment. Sciarid flies have been shown to achieve approximately 50% survival on just A. bisporus mycelia and fruiting bodies

(O'Connor & Keil, 2005); so, being attracted to these fungi is not totally detrimental to survival

(Figure 2-3A). The goal of the last two choice-test experiments (Figures 2-3B, C) was related to

34 the anecdotal evidence that L. ingenua is a vector of T. aggressivum and transports this noxious green pest mold into mushroom growing houses, leading to infestations. If T. agressivum spores or mycelia themselves were attractive to L. ingenua females or males, then the flies might acquire green mold life stages and carry them into mushroom houses and infect the mushroom crop.

Previous studies showed that L. castanescens development is significantly reduced on compost heavily colonized by A. bisporus mycelia (Tibbles et al., 2005), suggesting that if fly populations are excluded from a mushroom growing house in the beginning of the cropping cycle, the chances of a large infestation may decrease as mycelia develop. These results, plus the currently demonstrated attraction of females to mature phase II compost over sterilized compost, may guide mushroom growers in their efforts to exclude L. ingenua flies. Growers may use this information to control the incoming colonizing female flies in the early crop stages, before they oviposit on the fresh phase II spawned compost that has just been added.

A study with another sciarid species, Bradysia impatiens (Johannsen), indicated that survival was significantly reduced when flies were forced to develop in sterilized compared to non-sterilized peat moss (Olson et al., 2002), suggesting that there is a fitness disadvantage to flies that lay their eggs in sterilized compost. Our future work will be aimed at isolating and identifying the odorant compounds of unspawned compost that are attractive to gravid L. ingenua. Our static-flow olfactometer design, with its ability to determine the attraction-related behavior of individual flies, may prove to be an effective tool for screening synthetic compounds for their potential attractiveness to L. ingenua as well as to other fungus gnat species, such as the phorid pest M. halterata.

Acknowledgments

This project was supported by USDA/NIFA/SCRI grant no. 2012-51181-19912 (project director

David M Beyer). We thank David Beyer, John Pecchia, and Nina Jenkins for their contributions to this work, and Dr. Julie Todd for editing a penultimate version of this manuscript.

Chapter 3

Attraction, oviposition and larval survival of the fungus gnat,

Lycoriella ingenua, on fungal species isolated from adults, larvae,

and mushroom compost

*Published December 2016: Cloonan KR, Andreadis SS, Chen HB, Jenkins NE, Baker TC. Attraction, oviposition and larval survival of the fungus gnat, Lycoriella ingenua, on fungal species isolated from adults, larvae, and mushroom compost. Plos One. 2016;11(12).

Abstract

We previously showed that the females of the mushroom sciarid, Lycoriella ingenua (Dufour,

1839) (Diptera: Sciaridae), one of the most severe pests of the cultivated white button mushroom, Agaricus bisporus (J.E. Lange) Emil J. Imbach (Agaricales: Agaricaceae), are attracted to the mushroom compost that mushrooms are grown on and not to the mushrooms themselves.

We also showed that females are attracted to the parasitic green mold, Trichoderma aggressivum. In an attempt to identify what is in the mushroom compost that attracts female L. ingenua, we isolated several species of fungi from adult males and females, third instar larvae, and mushroom compost itself. We then analyzed the attraction of females to these substrates using a static-flow two choice olfactometer, as well as their oviposition tendencies in another type of assay under choice and no-choice conditions. We also assessed the survival of larvae to

37 adulthood when first instar larvae were placed on each of the isolated fungal species. We found that female flies were attracted most to the mycoparasitic green mold, T. aggressivum, to

Penicilium citrinum isolated from adult female bodies, and to Mycothermus thermophilium isolated from the mushroom compost. Gravid female flies laid the most eggs on T. aggressivum,

Aspergillus flavus isolated from third instar larval frass, Aspergillus fumigatus isolated from adult male bodies, and on P. citrinum. This egg-laying trend remained consistent under no-choice conditions as females aged. First instar larvae developed to adulthood only on M. thermophilium and Chaetomium sp. isolated from mushroom compost, and on P. citrinum. Our results indicate that the volatiles from a suite of different fungal species act in tandem in the natural setting of mushroom compost, with some first attracting gravid female flies and then others causing them to oviposit. The ecological context of these findings is important for creating an optimal strategy for using possible semiochemicals isolated from these fungal species to better monitor and control this pestiferous mushroom fly species.

Introduction

The fungus gnat, Lycoriella ingenua (Dufour 1839) (Diptera: Sciaridae), formerly known as

Lycoriella mali (Wetzel et al., 1982), causes some of the most severe insect damage to cultivated white button mushrooms, Agaricus bisporus (J.E.Lange) Emil J. Imbach (Agaricales: Agaricaceae), in the United States. White button mushrooms are a high value crop, with an estimated 394,650 tons produced between 2012 and 2013 grossing nearly US $1.4 billion (USDA, 2014). Fungus gnats are especially effective pests of white button mushrooms considering this pests’ low

38 economic threshold (Hussey, 1968; White, 1986), high fecundity (Cantelo, 1988), and the range of damage characteristics it inflicts on the crop. This damage includes: direct larval feeding on developing A. bisporus mycelia in the growing compost media (Cantelo, 1979b; Grewal et al.,

1993; Kielbasa & Snetsinger, 1980); larval competition with developing A. bisporus mycelia for nutrients in the compost (Binns, 1980b); and a reduction in mycelial growth due to larval excrement (frass) (Hussey & Gurney, 1968). Mushroom growers have reported that L. ingenua adults vector into mushroom houses the green mold, Trichoderma aggressivum, one of the most severe fungal pathogens of A. bisporus, (Jess & Schweizer, 2009). Although no experiments have examined the ability of L. ingenua to vector T. aggressivum, scanning electron microcscopy images revealed that adult L. ingenua adults carry spores of the plant pathogen Fusarium fungicola on their bodies (Shamshad et al., 2009) thus may potentially vector the spores of other pathogenic fungi.

Methods to control this fly include compost drenches with the insecticide Imidacloprid

(Shirvani-Farsani et al., 2013), the juvenile hormone (JH) analog methoprene, and the chitin synthase inhibitor diflubenzuron (Cantelo, 1983). However, several factors have shown the necessity for changing standard pest management practices for this species. Early reports of rapid resistance development by the flies to pyrethroid insecticides (Bartlett & Keil, 1997; Brewer

& Keil, 1989b; White & Gribben, 1989a), the difficulty in excluding flies from mushroom houses, and a growing consumer demand for chemical-free mushrooms (Tibbles et al., 2005), have all contributed to a realization that novel, sustainable pest management tools are necessary to provide mushroom growers with biorational ways to keep pest populations down.

For instance, recent work has resulted in the successful isolation and partial identification of a female-produced sex pheromone component of L. ingenua that is attractive to male flies

(Andreadis et al., 2015a). Although this component is clearly different from the pheromone that was previously misidentified (Kostelc et al., 1980) and later shown to be behaviorally inactive

(Gotoh et al., 1999) the precise stereoisomeric structure of this compound needs to be characterized before it can be synthesized and used for monitoring and possibly mass trapping of males within and outside of mushroom houses. What would further add to an arsenal of semiochemical pest management tools for L. ingenua would be if attractants for female flies could be found for use in monitoring and mass trapping of females. Thus, the goal of this research was to isolate and identify potential fungal sources from the mushroom growing compost that are best at attracting gravid female L. ingenua flies. We wanted to make sure that in our efforts we would specifically include fungal species that are suspected of being vectored in and out of the mushroom-growing compost by both adult and larval flies.

We previously showed, using a static flow olfactometer, that gravid female L. ingenua flies were highly attracted to T. aggressivum both in culture and growing on spawned compost

(Cloonan et al., 2016a). “Spawned compost” refers to mushroom compost that has A. bisporus mycelia already developing in it. These results suggest that the belief by growers about L. ingenua vectoring green mold spores into a growing house may be true, and the flies may be responsible for high infestations of T. aggressivum. We also showed that gravid female L. ingenua flies are equally attracted to spawned compost and unspawned compost (compost with no A. bisporus mycelia). Attraction to unspawned compost was significantly reduced when it was sterilized, suggesting that gravid females are attracted to actively metabolizing microorganisms within the

40 unspawned compost, not the developing A. bisporus mycelia. These results support early work showing that L. ingenua development on sterilized compost is significantly reduced versus non- sterilized compost (Kielbasa & Snetsinger, 1981). Because of these findings, Kielbasa and

Snetsinger (1981) suggested that, “L. mali (now L. ingenua) obtains dietary supplementation from bacteria and/or fungi which survive pasteurization.” Considering that the microbial community of mushroom compost is both diverse and temporally dynamic (Zhang & Sun, 2014) it is reasonable to assume that females are attracted to the volatile profile of one or more of these metabolizing microorganisms.

As a result of the abovementioned findings we hypothesized that female flies might be attracted to one or several of the actively metabolizing fungal species we had found to be present in mushroom compost. Previous work showed that gravid L. ingenua females were differentially attracted to fungal species grown on a variety of substrates (Frouz et al., 2002). The authors suggested that this differential attraction was due to the different suite of volatiles produced as a result of the same fungal species utilizing different substrates. We therefore designed experiments to use some fungal cultures that we isolated from compost that we now grew only on potato dextrose agar (PDA) to see the extent to which female L. ingenua would be attracted to them and lay eggs in them.

We also included fungal species that we were able to obtain from the bodies of males and females that had been collected from compost, given previous evidence that L. ingenua carry other fungi on their bodies (Shamshad et al., 2009). Such phoretic fungi might have extra importance in the lives of L. ingenua and might therefore be more attractive to females or provide more powerful oviposition cues. For these experiments we used two-choice attraction

41 assays, two-choice oviposition assays, and no-choice oviposition assays in order to try to find fungal species that might be more attractive to female L. ingenua than T. aggressivum and might emit volatiles that can be isolated and identified for possible use in monitoring traps in mushroom houses.

Materials and Methods

Insects

The fungus gnats used in this study were from a three-year-old laboratory colony maintained at the University Park Campus of The Pennsylvania State University, Department of Entomology, and were positively identified by Dr. Seunggwan Shin (North Carolina State University,

Department of Entomology, Raleigh, NC). All L. ingenua culturing methods including colony initiation were identical to those described by Cloonan et al., (2016a). Flies were reared on a mixture of unspawned mushroom compost and nitrogen supplement (100:1) in an environmental growth chamber at 21°C, 70% r.h., and a L12:D12 photoperiod. Nine plastic Solo cups (355-ml, Solo, MI, USA) were filled to the top with the unspawned compost–nitrogen supplement mixture and placed into a mesh cage (BioQuip, CA, USA; 30 x 30 x 30 cm) with approximately 100 male and 100 female L. ingenua flies. These cups were then left under colony temperature and relative humidity conditions for two days to allow the flies to mate and the females to oviposit in the compost mixture provided. After two days, the cages were covered with plastic autoclave bags to prevent the compost from drying out. The cages were left under colony conditions until future adult flies emerged approximately 21 days later.

To obtain females of a specific age for experimentation, newly emerged adults (0-12 hours old) present on the screened walls of their cages had their nine plastic Solo emergence cups removed, and then a paper towel with a 10% table sugar solution was provided at the bottom of the cage to allow adults to feed ad libitum. The cages were then kept under colony conditions for the appropriate amount of time in order to obtain cages of gravid females of a specific age for different experiments. We found that adults would live under these conditions up to 12 days with ca. 50% of adults dying by the eighth day (unpublished data).

Fungal Cultures

Fungal species present in mushroom compost

Two cultures, Scytalidium thermophilum and Chaetomium sp., were obtained from the Plant

Pathology Department at Penn State University (Figure 3-1). These two fungal species were chosen because of their relatively high abundance in mushroom compost (Straatsma et al., 1994) and that therefore they might be likely candidates for attracting female L. ingenua to the compost. A culture of T. aggressivum was also obtained from the Plant Pathology Department at

Penn State University and was used as a control in all two-choice attraction and oviposition experiments because it is the most attractive substrate to gravid females that we have found thus far. Our goal was to isolate fungal species that were highly attractive to L. ingenua females, and to find those that might be even more attractive than T. aggressivum.

Fungal species isolated from adult flies

All fungi obtained from adult flies for these experiments were cultured on plates of potato dextrose agar (PDA) (8.5 cm in diameter; Crystalgen Inc, NY, USA) that were kept in an

43 environmental growth chamber at 21°C, 70% r.h., and a L12:D12 photoperiod for 1 hr. In order to ensure that the fungal species we isolated from flies actually did originate from the bodies of adult flies and not from contamination from elsewhere in the environment, we performed three different culturing treatments: 1) placing surface-sterilized adult flies on PDA; 2) placing non- sterilized adult adult flies on PDA; and 3) exposing empty plates of PDA to our work station environment. To obtain surface-sterilized flies, ten male and ten female newly emerged adults from the mesh colony cages were aspirated and immediately placed in a freezer (18°C) for 1 min to temporarily immobilize them. They were then set on top of two layers of dry paper towels and drenched in a 5% bleach solution, then immediately rinsed with deionized (DI) water. The adults were then transferred to another set of paper towels to dry. These dry, surface-sterilized flies were then individually placed on plates of PDA (8.5 cm in diameter) and kept in an environmental growth chamber at 21°C, 70% r.h., and a L12:D12 photoperiod for 1 hr. After 1 hr, sterilized flies were discarded and the plate was covered with parafilm, then placed back into the growth chamber for 72 hr to monitor for colony formations.

To obtain non-surface-sterilized adults, ten male and ten female flies were treated the same way except they were not drenched in bleach or DI water before being placed on plates of

PDA. They were then put into the growth chamber under the same conditions. We also exposed

10 plates of PDA to the open air of our work station for ca. 10 sec, as well as exposing them briefly to the clean tips of the brushes used to transfer sterilized and non-sterilized adults to ensure any fungal growth we observed originated from the adult fly bodies. After 72 hr Individual fungal colonies from all treated plates were then isolated under sterile conditions by placing them onto

44 new Petri plates of PDA that were kept under colony conditions for ca. 2 weeks or until we could verify visually that we had isolated a single individual fungal species.

Fungal species isolated from larvae

To obtain sterilized larvae, we collected ten third instars reared on unspawned compost and drenched them in bleach and DI water the same way as we treated the sterilized adults above, except larvae were not cold-treated first to immobilize them as we had done for adults. Larvae were then transferred individually to plates of PDA that were then placed in an environmental growth chamber at 21°C, 70% r.h., and a L12:D12 photoperiod for 1 hr. After 1 hr, sterilized larvae were removed from the plates of PDA which were then covered with parafilm and placed back into the growth chamber for 72 hr to monitor for colony formations.

Ten non-sterilized third instars were treated the same way as the larvae above except they were not drenched in bleach or DI water before being placed on plates of PDA and put into the growth chamber under the same conditions. After 72 hr individual fungal colonies from all treated plates were then isolated under sterile conditions by placing them onto new Petri plates of PDA and kept under colony conditions for ca. 2 weeks, or until we could verify visually that we had isolated a single individual fungal species.

Figure 3-1: Pure cultures of the seven fungal species used in these experiments. These fungi are grown on potato dextrose agar, a nutrient agar that many species of fungi are able to thrive on.

DNA Extraction, PCR, Sequencing and Identification of Isolated Fungi

DNA was extracted from mycelial mats grown in potato dextrose liquid media. Small plugs (2 mm in diameter) of each culture, grown on plates of PDA for 14 days, were placed into 250 ml flasks of potato dextrose liquid media and allowed to grow at 27 ± 1 ° C on a shaker rotating at 180 rpms for 14 days. After 14 days mycelial mats were harvested by filtering the liquid media through filter paper (Whatman® qualitative filter paper, Grade 1, 42.5 mm diam) using gravity and stored at -80°C. Samples were then freeze-dried in a Virtis Advantage XL Lyophilizer, held for

5 hr after the pressure dropped below 200 mT, and slowly increased to room temperature over

7 hours. DNA was extracted from each sample by grinding approximately 40 mg of tissue in liquid nitrogen then adding 600 μl of Nuclei Lysis Solution (1% SDS, 10 µl EDTA). These samples were then incubated at 65°C for 30 min. After incubation, 1 μl of 1 mg/mL RNase was added and samples were then incubated at 37°C for 15 min. Samples were then cooled to room temperature for 5 min. To this cooled sample 200 μl of a 10M ammonium acetate solution was added. The sample was then cooled and incubated for 5 min on ice, then centrifuged at 16,000 x g for 3 min. The supernatant was then transferred to clean Eppendorf tubes containing 600 μl of room temperature isopropanol and centrifuged at 16,000 x g for 3 min. A 70% ethanol solution was then added to the samples and they were centrifuged at 16,000 x g for 1 min. The pellet was then air dried and suspended in 100 μl of sterile water and rehydrated overnight at 4°C.

These samples and two sets of primers (Bt2a/Bt2a) and (cmd5/cmd6) were amplified by

PCR. These PCR products were then sequenced by Sanger DNA sequencing (Applied Biosystems

3730XL) at Genomics Core Facility service at The Pennsylvania State University. Results were

46 edited using Geneious 9.0.5 software. These were then compared to Genbank to determine their identity.

Two-Choice, Static Flow Olfactometer Attraction Assays

To examine the relative attractiveness to gravid females of all isolated fungal species, we used a modified static-flow two choice olfactometer previously described in Cloonan et al. (2016a)

(Figure 3-2). Each olfactometer was precisely machined to be as identical as possible. Each one consisted of a glass Petri plate “release arena” (VWR, Radnor, PA, USA), 5 cm in diameter, connected at the bottom to two 2 ml glass vials (12 x 32 mm ) (Supelco, Bellefonte, PA, USA) acting as pitfall traps. The pitfall traps were connected to the release arena by two 3 mm diam. holes spaced 3.5 cm apart that had been drilled into the release arena floor. The pitfall traps were then connected to these holes in the arena by means of 4 mm i.d. glass tubes (100-200 µl microdispensor replacement tubes; Drummond Scientific, Broomhall, PA, USA). The tubes were

3 cm in length and open at both ends. These tubes were affixed into the lid (screw thread

PTFE/Silicone septum 9 mm in diameter) of the pitfall traps such that they extended 1 cm into the interior of the traps. Because the outer diameter of each if these glass connecting tubes was

1 mm wider than the 3 mm holes drilled into the release arena floors, the release arenas rested directly on top of the two pitfall traps, with an opening that went directly from the release arena floor into the pitfall traps (Figure 3-1). The utility of these static-flow olfactometers lies in the fact that once a gravid female fly makes her choice and enters one of the pitfall traps containing an attractive substrate, she is unable to pass back through to the release arena and her choice can be easily and accurately recorded.

Figure 3-2: (a) The olfactometer consists of a 5-cm-diameter glass Petri dish release arena attached to two 2-ml glass vial pitfall traps spaced 3.5-cm apart through two 3-mm-diameter holes. A 4-mm-diameter, 1.5-cm-long glass tube extends into each of the two 2-ml glass vial pitfall trap lids. These glass tubes are affixed into the lids of the 2-

ml glass vial pitfall traps. (b) The tips of the tubes extend out of the 2-ml glass vial pitfall traps and are positioned

directly under 3-mm diameter holes drilled into the release arena floor. Because the tubes are 1-mm diameter

larger than the drilled holes, the release arena is able to be set directly over these holes such that they lay flush

with the release arena floor. This is critical in facilitating female Lycoriella ingenua entrance into the pitfall traps.

Each pitfall trap contained one disk of test substrate punched out of the above-described

2-week-old fungal cultures grown on PDA. These disks (7 mm in diameter) were punched out of

PDA cultures using the wide end of a yellow 1-200 µl pipette tip (VWR International, Radnor, PA,

USA). For each two-choice test, one pitfall trap always contained T. aggressivum and the other pitfall trap contained one of the other five fungal cultures described. Individual 2-day-old gravid female flies were aspirated into the release arena, whose holes were then gently positioned so

48 that they rested over the two connecting tubes of the two pitfall traps that contained disks of fungal cultures. Once set in position, the olfactometers were placed under colony conditions and females were allowed to make their choice over a 12 hr period of 6 hr of light and 6 hr of dark.

At the end of each 12 hr experiment the locations of females were recorded and the flies were then discarded. Fungal disks were discarded, and the release arenas and pitfall traps were then soaked and washed in detergent and water, rinsed, and then soaked in a 70% ethanol bath for 1 hr before being air dried overnight before re-use.

These olfactometer experiments were conducted in 30 concurrently-running olfactometers each night. Paired fungal cultures were presented to individual female flies in a complete-block replicate having the six following treatments:

1) Trichoderma aggressivum versus Aspergillus niger isolated from adult male flies

2) Trichoderma aggressivum versus Aspergillus flavus isolated from larval frass

3) Trichoderma aggressivum versus Aspergillus fumigatus isolated from adult female flies

4) Trichoderma aggressivum versus Penicillium citrinum isolated from adult female flies

5) Trichoderma aggressivum versus Mycothermus thermophilium obtained from Plant

Pathology at The Pennsylvania State University

6) Trichoderma aggressivum versus Chaetomium sp. obtained from Plant Pathology at The

Pennsylvania State University

Five replicates were run each night (1 female per each of 6 treatments X 5 replicates = 30 females per night) over 12 nights, using a different batch of females each night. Thus, overall, 60 different

49 females were tested for their response to each treatment (the six fungal culture choices) over 12 nights.

Two-Choice Oviposition Assays

We also examined the oviposition preferences of individual two-day-old gravid females in response to the above-described treatments, i.e., the six fungal culture pairings, grown on PDA.

These assays were performed in covered 5-cm-diameter Petri dishes from the same manufacturer as used in the olfactometer experiments above. These Petri dishes had no holes drilled into their floors. For each replicate two 7-mm-diameter disks of two-week-old fungal isolates grown on PDA, obtained as described above, were placed 4 cm apart on the Petri dish floor. An individual two-day-old gravid female was then aspirated into the Petri dish that was then promptly covered with its glass lid and placed under colony conditions for 6 hr of light and

6 hr of dark. At the end of each 12 hr experiment, the numbers of eggs were counted on each fungal disk, and the disks were then discarded. The Petri dishes were then soaked and washed with detergent-water, rinsed, then soaked in a 70% ethanol bath for 1 hr before being air dried overnight before re-use.

Fungal cultures were presented to 18 individual female flies in the described paired combinations each night, forming 3 complete-block replicates of the six treatments each night.

The experiment was run over 5 nights, using a different batch of females each night. Thus, overall,

15 different females were tested for their response to each treatment over 5 nights.

No-Choice Assays

We also performed no-choice oviposition assays to determine the degree to which females, if they had no choice, would oviposit on both the preferred and non-preferred fungal isolates that were used in the two-choice oviposition assays above. We conducted these experiments because the attractive presence of T. aggressivum in every two-choice oviposition assay above may have affected our ability to measure females’ acceptance or rejection behaviors related to these fungal substrates. In these no-choice assays we recorded the daily number of eggs laid by females to see if their egg deposition might become greater even on less-preferred fungi as females matured and died. These assays were performed in the same glass Petri dishes that were used in the two- choice oviposition assays. Individual two-week old disks (7 mm in diameter) of each of the following fungal cultures were placed directly in the center of the glass Petri plate:

1. Blank PDA

2. Aspergillus fumigatus isolated from adult female flies

3. Aspergrillus flavus isolated from larval frass

4. Trichoderma aggressivum

5. Chaetomium sp. obtained from Plant Pathology at The Pennsylvania State University

6. Penicillium citrinum isolated from adult female flies

7. Mycothermus thermophilium obtained from Plant Pathology at The Pennsylvania State

University

8. Aspergillus niger isolated from adult male flies

The goal of these no-choice experiments was to examine whether female flies would oviposit on less-preferred fungal isolates as they aged. However, we were unable to monitor the oviposition behavior of individual female flies throughout their lifetime, because even unmated flies will oviposit all of their eggs on a preferred substrate and die within a few hours (unpublished data). Because of this we developed a method to monitor the oviposition behavior of replicated individual flies of the same age under no-choice conditions, from newly emerged females to seven-day-old females. To obtain mated females of the same age, individual cages of male and female flies of a specific age were held under colony conditions until they were used for experimentation. All of the females in each one of these cages were thus considered mated as they were held at a ratio of ca. 50:50 male to females within each cage of specifically aged individuals. Individual female flies, between the ages of newly emerged to seven-days-old, were aspirated into a Petri dish containing a single fungal culture disk placed in the center. The Petri dishes were then placed under colony conditions for 12 hr (6 hr of light and 6 hr of dark). At the end of each 12 hr experiment the numbers of eggs were counted on each fungal disk, and the disks were then discarded. The release arenas were then cleaned as before.

Three replicates with respect to female age were run during each no-choice oviposition assay

(one female from each of the eight age classes X 3 replicates = 24 females per night), using a different fresh batch of females aged 0-7 days each session. Thus we used 24 Petri-dish assay chambers during each 12-hr session. Individual females from all 8 age groups were tested during each session against three of the eight different fungal cultures in our regime (including blank

PDA). Due to space and supply constraints we could not test all the fungal cultures simultaneously against all seven age-group females during a single 12-hr session. The array of three fungal

52 treatments used in each session was chosen randomly until 15 females had been assayed against each culture during five different sessions.

Larvae-to-Adult Survival Assays

We transferred individual first instar larvae to agar disks of each isolate to examine larval survival to adulthood. First instar larvae were carefully transferred to two-week-old cultures grown on

PDA (7 mm in diameter) that were placed in a Petri dish (the same as was used in the no-choice oviposition assays; 5 cm in diameter). The Petri dishes were then covered in parafilm and placed back in colony conditions at 21°C and a 12:12 photoperiod (light:dark). These dishes were then monitored each day for larval development and subsequent adult emergence. 30 individual first- instar larvae were tested per isolated fungal species and a PDA control. For each replicate, five first-instars were placed on each of the seven fungal cultivars plus blank PDA and observed through to adulthood or death. Six of these replicates were run over the course of the experiment.

Statistical Analysis

All statistics were conducted using Prism 5.0 software (GraphPad Software Inc., San Diego, CA,

USA). All datasets were first analyzed for normality via the D'Agostino & Pearson Omnibus

Normality Test. Differences among proportions of female flies found in individual pitfall traps for the two-choice olfactometer experiments were assessed via Chi-square goodness of fit test after correcting for continuity with Yates’ correction factor. Non-responders remaining in the release chamber were excluded from these analyses. The two-choice oviposition data were compared via the Mann–Whitney U test. The differences in mean egg numbers among each age group for the no-choice oviposition data were analyzed via the Kruskal-Wallis test, and differences in

53 means were then compared via Dunn’s Multiple Comparisons. The differences in mean survival from larvae to adult were analyzed via the Kruskal-Wallis test, and differences in mean survival on different fungal substrates were then compared via Dunn’s Multiple Comparisons.

Results

Sequence comparisons referencing GenBank positively identified the following fungal isolates to species: Aspergillus flavus was isolated from larval frass, A. niger from adult male bodies, and A. fumigatus from adult female bodies, and Penicillium citrinum was isolated from adult female bodies.

Females were attracted more to T. aggressivum than to Aspergillus niger in two-choice pitfall traps (P < 0.0001; χ2 = 22.41 ) (Fig 3-3). This greater relative attraction to T. aggressivum was reflected also in our two-choice oviposition assays, in which female flies laid significantly more eggs on T. aggressivum than on A. niger (P = 0.02) (Fig 3-4). A different relationship between attraction and oviposition was seen when females had to choose between T. aggressivum and A. flavus isolated from third instar frass. In this case, females were more attracted to T. aggressivum than to A. flavus in two-choice olfactometer assays (P = 0.0005 ; χ2 =

10.95 ) (Fig 3-3), but they oviposited significantly more eggs on A. flavus than on T. aggressivum

(P = 0.001) (Fig 3-4).

Figure 3-3: Mean number (± SEM) of two-day-old gravid female Lycoriella ingenua flies attracted to various pure fungal cultures grown on potato dextrose agar in two-choice, static-flow olfactometer assays. Each horizontal bar

is the mean of responses of 5 females to each treatment over 12 replicates (N=12). The mean number of non-

responders (± SEM) for each combination is included in parentheses to the right. Female choices for each pair of

cultures were analyzed via chi square . All non-responders were excluded from the analysis.

Figure 3-4: Mean number (± SEM) of eggs laid by two-day old gravid female L. ingenua flies on various pure fungal

cultures grown on potato dextrose agar in two-choice oviposition assays. Each horizontal bar shows the mean number of eggs laid on the two choices of fungi by 15 two-day-old gravid female flies, tested in groups of three in

three individual chambers with 5 different cohorts of females being tested over 5 different nights. All data were

non-normally distributed and differences between the mean number of eggs deposited on each pair of fungal

cultures were analyzed via the Mann–Whitney U test (two-tailed, df = 14).

For the third fungal species isolated from L. ingenua, females were more attracted to T. aggressivum than to A. fumigatus isolated from adult female bodies (P < 0.001; ; χ2 = 15.02) (Fig

3-3), but they oviposited similar numbers of eggs on these two fungal species (P = 0.09) (Fig 3-4).

In response to P. citrinum, the Penicillium fungal species isolated from adult female flies, females were equally attracted to P. citrinum and to T. aggressivum (P = 0.17; χ2 = 1.42 ) (Fig 3-3), and they also laid similar numbers of eggs on both of these fungal species in the two-choice oviposition assay (P = 0.26) (Fig 3-4).

For one of the fungal species isolated from compost, M. thermophilium, females were equally attracted to T. aggressivum and M. thermophilium in our two-choice olfactometer tests

(P = 0.08; χ2 = 2.42) (Fig 3-3). Mycothermus thermophilium is the fungal species known to be present in high abundance in mushroom compost and that we did not find on adult bodies.

Despite these equal attraction levels, females laid significantly more eggs on T. aggressivum (P =

0.0002) (Fig 3-4) than on M. thermophilium, almost completely avoiding oviposition on this species in two-choice oviposition assays. In response to the Chaetomium sp., another fungus we isolated from mushroom compost but that we could not isolate from adults’ bodies, females were significantly more attracted to T. aggressivum than to Chaetomium sp., in the olfactometer assays (P = 0.002; χ2 = 8.42 ) (Fig 3-3). The attraction results were reflected later in the two-choice oviposition assays: when given a choice between these two species, females laid significantly more eggs in T. aggressivum than Chaetomium sp. (P = 0.007) (Fig 3-4).

The no-choice oviposition experiment showed females’ pure responses to invest or not invest significant numbers of eggs on various fungal substrates. Female L. ingenua that were newly emerged to three-days-old laid similar, small numbers of eggs on all fungal cultures under no-choice conditions (P > 0.5) (Fig 3-5 a-d). However, after reaching the age of four-days-old, females began laying significantly more eggs on those fungal substrates they had shown a preference for in the two-choice oviposition assays, which included A. fumigatus, A. flavus, T. aggressivum, Chaetomium sp., and P. citrinum. The non-preferred fungi from those assays, M. thermophilium and A. niger (P < 0.5) (Fig 3-5 e-h), evoked very little oviposition even from older females in these no-choice tests.

Figure 3-5: Mean number (± SEM) of eggs laid by 0- (day of emergence) to-7 day-old female L. ingenua flies on various pure fungal cultures grown on potato dextrose agar in no-choice oviposition assays. A total of 15 flies of a

particular age were tested for their tendency to lay eggs on each of the 8 fungal cultures under no-choice

conditions, resulting in a total of 120 females (15 flies x 8 fungal cultures) of each age that were tested for oviposition on all fungal cultures. There were 8 age groups tested (panels a-h), and thus 960 different female flies

were used in this experiment. All data were non-normally distributed and differences among mean egg numbers within each age group were analyzed via the Kruskal-Wallis test. Differences between mean egg numbers for each

age group were compared using the Dunn’s Multiple Comparisons Test. No comparisons were made between

mean numbers of eggs in different age groups. Different letters above histograms within the same age group

indicate a significant difference (df =14; P < 0.05).

Thus, in general, those fungal species that were not preferred for oviposition under two- choice conditions had very few eggs laid on them under no-choice conditions by females of all age groups. For example, females laid almost no eggs on A. niger under no-choice conditions

58 across all age groups and they had almost completely avoided laying eggs on A. niger in the two- choice oviposition assays. There were no statistically significant differences in egg numbers between blank PDA and A. niger across all age groups (Fig 3-5 a-h). Even at the end of their lives, at seven-days-old (Fig 3-5 h), females chose to lay their eggs on the plastic sides of Petri dishes before they died, rather than laying them on A. niger. A similar result was seen in no-choice oviposition tests for M. thermophilium. There were no significant differences in the very low egg numbers deposited by females aged 1-7 days on M. thermophilium compared to either blank PDA or A. niger (Fig 3-5 h).

There were two curious anomalies in no-choice oviposition results compared to the two- choice assays. In the two-choice assays, both T. aggressivum and P. citrinum had received similarly large numbers of eggs. Under no-choice conditions P. citrinum received numbers of eggs similar to the blank PDA controls and to T. aggressivum across all age groups (Fig 3-5 a-h).

However, T. aggressivum only received similar numbers of eggs to the blank PDA control from newly emerged, one-day-old, and two-day-old females (Fig 3-5 a-c). Similar results were seen for

Chaetomium sp., the species of fungus found in high abundance in mushroom compost. Although not preferred for oviposition compared to T. aggressivum in two-choice assays, it was readily accepted as an oviposition site by females under no choice conditions. Our blank PDA controls received very few eggs from female flies across all age groups.

No larvae survived past the first instar on pure cultures of A. flavus, A. niger, A. fumigatus, or T. aggressivum (Fig 3-6). On P. citrinum 60% of the first instars survived to adulthood, 73% survived to adulthood on Chaetomium sp., and 70% survived to adulthood on S. thermophilum

(Fig 3-6). No larvae survived past the first instar on our PDA controls.

Figure 3-6: Mean survival (± SEM) of Lycoriella ingenua larvae to adults on various pure fungal cultures grown on potato dextrose agar. Each histogram is the mean survival of 5 newly emerged first instar larvae, replicated from 6

different cohorts of larvae. All data were non-normally distributed and differences among mean survival on

different fungal cultures were first analyzed via the Kruskal-Wallis test. Differences between survival on different fungal cultures were compared with the Dunn’s Multiple Comparisons Test. Different letters above bars indicate a

significant difference (df =5; P < 0.05).

Discussion

There is a growing body of experimental evidence showing that many species of insects are attracted to microbial volatile organic compounds produced by microbes present in their habitat

(Andreadis et al., 2015b; Davis & Landolt, 2013a; Davis et al., 2013; de Bruyne & Baker, 2008;

Hung et al., 2015). Sometimes these volatiles act additionally as oviposition cues that may provide the attracted female insects with information about the suitability of substrates for larval development (Becher et al., 2012).

The results from our study add to this evidence, and show that the growing mycelia of some species of fungi found in mushroom compost are attractive to gravid female L. ingenua and induce ovipostion. However, mycelia of other fungal species were not attractive and did not induce much oviposition, even in no-choice conditions. The goal of this research was to try to identify species that might elicit higher levels of attraction and oviposition from gravid female L. ingenua than to T. aggressivum, the pestiferous green mold fungus species already shown to be attractive to these females (Cloonan et al., 2016a). Among the species we tested, we found none that were more attractive than T. aggressivum. This fungal species thus must still be considered to have the most behaviorally active set of volatiles related to L. ingenua attraction and reproduction.

The growing mycelia from only a few species of fungi were as attractive as T. aggressivum in our study. These species were M. thermophilium, from the compost, and the P. citrinum that we found on adult female bodies. It is interesting that of these two species, only P. citrinum induced levels of oviposition equivalent to levels induced by T. aggressivum in addition to being as attractive to females as T. aggressivum. On the other hand, despite its attractiveness, M. thermophilium mycelia evoked very little oviposition from L. ingenua females even at seven days old in the no-choice arena. These results indicate distinct differences between this species’ volatile cues evoking significant levels of female L. ingenua attraction and the lack of cues that would induce females to oviposit.

Several fungal species were significantly less attractive to females than T. aggressivum, including A. niger that was phoretic on males, A. flavus that we isolated from larval frass, A. fumigatus found to be phoretic on adult females, and Chaetomium sp. Again in these cases there were some differences between attraction and the propensity to oviposit, but these differences were in the opposite direction from those in the P. citrinum and M. thermophilium results discussed above. Although females were less attracted to A. flavus compared to T. aggressivum, they were stimulated to oviposit more eggs on A. flavus than on T. aggressivum in two-choice oviposition tests.

These results might indicate that the volatiles from a suite of different fungal species act in concert in the natural setting of mushroom compost. Volatiles from some species such as M. thermophilium might cause gravid females to be attracted to the compost, and then volatiles from others, such as from A. fumigatus or A. flavus, might induce oviposition once females arrived at this area of compost, even though the volatiles from such species are not themselves very attractive to females. Penicillium citrinum seems to be similar in behavioral activity to T. aggressivum in being attractive to females as well as inducing high levels of ovipostion. It would be interesting to find out whether T. aggressivum is truly vectored by L. ingenua adults as has been conjectured. Then we might be able to see whether there is a relationship between fungal species known to be phoretic on L. ingenua adults — such as P. citrinum and possibly T. aggressivum — and high levels of both L. ingenua attraction and oviposition.

There is evidence of strong relationships in other dipteran-plant-host systems between microbial infection and reproductive behavior of adult females. Flies in the genus Bradysia

(Diptera: Sciaridae), the darkwing fungus gnats, can be severe pests of ornamental and vegetable

62 greenhouse production systems (Cloyd, 2008; Cloyd et al., 2009). Female Bradysia impatiens are more attracted to geranium seedlings infected with several plant-pathogenic Pythium spp. compared to healthy seedlings (Braun et al., 2012). In the case of the onion maggot fly, Delia antiqua (Diptera: Anthomyiidae), adult females are more attracted to diseased compared to healthy onion plants and their larvae develop better on these damaged and diseased plants than on healthy ones (Hausmann & Miller, 1989). Furthermore, insects from other orders vector specific fungal and other microbial symbionts to host substrates that assist the insect with nutrition acquisition and subsequently enhance its survival (Davis et al., 2011; Nasir & Noda,

2003; Witzgall et al., 2012).

Field evaluations of dipteran adults attracted to traps baited with Penicillium expansum showed high catches of flies in the families Chironomidae, Drosophilidae, Sarcophagidae, and

Syrphidae (Davis & Landolt, 2013b). However none of the flies captured in these traps were from the fungus gnat families, e.g., Sciaridae, Mycetophilidae, or Phoridae. Some studies have shown that there may be microbial semiochemicals that attract larval dipterans. Drosophila melanogaster larvae have been shown to be attracted toward pure cultures of P. expansum under laboratory conditions (Stotefeld et al., 2015).

A previous study found that L. ingenua will oviposit on Aspergillus versicolor, but that the hatched larvae do not survive past the first instar (Frouz & Novakova, 2001). Similar results for

Aspergillus species were seen in our choice and no-choice experiments for A. flavus and A. fumigatus, in which females would choose to lay eggs on the fungi but subsequent larval survival was zero. Larval survival on P. citrinum was fairly high (60%), corroborating previous results that

L. ingenua can complete its life cycle on some Penicillium species (Frouz & Novakova, 2001). It is

63 interesting that both P. citrinum and S. thermophilum were equally as attractive to L. ingenua females as T. aggressivum, but that larvae could only complete their life cycle to adulthood on P. citrinum and S. thermophilum, not on T. aggressivum.

One explanation for L. ingenua attraction to the isolated fungal species, other than direct acquisition of nutrients, may be that the fungus provides some type of detoxification benefit to feeding larvae. Previous research suggests that fungal species in the genus Attamyces, cultivated by some species of ants, and Symbiotaphrina vectored by some cigarette beetle species, may provide some detoxifying qualities to their host insect (Dowd, 1992). The attractive fungal species in our study may also provide increased substrate utilization for L. ingenua larvae. It has been suggested that the Asian Longhorn beetle, Anoplophora glabripennis, contains a soft-rot fungal gut symbiont in the species complex Fusarium solani/Nectria haematococca that aids in the breakdown and metabolism of lignin (Geib et al., 2008). Future work should examine whether any of the isolated species of fungi contain genes encoding for xylanases, proteinases, or cellulases, similar to such enzymes previously described for fungi associated with leaf cutting ants

(Schiøtt et al., 2008). Enzymatic activity may provide information for the possible benefit to attracted and ovipositing female L. ingenua flies.

Several different fly species have been shown to carry A. flavus, A. fumigatus, A. niger, and Penicillium sp. spores on their bodies (Kumara et al., 2013; Sakai et al., 2007; Srivoramas et al., 2012; Trienens et al., 2010; Yang et al., 2015). L. ingenua flies may be carrying spores of these fungi into mushroom houses from outside and act as a source of external inoculum. They may also act as vectors that spread the spores of these fungi between different mushroom growing rooms in a mushroom house. More efforts should be made to investigate a possible link between

L. ingenua flies and these human pathogenic and allergenic fungi in mushroom production systems (Bogart et al.; Fakruddin et al., 2015; Kwon-Chung & Sugui, 2013; Lai et al., 2002;

Llewellyn et al., 1983; Ramirez-Camejo et al., 2014; Su et al., 1999).

This research offers insight into potential sources of fungi in mushroom compost that elicit attraction and oviposition to the damaging sciarid fungus gnat, L. ingenua. In the future we hope to identify attractive head-space volatiles for the species of fungi we isolated, focusing specifically on those that were most attractive to female flies, i.e., T. aggressivum, P. citrinum isolated from adult female flies, and M. thermophilium found in compost. By isolating and identifying any volatiles emitted by these attractive species of fungi, it may be possible to develop some kind of synthetic lure that could be used in a mushroom growing house for monitoring and control of host-seeking female flies. Lycoriella ingenua are known to be attracted to mushroom compost the moment a mushroom house is filled with spawned compost, and the number of invading female flies decreases after 14 days (Mehelis, 1995). Thus, during the period in which a mushroom crop develops in the compost, L. ingenua survival significantly decreases (Binns,

1980a; b; Hussey & Gurney, 1968). If we can prevent this initial invading population within the first 14 days via some kind of attract and kill strategy, all other invading populations will not likely establish in the compost and cause damage to the mushroom crop.

Acknowledgements

We thank Dr. David Byer and Dr. John Pecchia for their contributions to this work. We also thank

Vija Wilkenson for supplying us with the initial fungal cultures isolated from mushroom compost.

Chapter 4

Isolation of a female-emitted sex pheromone component

of the fungus gnat, Lycoriella ingenua, attractive to males

*Published December 2015: Andreadis SS, Cloonan KR, Myrick AJ, Chen HB, Baker TC. Isolation of a female-emitted sex pheromone component of the fungus gnat, Lycoriella ingenua, attractive to males. Journal of chemical ecology. 2015;41(12)

Abstract

Lycoriella ingenua Dufour (Diptera: Sciaridae) is acknowledged as the major pest species of the white button mushroom, Agaricus bisporus, throughout the world. Previous reports concerning the identification of a sex pheromone comprised of saturated n-C15-n-C18 hydrocarbons useful for trapping L. mali, with the major component declared to be heptadecane, have proven to be questionable. The purpose of our present study was therefore to reinvestigate the sex pheromone of this species, beginning with the collection of extract from virgin females, and thereafter isolating the behaviorally active fractions of these extracts that would evoke wing- fanning plus copulatory abdomen-curling in males. Our investigations using coupled gas chromatography electro-antennographic detection, plus a rarely used gas chromatography- coupled behavioral bioassay, resulted in a behaviorally active pheromone component being isolated and partially characterized via gas chromatography-mass spectrometry. This component was found definitively to not be n-heptadecane or any of the other n-C15-n-C19 saturated

66 hydrocarbons previously erroneously identified, but rather appears to be a sesquiterpene alcohol having chemical characteristics that we found to be quite closely matched to those of some form of germacradienol.

Introduction

Lycoriella ingenua (Dufour) (Diptera: Sciaridae) (formerly known as L. mali Fitch) is acknowledged as the major pest species of commercial mushrooms throughout the world causing severe damage (Erler et al., 2011; Park et al., 2008). Larvae of the flies feed on the compost, mycelium and sporophores, and tunnel into the caps and stems of mushrooms (Shamshad, 2010; Shamshad et al., 2008). Moreover, adults of L. ingenua vector fungus spores of Trichoderma aggressivum

(Samuels & W. Gams) (Hypocreales: Hypocreaceae), which cause severe epidemics of “green mold” and lead consequently to additional crop losses (Shamshad, 2010).Thus, control of L. ingenua is a necessity worldwide, and so far control efforts primarily rely on applications of conventional synthetic pesticides (Cantelo, 1979a; 1983b; Shamshad, 2010; Shamshad et al.,

2008). However, insecticide options are limited because they are subject to label restrictions of the number of applications per season or the total amount of active ingredient applied.

Moreover, the efficacy of insecticides is inconsistent, because larvae move away from the hatching site to feed inside the caps and stems of mushrooms, where they are well protected. In addition, repeated applications produce undesirable effects, such as insecticide residues and reduced populations of natural enemies as well as insecticide resistance (Bartlett & Keil, 1997;

Brewer & Keil, 1989b). It is therefore essential to develop efficient monitoring methods, including damage thresholds and alternative control strategies such as the use of semiochemicals such as

67 sex pheromones.

Sex pheromones involved in intraspecific communication are widely used for mating disruption and attraction to point-source lures, for population monitoring, and for control by mass trapping and attract-and-kill (Carde & Minks, 1995; Witzgall et al., 2010). In the dipteran suborder Nematocera, sex pheromones are used by flies in only three families thus far: the

Cecidomyiidae, the Psychodidae and the Sciaridae (Wicker-Thomas, 2007). The sex pheromones of nematoceran flies are female-emitted, and cause attraction of males over long distances. In cecidomyiid species such as the Hessian fly, Mayetiola destructor (Say) (Diptera: Cecidomyidae)

(Foster et al., 1991) and the swede midge, Contarinia nasturtii (Kieffer) (Diptera: Cecidomyidae)

(Hillbur et al., 2005), the components are derived from short-chain alkanes having acetoxy groups. Alternatively the sex pheromone components of sand flies (Psychodidae) are sesquiterpenes, as in Lutzomyia longipalpis (Lutz & Neiva) (Brazil et al., 2009; Hamilton et al.,

1996)) and L. cruzi (Mangabeira) (Brazil & Hamilton, 2002).

In sciarids only one pheromone has been identified, that of L. ingenua (Kostelc et al.,

1980), although sex pheromones have been implicated in the mate-finding behavior of several species of Bradysia (Alberts et al., 1981; Frank & Dettner, 2008; Li et al., 2007). The previous identification of the sex pheromone of L. ingenua (called L. mali Fitch at that time) by Kostelc et al. (1980) has been shown in more recent studies to be questionable (Gotoh et al., 1999). The purported major pheromone component n-heptadecane (C17) plus companion saturated n- hydrocarbons n-C15, C16, C18, C20 were stated to be the active constituents in the blend (Kostelc et al., 1980). However, Gotoh et al. (1999) found that n-heptadecane was completely pheromonally inactive; virgin male flies did not make any copulatory responses to doses ranging from 10-4 to

10-11g of n-heptadecane, and this pheromone blend has never been shown to be effective for any kind of monitoring of L. ingenua populations in mushroom houses anywhere in the world.

In our current study, we report the collection of extract from unmated L. ingenua females and the isolation and identification of behaviorally active fractions of these extracts for inducing courtship behavior in males. We used coupled gas chromatography-electroantennographic detection (GC/EAD) and gas chromatography coupled with a behavioral bioassay (GC/BB) to pinpoint the elution times on two different GC columns of one pheromononally active compound. Long-distance upwind flights of males in an olfactometer bioassay in response to unmated, mated, or dead unmated females were also characterized.

Materials and Methods

Insect Rearing

The insects used in this study were from a 2-year-old laboratory colony maintained at the

University Park Campus of Penn State University. This colony was initiated in 2012 using gravid adult female flies that had been aspirated from the beds of spawned A. bisprous compost in Berks

County (PA, USA). Flies were positively identified as L. ingenua by Dr. Seunggwan Shin (North

Carolina State University, Department of Entomology, Raleigh, NC).

We found that for rearing L. ingenua, we could maintain a robust and healthy colony by using unspawned compost, e.g., compost lacking A. bisporous mycelia (unpubl. data). Nine, 355- ml plastic drinking cups were filled with Phase II mushroom compost with an added nitrogen

69 supplement (100:1, compost:supplement) which were placed into mesh cages (BioQuip, CA,

USA; 30 x 30 x 30 cm). Approximately 50 male and 50 female L. ingenua flies were added to each cage, and the cages were then left under colony conditions of 21°C, 70% r.h., 12:12 L:D photoperiod regime for 2 days to allow the females to oviposit in the compost mixture in the cups. After 2 days, the cages were covered with plastic autoclave bags (VWR International,

Atlanta, GA, USA; 30.5 x 60 cm) to prevent the compost from drying out. The cages were left under colony conditions until future adult flies emerged approximately 21 days later. This process was repeated until we had a continuously emerging colony.

In order to obtain virgin adults of both sexes, we collected pupae, which were difficult to segregate according to sex, sorting them from compost by means of a camel’s-hair brush, and placed them into individual 10 ml disposable culture tubes (15 x 85 mm) (VWR International, PA,

USA) covered with Parafilm M (Bemis Healthcare Packaging, WI, USA). Individual virgin males and females could then be identified after they emerged as adults in these culture tubes and these were then used for obtaining virgin female pheromone extract and for male GC/EAD and GC/BB experiments.

Olfactometer Choice Assays

A Y-tube olfactometer (70 x 35 x 6 cm) was used for the behavioral assays under laboratory conditions (23°C and daylight intensity) (Figure 4-1). Male L. ingenua were released and assayed one-at-a-time, beginning with their introduction at the downwind end of the main trunk of the

Y-tube. Charcoal-purified air was pushed through both arms of the Y-tube at a rate of 0.75 liters/sec via a portable air-pump system (PVAS22, Volatile Assay Systems, NY, USA). Starved,

70 virgin female L. ingenua, 1-2-days old, were used in the study. Three females were placed in a glass cylinder open at both ends (1.5 cm in diameter x 5.0 cm in height), the ends being covered with fine cloth mesh. The vial containing the females was then positioned inside either of the arms of the Y-tube at their extreme upwind end. Choice assays were performed using as treatments, live virgin females, live mated females, or dead virgin females. Dead females were used because males had been observed occasionally in the lab being attracted to and clasping dead females during abdomen-curling courtship. We thus wanted to quantify this behavioral activity compared to live females to help us learn whether we could obtain sufficient amounts of pheromone from dead virgin females as well as from live females. A blank (empty glass vial) was used as a control in some assays in addition to direct comparisons between live virgin females and dead ones, as well as live virgin females vs. live mated females. All dead virgin females that were used in these experiments had died within the preceding 24 hours before assay. In total six replicates of five individuals were performed for each combination.

Figure 4-1: Schematic drawing of the Y-tube olfactometer used in these studies. Humidified air was blown through

the two arms of the Y-tube (left) and over glass vials that were covered with fine mesh at both ends. The vials

contained either three live virgin females, three live mated females, three dead virgin females, or were empty,

depending on the experiment. Virgin male flies were released one-at-a-time at the far downwind end (right) and their progress upwind into either arm was monitored. (b) Mean percentages of male L. ingenua (± S.E.) that were

attracted in the Y-tube olfactometer choice assays to (i) live virgin female flies versus a blank control, (ii) dead virgin female flies versus a blank control, (iii) live mated female flies versus a blank control and iv) live virgin female flies versus dead virgin female flies. Asterisks denote significantly different levels of attraction (χ-square 2 x 2 test

of independence; *P<0.05; **P<0.01; ***P<0.001; N = 30).

Males were assayed individually for their ability to perform upwind flight and landing on or near the glass vial containing females. Typically, the males flew all the way up the tube with very little touching or landing on the glass walls before they landed on, or within a few cm downwind of, the females while wing fanning while walking during the last few cm of their approach. Males had to progress all the way up the tube with upwind flight to a position level with the females’ vial in the tube in order to receive a positive score. Thus, we describe the percentage of males exhibiting successful attraction all the way to the source as performing successful “upwind flight”.

Each male was given three minutes to respond and males that did not initiate upwind flight within the three-minute assay period (fewer than three individuals per treatment over the entire course of the experiment) were discarded and not used for the statistical analyses. No males went only part way up the tube; they either did not respond at all or made it all the way up to the females.

Responders were removed from their location near the females using an aspirator and were not re-used. The orientation of the arms of the Y-tube was reversed 180ο after 5 individual flights to avoid lighting-related bias.

Extraction of Sex Pheromone Component

For the isolation of the sex pheromone component from the females of L. ingenua we used hexane as an extraction solvent. Virgin female flies 0-to-1-day- post-emergence were collected from their individual emergence vials as described above and then transferred into 2 ml clear glass vial (12 x 32 mm) (Supelco, PA, USA) to form groups of five to ten females in a single vial.

These females were held for 24 to 48 h in an environmental chamber maintained at 21±1°C, 70%

RH, on a 12:12 (light:dark) photoperiod regime. Afterwards, both the female bodies as well as the glass vial that had contained them were washed with 50 μl distilled hexane (n-hexane,

OmniSolv, Germany) for approximately 5 min. The supernatant was transfered into a new clear micro-test-tube that was placed inside a 1-dram vial with a Teflon-lined cap and stored at -20°C.

For analyses on the GC, the collected extract was concentrated under a stream of nitrogen to obtain a concentration of approximately 1 female equivalent (FE) per microliter (1 FE/µl).

GC Analyses

Gas chromatographic analyses were performed in splitless mode on a Hewlett-Packard 5890

Series II gas chromatograph outfitted with ChemStation software (version D.01.00 Build 75,

Agilent Technologies, Wilmington, DE, USA). The GC analyses were performed using either a relatively non-polar EC-5 column (30 m x 0.320 mm; 1.0 μm) (Agilent Technologies) or a polar DB-

225 column (30 m x 32 mm; 0.25 μm) (Agilent Technologies). In all analyses, from one to two µl of extract was injected to attain ca. 1 ng of pheromone component per injection for each GC run, because the concentration of pheromone component in the 1FE/µl extract varied slightly between different cohorts of females that had been used to obtain extract on different days.

With the EC-5 column, the carrier gas (He) linear velocity was 45 cm/sec before being equally split via a fused silica “Y GlasSeal” connector (Supelco, Bellefonte, PA, USA) 1 m before the FID and EAD sensor outlets. The injector temperature was set at 250°C, with a purge 1 min after injection. The initial column temperature was 100°C and upon injection was then immediately ramped up to 220°C at a rate of 10°C/min. Final temperature was held for 25 min. With the DB-

225 column we used a carrier gas linear velocity of 45 cm/sec and an injector temperature of

250°C with a purge 1 min after injection. The initial column temperature was 80°C and upon injection was then immediately ramped up to 200°C at a rate of 15°C/min. Final temperature was held for 25 min.

All GC/EAD-active compounds, as well as GC/BB-active compounds had their Kovats retention indices calculated. We injected a hydrocarbon “ladder” consisting of 20 ng each of a series of n-C7-to-n-C30 saturated hydrocarbons using the same GC flow and temperature conditions used on the EC-5 and DB-225 columns, respectively, for the GC/EAD and GC/BB analyses. We injected also the single hydrocarbon, n-eicosane (n-C20) to identify the elution times of each of the hydrocarbons in the series in order to accurately place the position of our active compounds with respect to the compounds in the hydrocarbon ladder. The Kovats retention indices of our GC/EAD-active pheromone compound on both columns were calculated using the formula described by Robards et al. (1994).

Mass Spectrometry Analyses

Coupled gas chromatography-mass spectrometry (GC/MS) of EAD-active compounds was performed in electron impact (EI) mode using a Hewlett-Packard 6890 gas chromatograph equipped with an HP-5MS bonded phase capillary column (0.25 mm x 0.25 µm x 30m; Agilent

Technologies, Little Falls, DE, USA) interfaced to a Hewlett-Packard 5973 mass spectrometer

(Hewlett-Packard, Palo Alto, CA, USA). The column temperature was programmed from an initial temperature of 100°C and increased by 10°C min-1 to 220°C with a 5 min hold time. An injection of 20 FE of virgin female extract that had been concentrated to 2µL of hexane was made with the inlet in splitless mode at 250°C with a split time of 0.75 min and helium carrier gas flow rate of

0.7mL/min. EI analysis used the default settings (ion source: 230°C, quadrupole: 150°C, and with spectra generated at 70eV). Tentative identification was performed using the NIST 08 library as well as published retention indices and spectra of sesquiterpene alchols, including germacradienols.

GC/EAD Assays

In order to discriminately and sensitively record the responses of male L. ingenua antennae to whole-female extract eluting from the GC, we used a portable, Quadroprobe EAG recording system (Syntech®, The Netherlands) (Park & Baker, 2002) for simultaneous EAG recordings from eight male antennae placed in parallel (Figure 4-2). The stainless-steel probes were comprised of a single, common reference electrode and four independent recording electrode channels.

However, only one of the four channels was used, whereby we placed four males between the reference and the recording electrodes without cutting the antennal tips (Figure 4-2). A salt-free hypoallergenic electrically conductive gel (Spectra 360, Parker Laboratories Inc., USA) was used to establish electrical contact between the flies’ antennae, abdomens, and the electrodes (Figure

4-2). The gel was essential in restraining the positions of the males’ bodies and antennae with high stability between the recording and reference electrodes. The 8-antenna EAG preparation was positioned on the inside of, and in the middle of, a stainless steel air tube (0.8 cm inner diameter x 14 cm long) that carried a humidified air stream (charcoal-filtered house-air) flowing at a constant rate of ca. 20 cm/s over the antennae. This airstream carried the GC effluent emanating from the GC column via the heated EAG transfer line from the GC to then flow over the antennal preparation. The EAG signals from the Quadroprobe were amplified and monitored

76 with a four-channel portable preamplifier unit (Syntech B, The Netherlands) and digitized and stored on a PC using custom, in-house software. The stored signals were further analyzed with custom, in-house PC-based data processing software designed for four-channel EAG recording analysis.

Figure 4-2: Close-up image of the setup used for GC/EAD recordings. Four L. ingenua males’ abdomens formed an electrical connection with one of the stainless steel electrodes on the Quadroprobe using EKG gel, and their eight

antennae contacted the second, recording electrode such that the eight antennae were placed in parallel. Using multiple antennae in parallel significantly improved the signal-to-noise ratio and was a key to obtaining repeatable

and reliable GC/EAGs with this species.

GC/BB Experiments

The GC/BB technique was used for behavioral assays of compounds continuously eluting from the GC, as previously described {Hummel, 1973 #2251;Leal, 1994 #2255. GC/BB was carried out using the same GC EAD-FID-effluent-splitter and airstream delivery tube that usually flowed over

77 an EAD preparation. However, for GC/BB, the exit of the airstream delivery tube normally over the EAD was instead made to flow into a cylindrical glass vial (1.5 cm i.d. x 5.0 cm long) open at both ends. Each such vial contained three virgin starved males aged 1-2 days old and was covered with fine mesh at both ends to contain the males. One of the meshed ends was positioned to be touching the end of the airstream delivery tube so that the GC effluent could flow over the males in the vial. For each GC/BB run, the vial containing the males was not placed in the airstream until immediately after the solvent peak had eluted. Males were used for only one GC/BB run, and then discarded. GC/BB on the EC-5 and DB-225 columns were performed using exactly the same flow rates and temperature programs as were used for the GC/EAD investigations as described above.

Statistical Analyses

For each Y-tube olfactometer experiment a χ2-test was used to test whether the distribution of the total number of male flies that chose the one arm or the other one across all replicates differed from a 50:50 distribution at α=0.05. Beforehand, a heterogeneity χ2 test was conducted to ensure that data from each replicate were homogenous. Statistical analyses were done using

SPSS Ver. 22 (IBM).

Results

Olfactometer Choice Assays

Males exhibited significant upwind flight attraction responses in the olfactometer assays (Figure

4-1 a), flying quickly upwind and landing on or near the glass vial containing the females. In most cases, males took less than 10 seconds to land on the glass vial with the females and displayed characteristic courtship behavior of wing fanning and abdomen curling.

A significantly greater number of male flies was attracted to live virgin females compared to the blank control arm (3.3 %) (x2=26.13, df=1, P<0.001, N=30) (Figure 4-1 b). In response to dead virgin females, significantly more male flies flew upwind to the dead females compared to the blank control arm (x2=14.72, df=1, P<0.001, N=30) (Figure 4-1 b). Male flies did not show any attraction to live, mated females. The percentage of males flying to the mated females was not significantly different from the percentage flying into the blank control arm (x2=0.03, df=1,

P>0.05, N=30) (Figure 4-1 b). In the comparison between live versus dead virgin females, a significantly greater percentage of males flew upwind towards and landed on live females compared to dead ones (x2=11.63, df=1, P<0.001, N=30) (Figure 4-1 b). It would appear that there is a greater amount or higher quality of pheromone emanating from live virgin females compared to that remaining in dead virgin females, and this result guided our decision to optimize collection of pheromone by using live virgin females instead of dead ones.

GC/EAD Assays of Virgin Female Extract

GC/EAD analyses of virgin female extracts of L. ingenua using the Quadroprobe electrode arrangement with 8 male antennae in parallel (Figure 4-2) revealed one EAG-active compound eluting from the EC-5 column at 6.94 min on more than 10 different GC/EAD runs using several different aliquots of extract obtained from different batches of virgin females (Figure 4-3a). The amounts extracted from different groups of females over the many months of this study never exceeded 1ng per FE. For instance, the amounts extracted from three different representative batches of 50 or more females in different months were 1.0 ng/FE, 0.6 ng/FE and 0.25 ng/FE, respectively. The Retention Index of the EAD-active component on the EC-5 column was 1659, halfway between the n-C16 and n-C17 standards (Figure 4-3 b). No detectable amounts of n- heptadecane (n-C17) were observed in our extracts, and no EAD activity was observed during any

GC runs at the elution time of n-heptadecane.

On the DB-225 column, there was a single GC/EAD-active compound that eluted consistently at 7.54 minutes (Figure 4-4 a) on more than 10 different GC/EAD runs from different batches of female extract. Other possible GC/EAD-active compounds were present but not consistently so (Figure 4-4 a). The amount of compound present in this peak from a 1FE injection was consistently ca. 1 ng or less, just as on the EC-5 column, and its Retention Index was 2065, more than halfway between the n-C20 and n-C21 hydrocarbons (Figure 4-4 b). As with the analyses on the EC-5 column, no significant amounts of n-heptadecane were detected from our extract during any GC runs, nor was any EAD activity observed at the elution time of n-heptadecane.

Figure 4-3: Five representative GC-EAD recordings (top) from male antennae in response to virgin female L.

ingenua extract injected on the EC-5 column. Bottom tracing shows a representative FID output from the extract on this column. (b) The elution time, 6.93 min, of the GC/EAD-active peak on the EC-5 column compared to elution

times of the compounds from the hydrocarbon ladder. The Kovats index for the GC/EAD active peak on this

column is calculated to be 1659.

Figure 4-4: Five representative GC-EAD recordings (top) from male antennae in response to virgin female L.

ingenua extract injected on the DB-225 column. Bottom tracing shows a representative FID output from the extract on this column. (b) The elution time, 7.54 min, of the GC/EAD-active peak on the DB-225 column compared to elution times of the compounds from the hydrocarbon ladder. The Kovats index for the GC/EAD active peak on

this column is calculated to be 2065.

GC/BB

When 1 FE of female extract was injected on the EC-5 column on ten different GC/BB replicate runs, 90% of the males (27/30) exhibited immediate wing fanning, abdomen curling, and clasper extension while trying to copulate with each other as soon as the compound at 6.93 min began eluting (Figure 4-5 a). Males were quiescent or exhibited only occasional, slow walking in the holding tube during the minutes before the 6.93-min compound eluted. The courtship and copulatory behavior then persisted for ca. 30-45 sec after the 6.93-min compound eluted (Figure

4-5 a) and males then became quiescent again, with occasional slow walking. We had expected that there would be long-lasting behavioral excitation for tens of seconds, not only because this long-duration wing fanning is typical of male insects excited by pheromone, but also because the airflow through the bioassay cylinder will have been slowed considerably by the fine-mesh gauze coverings at both ends of the cylinder that kept the males from escaping. The compound eluting at 6.93 min had the same retention time as the one that had previously evoked EAD activity on

GC/EAD runs on this same EC-5 column. It therefore has the same Kovats retention index, 1659, as the GC/EAD-active compound on the EC-5 column.

Figure 4-5: GC/BB analyses (N = 30 males) of extracts of L. ingenua sex pheromone using an EC -5 column evoked

maximum behavioral activity (wing-fanning and abdomen curling) from conspecific males at the exact same retention time as occurred for the GC/EAD-active peak (6.93 min) on this column. (b) Likewise, GC/BB analyses (N

= 30 males) of extracts of L. ingenua sex pheromone using a DB-225 column evoked maximum behavioral activity

from males at the exact same retention time as occurred for the GC/EAD-active peak (7.54 min) on this column.

On ten different GC/BB injections on the DB-225 column, the same courtship and copulatory behaviors as above were exhibited by 80% of the males (24/30), now in response to the 7.54-min compound on this column (Figure 4-5 b). These wing fanning, abdomen-curling, and clasper extension behaviors did not occur during the minutes before the 7.54-min compound eluted, and they persisted for ca. 45 sec after it eluted before returning towards baseline quiescence levels (Figure 4-5 b). This compound at 7.54 min on the DB-225 column had the same retention time as the one that had evoked strong EAD responses during the previous GC/EAD runs. It therefore has the same Kovats retention index, 2065, as the GC/EAD-active peak on DB-

225.

GC/MS Analyses

The E.I. mass spectrum of the compound eluting at the Kovats index closest to our GC/EAD-active compound on the EC-5 column is shown in Figure 6. It has a molecular ion of m/z 222, which is consistent with a sesquiterpene alcohol. The base peak of this compound is m/z 82 (Figure 4-6) with strong signals for fragments at m/z 67, 93, 107, 121, 149, and 161. This fragmentation pattern is similar to published mass spectra of a type of sesquiterpene alcohol in the group called germacradienols {Cane, 2006 #2209;Cornwell, 2001 #1962;He, 2004 #2249}. We installed our DB-

225 column on the GC/MS and the compound eluting at the Kovats retention index of our

GC/EAD-active compound exhibited a mass spectrum identical in all respects to the mass spectrum from the HP-5MS column shown in Figure 4-6.

Figure 4-6: Electron-Impact mass spectrum, from 20 FE of female extract, of the GC/EAD-active and GC/BB-active

compound on a DB-5 column.

Discussion

Using GC/EAD analyses and GC/BB assays on two different types of GC column, we have isolated a behaviorally active sex pheromone component of L. ingenua that is obviously not the C17 hydrocarbon previously stated to be the sex pheromone of this species by Kostelc et al. (1980).

The Kovats index of our behaviorally active compound on EC-5 is 1659 and thus its molecular weight lies between saturated n-C16 and saturated n-C17 hydrocarbons (Figure 4-3 b). The Kovats index of our active compound on DB-225 is 2065, which, given its elution time between n-C16 and n-C17 on EC-5, places it as a polar C16-C17 compound eluting between saturated n-C20 and saturated n-C21 on DB-225 (Figure 4-4 b).

Even based on just the Kovats indices of our behaviorally active compound, it is apparent that the Kostelc et al. (1980) pheromone identification of this species was incorrect. The lack of pheromonal activity of the n-C17 hydrocarbon in the studies of Gotoh et al. (1990) of course raised early suspicions that there was an erroneous identification by Kostelc et al. (1980).

Unfortunately, this misidentification has persisted in the literature over decades without anyone attempting to verify or reanalyze it. Contradictory results and concern about the Kostelc et al.

(1980) findings arose in the study by Gotoh et al. (1990) when the behavioral activity of these compounds for this species was shown to be nil; however, they did not try to investigate this discrepency further. The Kostelc et al. (1980) pheromone blend has otherwise been innocently referred to as the sex pheromone of this species in further publications such as the excellent review of fly pheromones by Wicker-Thomas (2007). Our results confirm that the concern over the inactivity of this pheromone by Gotoh et al. (1990) was well-justified, and indicate that the major sex pheromone component appears to be instead a germacradienol that by itself is highly behaviorally active in eliciting attraction and courtship behavior in males

The E.I. mass spectrum of the GC/EAD- and behaviorally active compound having these

Kovats retention indices on these two different types of columns has a fragmentation pattern consistent with that of a sesquiterpene alcohol, and is an especially close match to mass spectra of germacradienols (Cornwell et al., 2001; He & Cane, 2004). The molecular ion of our active compound has an m/z of 222, with a pronounced base peak of m/z 82 (Figure 4-6) with strong signals for fragments at m/z 67, 93, 107, 121, 149, and 161. Other sesquiterpene alcohols that, like the germacradienols, have a molecular weight of 222, such as β–eudesmol and its related stereoisomers (Yu et al., 2008) or zingiberenol (Borges et al., 2006; de Oliveira et al., 2013) (a sex

87 pheromone component of the rice stink bug, Oebalus poecilus) are unlike the germacradienols in that they have very weak base peaks in the m/z 80 - m/z 90 region. The mass spectrum of alpha cadinol (Cornwell et al., 2001) also lacks large fragments in this region and differs from the mass spectra of germacradienols in many other respects.

Although some of the characterized germacradienols such as β-germacrenol and 6β- hydroxygermacra-1(10), 4-diene (kunzeaol) (Cornwell et al., 2001), have strikingly similar overall mass spectral patterns to our active compound, they differ slightly in that their base peaks are m/z 81 (β-germacrenol) or m/z 84 (kunzeaol) (Cornwell et al., 2001) instead of m/z 82. He and

Cane (2004), however, isolated a germacradienol from the volatiles of Streptococcus coelicolor bacteria that exhibited the same strong 82 base peak that our compound exhibits. In addition, all the other fragments of He and Cane’s (2004) germacradienol were present in nearly identical proportions to ours, except for a small difference in the ratio of m/z 161 to 164.

It is interesting that the pheromones of some other nematoceran flies, i.e., the psychodid sandfly species, L. cruzi, and L. longipalpus, are also a germacrene-related sesquiterpene, having been identified as ±9-methylgermacrene-B (Brazil et al., 2009; Brazil & Hamilton, 2002). Further work to completely characterize the geometry and stereoisomeric identity of our compound is currently in progress.

Our Y-tube bioassays showed that the L. ingenua sex pheromone is emitted only by virgin females and attracts males from at least a short distance of 70 cm via upwind flight to the females followed by landing, wing fanning, abdomen-curling, and clasper extension by these males. The pheromone is retained on the surface of dead females for at least 24 hours at sufficient

88 concentrations to evoke attraction in males. Mated, live females do not attract males at all. The single pheromone component eluting on EC-5 and DB-225 is highly behaviorally active with regard to these courtship and copulatory behaviors, and therefore we do not have any reason to believe that any other components are needed to evoke optimal sex pheromone behavior in males. However, we are not at all ruling out the possibility that one or more minor components to this sex pheromone might exist.

The seldom-used GC/BB technique (Leal et al., 1992; Leal et al., 1994) was helpful in our study in allowing us to follow with confidence the pheromonal activity of our suspected pheromone component in our extract on two different GC columns. With GC/EAD alone, one can pinpoint potentially behaviorally important constituents in an extract such as this, but GC/BB was able to inform us that the GC/EAD-active compound we were following was in fact a pheromone component; this single compound elicited stereotypcial pheromone-type behavior in males.

The GC/BB technique couples the sensitivity and specificity of insect behavior with the analytical accuracy and reproducibility of gas chromatography. This GC-coupled behavioral assay had thus far been of use only in identifications of pheromone components in a limited number of moth and beetle species. This limited usage is due to the fact that in most cases, single pheromone components in an insect pheromone blend are not overtly behaviorally active, and so GC/EAD is always more useful for indicating which compounds in an extract have the potential to be pheromone components.

When single components do have behavioral activity, the GC/BB assay can be highly informative in verifying the behavioral (pheromonal) activity of a consitutent in an extract that

89 has shown to have GC/EAD activity, as in our current study. Indeed, GC/BB on our EC-5 column showed behavioral activity starting at the exact same retention time as the GC/EAD-active compound. Likewise, GC/BB on our DB-225 column showed behavioral activity being inititated at the exact same retention time as when the GC/EAD-active compound eluted on this same column.

Our knowledge of the pheromone chemistry of sciarid flies (Diptera: Sciaridae) is limited to just this one species, L. Ingenua (Wicker-Thomas, 2007), despite the progress we have made in our current study in characterizing this species’ sex pheromone. Evidence for sex pheromone communication is strong for sciarids in another sciarid genus, Bradysia. Studies have shown that females or female extract evoke male attraction in B. impatiens (Alberts et al., 1981) and four other species (Frank & Dettner, 2008; Li et al., 2007; 2008). Female pheromones of B. difformis,

B. optata, B. tilicola and B. odoriphaga also evoke characteristic courtship behaviors including wing fanning, abdomen curling, and males grasping females’ abdominal tips with their claspers.

The pheromones of some of these species appear to be comprised of several components, as indicated by thin layer chromatography (TLC) separation of the pheromone constituents and male bioassays of active compounds eluting on the TLC plates. Further isolation and identification of any of the pheromone components of these Bradysia species has not been reported.

If such studies were to be undertaken on other sciarid species such as those in the genus

Bradysia, perhaps the use the GC/EAD techniques using multiple antennae connected in parallel as in our study might be useful in optimizing the signal-to-noise ratio during GC/EAD runs (Park

& Baker, 2002). The GC/BB setup that we used might also be another useful technique, if, as in our species, single components have significant behavioral activity. Our current work is aimed

90 now at definitively characterizing the geometry and stereochemistry of the apparent germacradienol that we have implicated as being the major pheromone component of L. ingenua.

Acknowledgements

We thank Nate McCartney of the Jim Tumlinson laboratory here at Penn State for his help in obtaining mass spectra on the HP-5MS and DB-225 columns. We are grateful for the many helpful discussions concerning mass spectra of germacradienols that we had with Dr. Cam Oehlschlager of ChemTica Internacional, as well as several similar helpful discussions that we had with Prof.

Jocelyn Millar at the University of California, Riverside, Prof. David Cane and Dr. Wayne Zhou of

Brown University, and Prof. Dr. Stefan Schulz of the Technical University Braunschweig.

Collaboration continues with the latter two laboratories to try to pinoint the stereoisomeric structure of this germacradienol. This project was supported by USDA/AFRI/SCRI grant No. 2012-

51181-19912 entitled, “Addressing Management Gaps with Sustainable Disease and Pest Tactics for Mushroom Production”, Prof. David M. Beyer, Penn State University, Principal Investigator.

Chapter 5

Conclusions and future directions

Kevin R. Cloonan

Department of Entomology, Pennsylvania State University

CONCLUSION

The research presented in this dissertation aimed to identify substrates that might emit volatiles that are optimally attractive to host- and mate-seeking L. ingenua flies. In Chapter 1 I described the biology and ecology of sciarid flies in the genera Lycoriella and Bradysia, pests of mushroom and greenhouse crops respectively. In Chapter 2 I found that L. ingenua females were attracted to phase II mushroom compost that A. bisporus is grown on (Cloonan et al., 2016a), not the A. bisporus mycelia or mushrooms as previously thought (Kim et al., 1999). I also showed that female L. ingenua flies were attracted to pure cultures of the mycoparasitic green mold T. aggressivum and to mushroom compost infested with T. aggressivum (Cloonan et al., 2016a).

The Cloonan et al. (2016a) publication offers the first experimental data supporting a long- standing belief by mushroom growers that L. ingenua vectors T. aggressivum into and between mushroom growing houses (Kim et al., 1999; Rinker et al., 1995; Smith et al., 2006; Yi et al., 2015).

Chapter 3 expanded on the results from Chapter 2 in that I was able to isolate and identify other fungal species found in mushroom compost that were as attractive to female L. ingenua flies as T. aggressivum. Six fungal species were isolated from mushroom compost, adult fly

92 bodies, and larval frass (Cloonan et al., 2016b). In two-choice olfactometer tests I found that female L. ingenua flies were equally attracted to T. aggressivum and P. citrinum isolated from adult fly bodies, and flies were equally attracted to T. aggressivum and M. thermophilium isolated from mushroom compost. In two-choice oviposition, no-choice oviposition, and survivorship assays I found an unexpected mismatch between attraction of females to certain fungal species and their acceptance of these same species for oviposition. There was a further unexpected mismatch between oviposition on certain fungal species and the subsequent survivorship of their larval offspring on these fungi (Cloonan et al., 2016b).

In chapter 4 I reported on my work toward isolating and identifying the L. ingenua female- emitted sex pheromone (Andreadis et al., 2015a). Using both gas-chromatography coupled with electroantennography (GC-EAD) and gas-chromatography coupled with behavior (GC-BB) a single component was found that elicited significant electroantennographic (EAG) signals in male antennae, and produced robust, stereotypical sexual behavior in males. The mass spectrum of this compound indicated it to be an unusual isomer of germacradienol, a form of this sequiterpene alcohol that has never before been reported in nature. Although further chemical characterization is needed to pinpoint this germacradienol’s exact structural and stereoisomeric configuration, my research on this project helped correct a 35-year-old misidentification of this species’ pheromone that had created a behaviorally inert synthetic pheromone of no use to growers. Ultimately this new, highly active pheromone component, once completely characterized and synthesized, will be useful for monitoring populations of this pest, and possibly for mass trapping and mating disruption inside mushroom houses for novel means of L. ingenua control.

Implications for L. ingenua control in mushroom houses

The work presented in chapter 2 is the first experimental data supporting mushroom grower observations that L. ingenua adults are initially attracted to phase II unspawned compost

(O'Connor & Keil, 2005). Developing L. ingenua larvae compete with the developing A. bisporus mycelia for nutrients in the compost (Grewal & Richardson, 1993; White, 1986). This means that early stages of A. bisporus are the most vulnerable to L. ingenua damage. Developing L. ingenua larvae rapidly consume nutrients before A. bisporus establishes in the compost.

It may also be possible to delay initial fly infestations in a mushroom house. When L. ingenua eggs are introduced into spawned mushroom compost two (Hussey & Gurney, 1968) and seven (Cantelo & Antonio, 1982) days after A. bisporus establishment fly populations are significantly reduced. Populations of L. ingenua flies can therefore be significantly reduced by delaying this initial attraction to freshly spawned compost by just 2 and 7 days. Future work should examine if intense and targeted control of these initial invading populations will result in lower fly populations.

Applications of the A. bisporus metabolic by-products calcium oxalate and mannitol to freshly spawned mushroom compost may also reduce the populations of these initial infesting adults. Both of these compounds are toxic to L. ingenua flies and reduce overall fly populations

(Binns, 1980b; White, 1997). Gravid L. ingenua flies are not attracted to and will not lay eggs on pure cultures of A. niger, the species of fungi isolated from adult L. ingenua males (Cloonan et al., 2016b). This fungus also produces large amounts of mannitol (Aguilar-Osorio et al., 2010).

Oviposition repellency of A. niger may be due to the production of this toxic compound.

Future research should explore applications of both calcium oxalate and mannitol on a full-scale mushroom farm. These trials could examine the effects of calcium oxalate and mannitol applications on L. ingenua immigration rates and overall population development. Mannitol and calcium oxalate may repel or suppress initial infesting L. ingenua populations. This may allow A. bisporus to establish in the compost enough to suppress future populations of L. ingenua. As metabolic by-products, however, mannitol and calcium oxalate may hinder the early development of A. bisporus.

Possible mutualistic associations between L. ingenua and various fungal species

Many insect-fungi mutualisms exist, such as in ant (Abril & Bucher, 2002; 2004), beetle (De Fine

Licht & Biedermann, 2012; Endoh et al., 2011), and termite (Aanen et al., 2002) systems. Insects provide fungi with increased spore dispersal and fungi provide insects with fitness benefits including increased nutrient availability and substrate detoxification. Female insects obtain information about host suitability from semiochemicals that may affect the fitness of their offspring called the preference-performance, or “mother knows best” hypothesis. It predicts that there is a tight evolutionary linkage between female host choice and the nutritional needs of their offspring (Jaenike, 1978).

Unpublished data by Penn State graduate student Maria Mazin shows that larvae reared on spawned mushroom compost infested with T. aggressivum survive better than on un-infested spawned mushroom compost. Thus, female L. ingenua flies transport T. aggressivum spores, are preferentially attracted to mushroom compost infested with T. aggressivum, and survive better

95 on mushroom composted infested with T. aggressivum. This possible L. ingenua-T. aggressivum insect-fungi mutualism may be an example in support of the preference-performance hypothesis.

Future work should now investigate the dispersal of T. aggressivum spores into and around a mushroom house. Questions that interest me include: are infestations of T. aggressivum originating from flies coming from the outside of a mushroom house? If so, where on the landscape are L. ingenua adults acquiring these spores? Can L. ingenua flies carry spores between mushroom growing houses, or can flies only carry spores a short distance within the same growing room?

Using quantitative real-time PCR researchers were able to detect Fusarium oxysporum f. sp. cucumerinum on Bradysia sp. adult flies entering a cucumber greenhouse (Scarlett et al.,

2014). Similar methods could be used across several mushroom houses analyzing incoming populations of L. ingenua for traces of T. aggressivum and other A. bisporus diseases. These data could broaden mushroom growers understanding of the disease dynamics within a mushroom growing house.

The data presented in chapter 3 begins to reveal the complex microbial environment of mushroom compost (Cloonan et al., 2016b) that may attract initial populations of gravid L. ingenua flies. A mutualist insect-fungi relationship may exist between the fungal species isolated from adult fly bodies, P. citrinum, and L. ingenua flies. This species of fungus has been found on other agricultural commodities including coffee (Batista et al., 2003) and some citrus fruits (du

Plooy et al., 2009). I am curious if this fungus will grow on mushroom compost. If P. citrinum can

96 go on mushroom compost I would like to investigate if L. ingenua females are more attracted to compost infested with P. citrinum versus un-infested mushroom compost.

Oviposition and larval survival experiments in chapter 3 showed a disparity between fungal species that elicited oviposition and those that sustained larvae to adulthood. Similar results were seen with L. ingenua by previous authors on several fungal species (Frouz &

Novakova, 2001). Pure fungal cultures of some species elicited a strong oviposition preference but larvae reared on these species did not survive past the first larval instar, as in the case of A. flavus, A. fumigatus, and T. aggressivum (Cloonan et al., 2016b). Conversely other fungal species did not elicit oviposition, but when placed directly onto pure cultures first instar larvae survived to adulthood, as in the case of M. thermophilium and Chaetomium sp. (Cloonan et al., 2016b). I think that attempts should be made to grow A. flavus and A. fumigatus on mushroom compost.

If they can successfully colonize mushroom compost I would attempt to rear L. ingenua larvae on this infested compost. It may be that L. ingenua larvae are unable to complete their development on pure cultures of A. flavus, A. fumigatus or spawned compost. However, mushroom compost infested with these fungi may sustain the larvae to adulthood as in the case of T. aggressivum.

Previous authors showed that phase II mushroom compost without M. thermophilium had a significantly lower A. bisporus yield (Coello-Castillo et al., 2009). Female L. ingenua flies are attracted to M. thermophilium, lay eggs on pure cultures of M. thermophilium, and can fully develop from larvae to adults on M. thermophilium (Cloonan et al., 2016b). Larvae are likely directly consuming the M. thermophilium in the compost. Developing L. ingenua larvae and developing A. bisporus mycelia may therefore be competing for M. thermophilium as a source of

97 nutrition from the compost. Hence, if larvae are directly feeding on this fungus in the compost this may be an undescribed damage characteristic of L. ingenua as a pest of A. bisporus.

Early ecology studies showed that samples of rabbit dung on the forest floor in the U.K. infested with Lycoriella sp. larvae contained significantly less Podospora tetraspora and

Lasiobolus intermedius mycelia compared to those droppings that were not infested with

Lycoriella sp. (Wicklow & Yocom, 1982). This is the first and only reference of these coprophilous fungi in association with sciarid larvae. In the future I would like to obtain samples of both of these fungi and perform similar two-choice olfactometer tests from chapters 2 (Cloonan et al.,

2016a) and 3 (Cloonan et al., 2016b). These data would build upon work (Cloonan et al., 2016b;

Frouz et al., 2002; Wicklow & Yocom, 1982) furthering our understanding of the host range of L. ingenua in a continued attempt to find a substrate that is more attractive to L. ingenua females than T. aggressivum.

Difficulties in proving mutualism

There are several examples in which fungi seem to “deceive” insects. These fungi attract insects to spore-rich areas through olfactory and gustatory cues. Spores then attach to the insect body and are subsequently dispersed as the insect moves across the landscape. Attracted insects do not receive a nectar reward or future fitness benefit through this spore dispersal process. In these cases a mutually beneficial relationship does not exist. By defining the parameters that make an insect-fungal interaction mutually beneficial, I hope to gain further insight into the relationship between sciarid flies and the fungal pathogens they vector to mushroom and plant hosts.

Honeydew excreted from sorghum infected with the ergot fungus, Claviceps purpurea, is attractive to several fly, beetle, and wasp species (Prom et al., 2003). In another example the small wasp, Polistes fuscatus, is attracted to dallis grass, Paspalum dilatatum, infected with

Claviceps. sp., the causal agent of ergot disease. Foraging wasps preferentially visit the honeydew of those infected grasses to feed on their honeydew excretions. After foraging and feeding on infected P. dilatatum, wasps land on uninfected grasses and perform grooming behaviors (Hardy,

1988). The authors suggest that it is during these grooming events that the wasps vector

Claviceps sp. spores to uninfected P. dilatatum grass heads. Similar observations were seen with a mycophagous fly, Minettia lupulina (Diptera: Lauxaniidae) and a mycophagous beetle,

Acylomus sp. (Coleoptera: Phalacridae), dispersing Claviceps purpurea spores to tall fescue,

Festuca arundinacea (Lemon, 1992).

These interactions are beneficial for the Claviceps sp. fungi and potentially beneficial for the attracted insects. In all cases the Claviceps sp. fungi acquire the benefit of spore dispersal from the insects. Attracted insects receiving the honeydew reward may not gain any advantage by visiting these infected grasses, however. Honeydew from infected grasses may have the same nutritional quality as the honeydew of un-infected grasses. In this case there would be no nutritional benefit to the attracted insects. Grasses infected with Claviceps sp. may simply increase the quantity of general attractive volatiles. This increased volatile output could then compete with non-infected grasses for insect visitation. It seems unlikely to me that this kind of an insect-fungal interaction is mutually beneficial.

In a previously described example, B. impatiens adults cannot vector Pythium sp. spores to healthy plants (Braun et al., 2010), but are attracted to geranium seedlings infected with P.

99 aphanidermatum (Braun et al., 2012). The authors did not measure B. impatiens larval survival on infected and uninfected seedlings. Onion maggot flies, Delia antiqua (Diptera: Anthomyiidae), are more attracted to onion seedlings with fungal and microbial growth versus onion seedlings that are surface-sterilized (Ellis et al., 1979), but larval survival was the same on seedlings with and without fungal growth. Similar results were seen with the seed corn maggot, Delia platura

(Diptera: Anthomyiidae), and squash seedlings infested with an unidentified fungal pathogen

(Eckenrode et al., 1975).

In no-choice oviposition assays significantly more L. ingenua females chose to lay their eggs on oyster mushrooms, Pleurotus ostreatus, that were left to rot versus fresh P. ostreatus mushrooms (Chidziya et al., 2013). The authors did not identify the causal microbial agents on these rotten P. ostreatus. Larvae manually placed on rotting and fresh P. ostreatus mushrooms had similar rates of survival to adulthood. These data show that although the rotten P. ostreatus mushrooms were more attractive to gravid females they did not offer a superior source of nutrition for the developing L. ingenua larvae.

In these cases flies are more attracted to host substrates infected with fungal pathogens but are unable to vector the spores of those fungal pathogens. Insects do not gain a fitness benefit because larvae do not survive better on infected host material. Unlike the Claviceps sp. and infected grasses example, there appears to be no fitness benefit to the fungi, through spore dispersal, or to the flies in any of these cases. It may be that B. impatiens, L. ingenua, H. antiqua, and H. platura are generally more attracted to plant material that is infected with microbial and fungal pathogens. These examples of insect-fungi interactions do not, therefore, appear to be mutually beneficial.

100

The previously mentioned B. coprophila can complete development from larva to adult on pure cultures of the causal agent of dry rot in garlic, Botrytis porri (Anas & Reeleder, 1988b).

Larvae are unable to complete their life cycle on the closely related fungus Botrytis cinerma, or on seedlings of several plant species including lettuce, Lactuca sativa, carrot, Daucus carota, leek,

Allium porrum, bean, Phaseolus vulgaris, or pea, Pisum sativum. Seedlings infected with B. cinerma are an adequate host substrate. When placed directly onto infected seedlings almost

100% of the larvae survive to adulthood (Anas & Reeleder, 1988b). Separately, the plant seedlings and B. cinerma are insufficient sources of nutrition for developing B. coprophila larvae.

Seedlings infected with B. cinerma may increase the nutrient availability for developing

B. coprophila larvae. The fungus may also detoxify compounds in the seedlings that would otherwise prove deadly to the developing larvae. Anas & Reeleder (1988b) did not examine whether B. coprophila adults are able to carry B. cinerma spores. They also did not examine if B. coprophila adult females were more attracted to plant seedlings infected with B. cinerma, although the closely related B. impatiens is more attracted to geranium seedlings infected with several Pythium species (Braun et al., 2012).

With further experimental data this interaction could prove to be an example of an insect- fungi beneficial mutualism. First, B. coprophila female flies would have to be preferentially attracted to lettuce, carrot, pea, bean, and leek seedlings infected with B. cinerma. Second, adult

B. coprophila would have to be competent vectors of B. cinerma spores. The fungus receives the fitness benefit of increased spore dispersal and the fly receives the benefit of exploiting an otherwise unavailable niche.

101

A similar example was previously discussed in which B. difformis adults vector Fusarium sp. to healthy seedlings (Hurley et al., 2007), and females prefer to oviposit on plant roots infected with Fusarium culmorum (Kuhne & Heller, 2009). The missing experimental work for this potential mutually beneficial insect-fungal interaction would include analysis of B. difformis larval survival. If B. difformis larvae develop better or faster on seedlings infected with F. culmorum versus non-infected seedlings a mutually beneficial relationship may exist. Again the fungus would receive the fitness benefit of increased spore dispersal and the fly would receive the fitness benefit of increased larval survival.

Identification of volatiles responsible for the attraction of L. ingenua females

The main goal of the research presented in this dissertation was to identify substrates that are attractive to L. ingenua flies that might be used for controlling their populations in mushroom houses. I have demonstrated several different compost and fungal substrates that attract females and cause them to oviposit. My research has set the stage for now trying to isolate and identify compounds that comprise the attractive blends from these substrates. For attracting gravid female L. ingenua flies with synthetic blends, future work should examine the headspace volatiles of the six fungal species that I isolated from mushroom compost, adult flies, and larval frass. Identifying attractive volatile blends can be extremely laborious, especially in polyphagous insects such as L. ingenua. For example in the case of the polyphagous fruit pest, Drosophila suzukii, ovipositing females rely on both the absolute and relative amounts of common host volatiles during the process of finding oviposition sites (Revadi et al., 2015).

102

This being said, I have already begun collecting the headspace volatiles from pure cultures of A. niger, A. flavus, A. fumigatus, P. citrinum, M. thermophilium, Chaetomium sp., and T. aggressivum using previously described methods (Hung et al., 2013). Next, I hope to analyze these volatile collections via GC-EAD and GC-MS to identify compounds that female L. ingenua antennae can detect following the protocols of similar work (Revadi et al., 2015). Lastly, I hope to test those compounds that elicited significant EAG responses for behavioral activity in female

L. ingenua flies. Single EAG-active compounds will first be tested in the static flow two-choice olfactometers from chapter 2 (Cloonan et al., 2016b). These single EAG-active compounds will always be tested for behavioral activity to assess their contribution to attraction of live females.

Large-scale and farm-wide trials should eventually be tested once compound(s) are identified that are highly attractive to gravid female L. ingenua flies.

Behavioral odor valence (positive=attraction; negative=repellency) of compounds cannot be determined by electrophysiological responses of olfactory receptor neurons on insect antennae or from the odorant receptor proteins (ORs) that are expressed on them (de Bruyne &

Baker, 2008; Knaden et al., 2012). For example GC-EAD assays using antennae from the locust bean moth, Ectomyelois ceratoniae, revealed significant antennal responses from host odors including ethyl hexanoate and 2-phenylethanol. However, only ethyl hexanoate elicited upwind flight behavior in adult E. ceratoniae (Cossé et al., 1994). Although the search for attractive volatile mixtures from natural sources may take many years, it could offer a powerful tool for mushroom growers to control populations of L. ingenua flies. However, the GC-EAD technique can reveal certain key compounds that are behaviorally important, to focus on the volatiles other

103 than the many inert ones that show up on GC tracings and can be further assessed for their contribution either to attraction or repellency.

Ecological contexts in which attraction and oviposition of L. ingenua females occur

In my view, more work needs to be done to examine what female L. ingenua flies are attracted to and laying eggs on in a more natural setting. Similarly, field surveys should investigate larval habitats in nature outside of mushroom growing houses. It may be that patches of fungi closely related to T. aggressivum, P. citrinum, and M. thermophilium across the natural landscape are even more attractive to female flies than fungi associated with mushroom compost. These basic ecology experiments could follow similar intensive sampling methods used by vector biologists to survey mosquito breeding and oviposition sites across a landscape (Ferraguti et al., 2016).

Fungi in both of the genera Chaetomium and Penicillium are producers of large amounts of geosmin (Behr et al., 2014; Mattheis & Roberts, 1992). Species of fungi in these genera were highly attractive to female L. ingenua flies (Cloonan et al., 2016b). Geosmin has been shown to repel insects (Holighaus & Rohlfs, 2016). No publications have ever reported geosmin to be attractive or neutral to insects. Future work should investigate if geosmin is attractive or repellent to adult L. ingenua flies.

It became increasingly apparent, after reviewing the available literature on sciarid flies in the genera Lycoriella and Bradysia, that ovipositing female flies are attracted to the microbial fauna of their host substrates, not the commodity they are associated pests of (i.e. fruit, vegetable, and mushroom crops). For example, in 2-choice olfactometer assays female B. impatiens adults were more attracted to older moist composted pine bark versus younger and

104 less-humid composted pine bark (Cloyd et al., 2009). Steam distillation and gas chromatography of the substrates showed that pinene and camphene were the primary components of the headspace from younger and less humid samples. In the older, more humid and attractive samples, there was significantly less pinene and camphene. Instead the primary components of these older pine compost samples were borneol and α-terpineol, both microbial metabolites of pinene and camphene (Cloyd et al., 2009).

Borneol (Roland et al., 1995), α-terpineol (Katerinopoulos et al., 2005), pinene (Erbilgin et al., 2016), and camphene (Silva et al., 2015) have all been described as insect attractants in previous studies. Future work should examine if borneol and α-terpineol elicit an attractive behavior in adult B. impatiens versus camphene and pinene. These two compounds may also be attractive to gravid female L. ingenua flies. Similar experiments with other sciarid flies and their associated host substrates may provide further evidence that Lycoriella sp. and Bradysia sp. are primarily attracted to the volatiles of microbial metabolism.

One hundred years ago Hungerford (1916) set out to better understand the biology and ecology of Sciara fungus gnats. As an applied entomologist it was Hungerford’s job to provide growers with knowledge and tools to control pestiferous populations of these flies. Although the technologies and techniques used in the experiments presented in this dissertation were unavailable to Hungerford in 1916, the goal remains the same: to provide growers with better tools for controlling populations of sciarid fungus gnats.

105

Bibliography

Aanen DK, Eggleton P, Rouland-Lefevre C, Guldberg-Froslev T, Rosendahl S & Boomsma JJ (2002) The evolution of fungus-growing termites and their mutualistic fungal symbionts. Proceedings of the National Academy of Sciences of the United States of America 99: 14887-14892.

Abril AB & Bucher EH (2002) Evidence that the fungus cultured by leaf-cutting ants does not metabolize cellulose. Ecology Letters 5.

Abril AB & Bucher EH (2004) Nutritional sources of the fungus cultured by leaf-cutting ants. Applied Soil Ecology 26.

Agriculture USDO (2015) Mushrooms Vol. 2017: National Agricultural Statistics Service (NASS).

Aguilar-Osorio G, vanKuyk PA, Seiboth B, Blom D, Solomon PS, Vinck A, Kindt F, Wösten HAB & de Vries RP (2010) Spatial and developmental differentiation of mannitol dehydrogenase and mannitol-1-phosphate dehydrogenase in Aspergillus niger. Eukaryotic Cell 9: 1398-1402.

Alberts SA, Kennedy MK & Carde RT (1981) Pheromone-mediated anemotactic flight and mating-behavior of the sciarid fly Bradysia impatiens (Diptera: Sciaridae). Environmental Entomology 10: 10-15.

Ali O, Dunne R & Brennan P (1999) Effectiveness of the predatory mite Hypoaspis miles (Acari: Mesostigmata: Hypoaspidae) in conjunction with pesticides for control of the mushroom fly Lycoriella solani (Diptera: Sciaridae). Experimental and Applied Acarology 23: 65-77.

Anas O & Reeleder RD (1988a) Consumption of sclerotia of Sclerotinia sclerotiorum by larvae of Bradysia coprophila: Influence of soil factors and interactions between larvae and Trichoderma viride. Soil Biology and Biochemistry 20: 619-624.

Anas O & Reeleder RD (1988b) Feeding-habits of larvae of Bradysia coprophila on fungi and plant-tissue. Phytoprotection 69: 73-78.

Andreadis SS, Cloonan KR, Myrick AJ, Chen HB & Baker TC (2015a) Isolation of a female-emitted sex pheromone component of the fungus gnat, Lycoriella ingenua, attractive to males. Journal of Chemical Ecology 41: 1127-1136.

106

Andreadis SS, Witzgall P & Becher PG (2015b) Survey of arthropod assemblages responding to live yeasts in an organic apple orchard. Frontiers in Ecology and Evolution 3.

Anon (1982) Mushroom Pests, Vol. No. 583: Leaflet (ed. by Ministry of Agriculture, Advisory), p. 14.

Bartlett GR & Keil CBO (1997) Identification and characterization of a permethrin resistance mechanism in populations of the fungus gnat Lycoriella mali (Fitch) (Diptera: Sciaridae). Pesticide Biochemistry and Physiology 58: 173-181. d

Batista LR, Chalfoun SM, Prado G, Schwan RF & Wheals AE (2003) Toxigenic fungi associated with processed (green) coffee beans (Coffea arabica L.). International Journal of Food Microbiology 85: 293-300.

Becher PG, Flick G, Rozpędowska E, Schmidt A, Hagman A, Lebreton S, Larsson MC, Hansson BS, Piškur J, Witzgall P & Bengtsson M (2012) Yeast, not fruit volatiles mediate Drosophila melanogaster attraction, oviposition and development. Functional Ecology 26: 822-828.

Behr M, Serchi T, Cocco E, Guignard C, Sergeant K, Renaut J & Evers D (2014) Description of the mechanisms underlying geosmin production in Penicillium expansum using proteomics. Journal of Proteomics 96: 13-28.

Binns ES (1972) Arctoseius cetratus (Acarina: Ascidae) phoretic on mushroom sciarid flies. Acarologia (Paris) 14: 350-356.

Binns ES (1980a) Biology and behaviour of sciarid fungus gnats (Dipttera: Sciaridae) in relation to swarming and migration. Entomologist's Monthly Magazine 115: 77-90.

Binns ES (1980b) Field and laboratory observations on the substrates of the mushroom fungus gnat Lycoriella auripila (Diptera: Sciaridae). Annals of Applied Biology 96: 143-152.

Bogart HGG, Ende G, Loon PCC & Griensven LJLD (1993) Mushroom worker's lung: Serologic reactions to thermophilic actinomycetes present in the air of compost tunnels. Mycopathologia 122: 21-28.

Borges M, Birkett M, Aldrich JR, Oliver JE, Chiba M, Murata Y, Laumann RA, Barrigossi JA, Pickett JA & Moraes MCB (2006) Sex attractant pheromone from the rice stalk stink bug, Tibraca limbativentris Stal. Journal of Chemical Ecology 32: 2749-2761.

Braun SE, Castrillo LA, Sanderson JP, Daughtrey ML & Wraight SP (2009) Lack of Pythium aphanidermatum transmission by adult fungus gnats (Bradysia impatiens) and investigation of larval vectoring capacity. Phytopathology 99: S16-S16.

107

Braun SE, Castrillo LA, Sanderson JP, Daughtrey ML & Wraight SP (2010) Transstadial transmission of Pythium in Bradysia impatiens and lack of adult vectoring capacity. Phytopathology 100: 1307-1314.

Braun SE, Sanderson JP, Daughtrey ML & Wraight SP (2012) Attraction and oviposition responses of the fungus gnat Bradysia impatiens to microbes and microbe-inoculated seedlings in laboratory bioassays. Entomologia Experimentalis et Applicata 145: 89-101.

Brazil RP, Caballero NN & Hamilton JGC (2009) Identification of the sex pheromone of Lutzomyia longipalpis (Lutz & Neiva, 1912) (Diptera: Psychodidae) from Asuncion, Paraguay. Parasites & Vectors 2.

Brazil RP & Hamilton JGC (2002) Isolation and identification of 9-methylgermacrene-B as the putative sex pheromone of Lutzomyia cruzi (Mangabeira, 1938) (Diptera: Psychodidae). Memorias Do Instituto Oswaldo Cruz 97: 435-436.

Brewer KK & Keil CB (1989a) A mixed function oxidase factor contributing to permethrin and dichlorvos resistance in Lycoriella mali (fitch) (diptera: Sciaridae). Pesticide Science 26: 29-39.

Brewer KK & Keil CB (1989b) Permethrin resistance in Lycoriella mali (Diptera: Sciaridae). Journal of Economic Entomology 82: 17-21.

Buczynska A, Sowiak M & Szadkowska-Stanczyk I (2008) Occupational exposure to mesophilic microorganisms associated with commercial processing of compost for mushroom production. Medycyna Pracy 59: 373-379.

Cantelo WW (1979a) Lycoriella mali (Diptera: Sciaridae) - Control in mushroom compost by incorporation of insecticides into compost. Journal of Economic Entomology 72: 703- 705.

Cantelo WW (1979b) Lycoriella mali (Diptera: Sciaridae) - control in mushroom compost by incorporation of insecticides into compost. Journal of Economic Entomology 72: 703- 705.

Cantelo WW (1983) Control of a mushroom-infesting fly (Diptera: Sciaridae) with insecticides applied to the casing layer. Journal of Economic Entomology 76: 1433-1436.

Cantelo WW (1988) Movement of Lycoriella mali (Diptera: Sciaridae) through mushroom- growing medium. Journal of Economic Entomology 81: 195-200.

108

Cantelo WW & Antonio JPS (1982) Effect of mushroom mycelium growth on population development of Lycoriella mali, (Diptera: Sciaridae) nematodes, and mites in compost. Environ Entomol 11: 227-230.

Cantwell GE & Cantelo WW (1984) Effectiveness of Bacillus thuringiensis var israelensis in controlling a sciarid fly, Lycoriella mali, in mushroom compost. Journal of Economic Entomology 77: 473-475.

Carde RT & Minks AK (1995) Control of moth pests by mating disruption - successes and constraints. Annual Review of Entomology 40: 559-585.

Chidziya E, Mutangadura D, Jere J & Siziba L (2013) A comparative evaluation of locally available substrates for rearing and studying biology of sciarid fly, Lycoriella mali. Academia Journal of Biotechnology 1: 057-061.

Choi WS, Park BS, Lee YH, Jang DY, Yoon HY & Lee SE (2006) Fumigant toxicities of essential oils and monoterpenes against Lycoriella mali adults. Crop Protection 25: 398-401.

Clift AD & Larsson SF (1987) Phoretic dispersal of Brennandania lambi (kcrzal) (Acari: Tarsonemida, Pygmephoridae) by mushroom flies (Diptera: Sciaridae and Phoridae) in New South Wales, Australia. Experimental & Applied Acarology 3: 11-20.

Cloonan KR, Andreadis SS & Baker TC (2016a) Attraction of female fungus gnats, Lycoriella ingenua, to mushroom-growing substrates and the green mold Trichoderma aggressivum. Entomologia Experimentalis et Applicata 159: 298-304.

Cloonan KR, Andreadis SS, Chen HB, Jenkins NE & Baker TC (2016b) Attraction, oviposition and larval survival of the fungus gnat, Lycoriella ingenua, on fungal species isolated from adults, larvae, and mushroom compost. PLoS One 11.

Cloyd RA (2008) Management of fungus gnats (Bradysia spp.) in greenhouses and nurseries. Floriculture and Ornamental Biotechnology.

Cloyd RA (2015) Ecology of fungus gnats (Bradysia spp.) in greenhouse production systems associated with disease-interactions and alternative management strategies Insects 6: 325-332.

Cloyd RA & Dickinson A (2006) Effect of Bacillus thuringiensis subsp israelensis and neonicotinoid insecticides on the fungus gnat Bradysia sp nr. coprophila (Lintner) (Diptera: Sciaridae). Pest Management Science 62: 171-177.

Cloyd RA, Marley KA, Larson RA & Arieli B (2009) Attractiveness of parboiled rice hulls to the fungus gnat, Bradysia sp. nr coprophila (Diptera: Sciaridae), adult relative to standard growing medium components. Hortscience 44: 1366-1369.

109

Cloyd RA & Zaborski ER (2004) Fungus gnats, Bradysia spp. (Diptera: Sciaridae), and other arthropods in commercial bagged soilless growing media and rooted plant plugs. Journal of Economic Entomology 97: 503-510.

Coello-Castillo MM, Sánchez JE & Royse DJ (2009) Production of Agaricus bisporus on substrates pre-colonized by Scytalidium thermophilum and supplemented at casing with protein- rich supplements. Bioresource Technology 100: 4488-4492.

Cornwell CP, Reddy N, Leach DN & Wyllie SG (2001) Germacradienols in the essential oils of the Myrtaceae. Flavour and Fragrance Journal 16: 263-273.

Cossé AA, Endris JJ, Millar JG & Baker TC (1994) Identification of volatile compounds from fungus-infected date fruit that stimulate upwind flight in female Ectomyelois ceratoniae. Entomologia Experimentalis et Applicata 72: 233-238.

Davis T & Landolt P (2013a) A survey of insect assemblages responding to volatiles from a ubiquitous fungus in an agricultural landscape. Journal of Chemical Ecology 39: 860-868.

Davis TS, Crippen TL, Hofstetter RW & Tomberlin JK (2013) Microbial volatile emissions as insect semiochemicals. Journal of Chemical Ecology 39: 840-859.

Davis TS, Hofstetter RW, Foster JT, Foote NE & Keim P (2011) Interactions between the yeast Ogataea pini and filamentous fungi associated with the western pine beetle. Microbial Ecology 61: 626-634.

Davis TS & Landolt PJ (2013b) A survey of insect assemblages responding to volatiles from a ubiquitous fungus in an agricultural landscape. Journal of Chemical Ecology 39: 860-868.

De Bruyne M & Baker TC (2008) Odor detection in insects: Volatile codes. Journal of Chemical Ecology 34: 882-897.

De Fine Licht HH & Biedermann PHW (2012) Patterns of functional enzyme activity in fungus farming ambrosia beetles. Frontiers in Zoology 9: 1-11.

De Groot PWJ, Visser J, Van Griensven LJLD & Schaap PJ (1998) Biochemical and molecular aspects of growth and fruiting of the edible mushroom Agaricus bisporus. Mycological Research 102: 1297-1308.

De Oliveira MWM, Borges M, Andrade CKZ, Laumann RA, Barrigossi JAF & Blassioli-Moraes MC (2013) Zingiberenol, (1S,4R,1 ' S)-4-(1 ',5 '-Dimethylhex-4 '-enyl)-1-methylcyclohex-2-en- 1-ol, Identified as the Sex Pheromone Produced by Males of the Rice Stink Bug Oebalus poecilus (Heteroptera: Pentatomidae). Journal of Agricultural and Food Chemistry 61: 7777-7785.

110

de Siqueira FG, Martos ET, da Silva EG, da Silva R & Dias ES (2011) Biological efficiency of Agaricus brasiliensis cultivated in compost with nitrogen concentrations. Horticultura Brasileira 29: 157-161.

Dowd PF (1992) Insect fungal symbionts - A promising source of detoxifying enzymes. Journal of Industrial Microbiology 9. du Plooy W, Regnier T & Combrinck S (2009) Essential oil amended coatings as alternatives to synthetic fungicides in citrus postharvest management. Postharvest Biology and Technology 53: 117-122.

Dutton MV & Evans CS (1996) Oxalate production by fungi: Its role in pathogenicity and ecology in the soil environment. Canadian Journal of Microbiology 42: 881-895.

Eckenrode CJ, Harman GE & Webb DR (1975) Seed-borne microorganisms stimulate seedcorn maggot egg-laying. Nature 256: 487-488.

El-Hamalawi ZA (2008) Acquisition, retention and dispersal of soilborne plant pathogenic fungi by fungus gnats and moth flies. Annals of Applied Biology 153: 195-203.

El-Hamalawi ZA & Stanghellini ME (2005) Disease development on Lisianthus following aerial transmission of Fusarium avenaceum by adult shore flies, fungus gnats, and moth flies. Plant Disease 89: 619-623.

Ellis PR, Eckenrode CJ & Harman GE (1979) Influence of onion cultivars and their microbial colonizers on resistance to onion maggot (Diptera: Anthomyiidae). Journal of Economic Entomology 72: 512-515.

Endoh R, Suzuki M, Okada G, Takeuchi Y & Futai K (2011) Fungus symbionts colonizing the galleries of the Ambrosia beetle Platypus quercivorus. Microbial Ecology 62.

Erbilgin N, Stein JD, Acciavatti RE, Gillette NE, Mori SR, Bischel K, Cale JA, Carvalho CR & Wood DL (2016) A blend of ethanol and (−)-α-pinene were highly attractive to native siricid woodwasps (Siricidae: Siricinae) infesting conifers of the Sierra Nevada and the Allegheny mountains. Journal of Chemical Ecology: 1-8.

Erler F, Polat E, Demir H, Catal M & Tuna G (2011) Control of mushroom sciarid fly Lycoriella ingenua populations with insect growth regulators applied by soil drench. Journal of Economic Entomology 104: 839-844.

Evans MR, Smith JN & Cloyd RA (1998) Fungus gnat population development in coconut coir and sphagnum peat-based substrates. Horttechnology 8: 406-409.

111

Fakruddin M, Chowdhury A, Hossain MN & Ahmed MM (2015) Characterization of aflatoxin producing Aspergillus flavus from food and feed samples. Springerplus 4.

Farsani NS, Zamani AA, Abbasi S & Kheradmand K (2013) Effect of temperature and button mushroom varieties on life history of Lycoriella auripila (Diptera: Sciaridae). Journal of Economic Entomology 106: 115-123.

Ferraguti M, Martínez-de la Puente J, Roiz D, Ruiz S, Soriguer R & Figuerola J (2016) Effects of landscape anthropization on mosquito community composition and abundance. Scientific Reports 6: 29002.

Fletcher JT, White PF & Gaze RH (1989) Mushrooms: Pests and Disease Control. Andover: Intercept Ltd.

Foster SP, Harris MO & Millar JG (1991) Identification of the sex-pheromone of the Hessian fly, Mayetiola destructor (Say). Naturwissenschaften 78: 130-131.

Frank J & Dettner K (2008) Sex pheromones in three Bradysia species (Diptera: Sciaridae): novel bioassays with female body extracts and fractions. Journal of Applied Entomology 132: 513-518.

Frouz J & Novakova A (2001) A new method for rearing the sciarid fly, Lycoriella ingenua (Diptera: Sciaridae), in the laboratory: possible implications for the study of fly-fungal interactions. Pedobiologia 45: 329-340.

Frouz J, Nováková A & Jones TH (2002) The potential effect of high atmospheric CO2 on soil fungi–invertebrate interactions. Global Change Biology 8: 339-344.

Gardiner RB, Jarvis WR & Shipp JL (1990) Ingestion of pythium spp by larvae of the fungus gnat Bradysia impatiens (Diptera: Sciaridae). Annals of Applied Biology 116: 205-212.

Geib SM, Filley TR, Hatcher PG, Hoover K, Carlson JE, Jimenez-Gasco MD, Nakagawa-Izumi A, Sleighter RL & Tien M (2008) Lignin degradation in wood-feeding insects. Proceedings of the National Academy of Sciences USA 105.

Gillespie DR & Menzies JG (1993) Fungus gnats vector Fusarium oxysporum f-sp radicis- lycopersici. Annals of Applied Biology 123: 539-544.

Gotoh T, Nakamuta K, Tokoro M & Nakashima T (1999) Copulatory behavior and sex pheromones in sciarid fly, Lycoriella mali (Fitch) (Sciaridae: Diptera). Japanese Journal of Applied Entomology and Zoology 43: 181-184.

112

Grewal PS & Richardson PN (1993) Effects of application rates of Steinernema feltiae (Nematoda: Steinernematidae) on biological-control of the mushroom fly Lycoriella auripila (Diptera: Sciaridae). Biocontrol Science and Technology 3: 29-40.

Grewal PS, Richardson PN, Collins G & Edmondson RN (1992) Comparative effects of Steinernema feltiae (Nematoda: Steinernematidae) and insecticides on yield and cropping of the mushroom Agaricus bisporus. Annals of Applied Biology 121: 511-520.

Grewal PS, Tomalak M, Keil CBO & Gaugler R (1993) Evaluation of a genetically selected strain of Steinernema feltiae against the mushroom sciarid Lycoriella mali. Annals of Applied Biology 123: 695-702.

Hamilton JGC, Ward RD, Dougherty MJ, Maignon R, Ponce C, Ponce E, Noyes H & Zeledon R (1996) Comparison of the sex-pheromone components of Lutzomyia longipalpis (Diptera: Psychodidae) from areas of visceral and atypical cutaneous leishmaniasis in Honduras and Costa Rica. Annals of Tropical Medicine and Parasitology 90: 533-541.

Hardy TN (1988) Gathering of fungal honeydew by Polistes spp. (Hymenoptera: Vespidae) and potential transmission of the causal ergot fungus. The Florida Entomologist 71: 374-376.

Harris MA, Gardner WA & Oetting RD (1996) A review of the scientific literature on fungus gnats (Diptera: Sciaridae) in the genus Bradysia. Journal of Entomological Science 31: 252-276.

Hausmann SM & Miller JR (1989) Ovipositional preference and larval survival of the onion maggot (Diptera: Anthomyiidae) as influenced by previous maggot feeding. Journal of Economic Entomology 82: 426-429.

He XF & Cane DE (2004) Mechanism and stereochemistry of the germacradienol/germacrene D synthase of Streptomyces coelicolor A3(2). Journal of the American Chemical Society 126: 2678-2679.

Hillbur Y, Celander M, Baur R, Rauscher S, Haftmann J, Franke S & Francke W (2005) Identification of the sex pheromone of the swede midge, Contarinia nasturtii. Journal of Chemical Ecology 31: 1807-1828.

Holighaus G & Rohlfs M (2016) Fungal allelochemicals in insect pest management. Applied Microbiology and Biotechnology 100: 5681-5689.

Hung R, Lee S & Bennett JW (2013) Arabidopsis thaliana as a model system for testing the effect of Trichoderma volatile organic compounds. Fungal Ecology 6: 19-26.

Hung R, Lee S & Bennett JW (2015) Fungal volatile organic compounds and their role in ecosystems. Applied Microbiology and Biotechnology 99: 3395-3405.

113

Hungerford HB (1916) Sciara maggots injurious to potted plants. Journal of Economic Entomology 9: (pp. 538-549).

Hurley BP, Slippers B, Coutinho TA, Wingfield BD, Govender P & Wingfield MJ (2007) Molecular detection of fungi carried by Bradysia difformis (Sciaridae: Diptera) in South African forestry nurseries. Southern Hemisphere Forestry Journal 69: 103-109.

Hussey NW (1968) Control of Sciarid flies in mushroom culture. Mushroom Growers Association Bull. 236: 338-340.

Hussey NW & Gurney B (1968) Biology and control of the sciarid Lycoriella auripila Winn. (Diptera: Lycoriidae) in mushroom culture. Annals of Applied Biology 62: 395-403.

Jaenike J (1978) On optimal oviposition behavior in phytophagous insects. Theoretical Population Biology 14: 350-356.

Jagdale GB, Casey ML, Grewal PS & Lindquist RK (2004) Application rate and timing, potting medium, and host plant effects on the efficacy of Steinernema feltiae against the fungus gnat, Bradysia coprophila, in floriculture. Biological Control 29: 296-305.

Jarvis WR, Shipp JL & Gardiner RB (1993) Transmission of Pythium aphanidermatum to greenhouse cucumber by the fungus gnat Bradysia impatiens (Diptera: Sciaridae). Annals of Applied Biology 122: 23-29.

Jess S & Kilpatrick M (2000) An integrated approach to the control of Lycoriella solani (Diptera: Sciaridae) during production of the cultivated mushroom (Agaricus bisporus). Pest Management Science 56: 477-485.

Jess S & Schweizer H (2009) Biological control of Lycoriella ingenua (Diptera: Sciaridae) in commercial mushroom (Agaricus bisporus) cultivation: a comparison between Hypoaspis miles and Steinernema feltiae. Pest Management Science 65: 1195-1200. joshi G, Mrig K, Singh R & Singh S (2011) Studies on seasonal abundance of adult mushroom flies. Crop Resistance 41: 218-221.

Kalb DW & Millar RL (1986) Dispersal of Verticillium alboatrum by the fungus gnat (Bradysia- impatiens). Plant Disease 70: 752-753.

Katerinopoulos HE, Pagona G, Afratis A, Stratigakis N & Roditakis N (2005) Composition and insect attracting activity of the essential oil of Rosmarinus officinalis. Journal of Chemical Ecology 31: 111-122.

114

Keates SE, Sturrock RN & Sutherland JR (1989) Populations of adult fungus gnats and shore flies in British Columbia container nurseries as related to nursery environment, and incidence of fungi on the insects. New Forests 3: 1-9.

Kennedy MK (1974) Survival and development of Bradysia impatiens (Diptera: Sciaridae) on fungal and non-fungal food sources. Annals of the Entomological Society of America 67: 745-749.

Kielbasa R & Snetsinger R (1980) Life-history of a sciarid fly, Lycoriella mali, and its injury threshold on the commercial mushroom. Pennsylvania Agricultural Experiment Station Bulletin: 1-14.

Kielbasa R & Snetsinger R (1981) The effect of mushroom mycelial growth on the reproduction of a mushroom infesting sciarid Lycoriella mali. Melsheimer Entomological Series: 15-18.

Kim S-R, Choi K-H, Cho E-S, Yang W-J, Jin B-R & Sohn H-D (1999) An investigation of the major dipteran pests on the oyster mushroom (Pleurotus ostreatus) in Korea. Korean Journal of Applied Entomology 38: 41-46.

Knaden M, Strutz A, Ahsan J, Sachse S & Hansson Bill S (2012) Spatial representation of odorant valence in an insect brain. Cell Reports 1: 392-399.

Kostelc JG, Girard JE & Hendry LB (1980) Isolation and identification of a sex attractant of a mushroom-infesting sciarid fly. Journal of Chemical Ecology 6: 1-11.

Krupke OA, Castle AJ & Rinker DL (2003) The North American mushroom competitor, Trichoderma aggressivum f. aggressivum, produces antifungal compounds in mushroom compost that inhibit mycelial growth of the commercial mushroom Agaricus bisporus. Mycological Research 107: 1467-1475.

Kuhne S & Heller K (2009) Sciarid fly larvae in growing media – Biology, occurrence, substrate, environmental effects and biological control measures: Proceedings of the International Peat Symposiumed. Amsterdam, Netherlands.

Kumara HNSM, S. , Thyagaraj NE & Ghosh SK (2013) Survey and Isolation of natural Incidence of different fungal pathogens against house files in different urban habitats. Journal of Biopesticides 6: 133-138.

Kwon-Chung KJ & Sugui JA (2013) Aspergillus fumigatus - what makes the species a ubiquitous human fungal pathogen? PLoS Pathogens 9.

Lai HY, Tam MF, Tang RB, Chou H, Chang CY, Tsai JJ & Shen HD (2002) CDNA cloning and immunological characterization of a newly identified enolase allergen from Penicillium

115

citrinum and Aspergillus fumigatus. International Archives of Allergy and Immunology 127: 181-190.

Leal WS, Hasegawa M & Sawada M (1992) Identification of Anomalas schonfeldti sex- pheromone by high-resolution GC-behavior bioassay. Naturwissenschaften 79: 518-519.

Leal WS, Hasegawa M, Sawada M & Ono M (1994) Sex-pheromone of oriental beetle, Exomala orientalis - identification and field-evaluation. Journal of Chemical Ecology 20: 1705- 1718.

Leath KT & Newton RC (1969) Interaction of a fungus gnat, Bradysia sp. (Sciaridae) with Fusarium spp. on alfalfa and red clover. Phytopathology 59: 257-258.

Lee H-S, Kim K-C, Park C-G & Shin W-K (1999) Description of fungus gnat, Lycoriella mali Fitch (Diptera: Sciaridae) from Korea. Korean Journal of Applied Entomology 38: 209-212.

Lemon KM (1992) Dispersal of the ergot fungus Claviceps purpurea by the Lauxaniid fly Minettia lupulina. Journal of the New York Entomological Society 100: 182-184.

Lewandowski M, Kozak M & Sznyk-Basalyga A (2012) Biology and morphometry of Megaselia halterata, an important insect pest of mushrooms. Bulletin of Insectology 65: 1-8.

Lewandowski M, Sznyk A & Bednarek A (2004) Biology and morphometry of Lycoriella ingenua (Diptera: Sciaridae). Biological Letters 41: 41-50.

Li HJ, He XK, Zeng AJ, Liu YJ & Jiang SR (2007) Bradysia odoriphaga copulatory behavior and evidence of a female sex pheromone. Journal of agricultural and urban entomology 24: 27-34.

Li HJ, He XK, Zeng AJ, Liu YJ & Jiang SR (2008) Evidence of a female sex pheromone in Bradysia odoriphaga (Diptera: Sciaridae). Canadian Entomologist 140: 324-326.

Lindquist RK, Faber WR & Casey ML (1985) Effect of various soilless root media and insecticides on fungus gnats. Hortscience 20: 358-360.

Llewellyn GC, Martin WA & Bean GA (1983) Aflatoxin production on two mushroom substrates. Journal of Food Safety 5: 113-118.

Ludwig SW & Oetting RD (2001) Evaluation of medium treatments for management of Frankliniella occidentalis (Thripidae: Thysanoptera) and Bradysia coprophila (Diptera: Sciaridae). Pest Management Science 57: 1114-1118.

Marín-Cruz VH, Cibrián-Tovar D, Méndez-Montiel JT, Pérez-Vera OA & Cadena-Meneses JA (2015) Control del mosco fungoso negro, Lycoriella ingenua (Dufour, 1839) y Bradysia

116

impatiens (Johannsen, 1912) (Dipteria: Sciaridae) en Pinus montezumae Lamb. Revista mexicana de ciencias forestales 6: 90-101.

Mattheis JP & Roberts RG (1992) Identification of geosmin as a volatile metabolite of Penicillium expansum. Applied and Environmental Microbiology 58: 3170-3172.

Meers TL & Cloyd RA (2005) Egg-laying preference of female fungus gnat Bradysia sp nr. coprophila (Diptera: Sciaridae) on three different soilless substrates. Journal of Economic Entomology 98: 1937-1942.

Mehelis N (1995) Biology and management of mushroom infesting sciarid fly (Diptera: Sciaridae) in relation to room - room dispersal: Oregon State University.

Mohrig W, Heller K, Hippa H, Vilkamaa P & Menzel F (2012) Revision of the black fungus gnats (Diptera: Sciaridae) of North America. Studia Dipterologica 19: 141-286.

Nasir H & Noda H (2003) Yeast-like symbiotes as a sterol source in anobiid beetles (Coleoptera, Anobiidae): Possible metabolic pathways from fungal sterols to 7-dehydrocholesterol. Archives of Insect Biochemistry and Physiology 52: 175-182.

Nickle WR & Cantelo WW (1991) Control of a mushroom-infesting fly, Lycoriella mali, with Steinernema feltiae. Journal of Nematology 23: 145-147.

O'Connor L & Keil CB (2005) Mushroom host influence on Lycoriella mali (Diptera: Sciaridae) life cycle. Journal of Economic Entomology 98: 342-349.

Olson DL, Oetting RD & van Iersel MW (2002) Effect of soilless potting media and water management on development of fungus gnats (Diptera: Sciaridae) and plant growth. Hortscience 37: 919-923.

Park IK, Choi KS, Kim DH, Choi IH, Kim LS, Bak WC, Choi JW & Shin SC (2006) Fumigant activity of plant essential oils and components from horseradish (Armoracia rusticana), anise (Pimpinella anisum) and garlic (Allium sativum) oils against Lycoriella ingenua (Diptera: Sciaridae). Pest Management Science 62: 723-728.

Park IK, Kim JN, Lee YS, Lee SG, Ahn YJ & Shin SC (2008) Toxicity of plant essential oils and their components against Lycoriella ingenua (Diptera: Sciaridae). Journal of Economic Entomology 101: 139-144.

Park KC & Baker TC (2002) Improvement of signal-to-noise ratio in electroantennogram responses using multiple insect antennae. Journal of Insect Physiology 48: 1139-1145.

117

Patel TK & Williamson JD (2016) Mannitol in plants, fungi, and plant–fungal interactions. Trends Plant Sci.

Prom LK, Lopez JD & Latheef MA (2003) Transmission of Claviceps africana spores from diseased to non-infected sorghum by corn earworm moths, Helicoverpa zea. Journal of Sustainable Agriculture 21: 49-58.

Pudelko K (2014) Effect of forced ventilation during composting on Agaricus bisporus substrate selectivity. International Biodeterioration & Biodegradation 93: 153-161.

Ramirez-Camejo LA, Torres-Ocampo AP, Agosto-Rivera JL & Bayman P (2014) An opportunistic human pathogen on the fly: Strains of Aspergillus flavus vary in virulence in Drosophila melanogaster. Medical Mycology 52: 211-219.

Revadi S, Vitagliano S, Rossi Stacconi MV, Ramasamy S, Mansourian S, Carlin S, Vrhovsek U, Becher PG, Mazzoni V, Rota-Stabelli O, Angeli S, Dekker T & Anfora G (2015) Olfactory responses of Drosophila suzukii females to host plant volatiles. Physiological Entomology 40: 54-64.

Richardson PN (1987) Susceptibility of mushroom pests to the insect-parasitic nematodes Steinernema feltiae and Heterorhabditis heliothidis. Annals of Applied Biology 111: 433- 438.

Rinker DL, Finley RJ, Wuest PJ, Snetsinger R, Tetrault R & Royse DJ (1984) The influence of phorid & sciarid populations on mushroom yield at commercial mushroom farms in Pennsylvania. Proceedings of the Pennsylvania Academy of Science 58: 70-72.

Rinker DL, Olthof THA, Dano J & Alm G (1995) Effects of entomopathogenic nematodes on control of a mushroom-infesting sciarid fly and on mushroom production. Biocontrol Science and Technology 5: 109-119.

Roland J, Denford KE & Jimenez L (1995) Borneol as an attractant for Cyzenis albicans, a tachinid parasitoid of the winter moth, Operophtera brumata l. (Lepidoptera: Geometridae). The Canadian Entomologist 127: 413-421.

Royse DJ & Beelman RB (2013) Six Steps to Mushroom Farming. The Pennsylvania State University, College of Agricultural Sciences, University Park, PA.

Sakai RK, Aparecida Rezende Vilela J, Sandes Pires M, Sanavria A & De Campos SG (2007) Occurrence of fungus Aspergillus flavus in colonies of Dermatobia hominis (Dipterous [Diptera]: Cuterebridae) in laboratory. Revista Universidade Rural Serie Ciencias da Vida 27: 161-162.

118

Savoie JM, Olivier JM & Laborde J (1996) Changes in nitrogen resources with increases in temperature during production of mushroom compost. World Journal of Microbiology & Biotechnology 12: 379-384.

Scarlett K, Tesoriero L, Daniel R & Guest D (2014) Sciarid and shore flies as aerial vectors of Fusarium oxysporum f. sp. cucumerinum in greenhouse cucumbers. Journal of Applied Entomology 138: 368-377.

Scheepmaker JWA, Geels FP, Smits PH & VanGriensven LJLD (1997) Location of immature stages of the mushroom insect pest Megaselia halterata in mushroom-growing medium. Entomologia Experimentalis et Applicata 83: 323-327.

Scheepmaker JWA, Geels FP, Van Griensven LJLD & Smits PH (1996) Substrate dependent larval development and emergence of the mushroom pests Lycoriella auripila and Megaselia halterata. Entomologia Experimentalis et Applicata 79: 329-334.

Schiøtt M, De Fine Licht HH, Lange L & Boomsma JJ (2008) Towards a molecular understanding of symbiont function: Identification of a fungal gene for the degradation of xylan in the fungus gardens of leaf-cutting ants. BMC Microbiology.

Shamshad A (2010) The development of integrated pest management for the control of mushroom sciarid flies, Lycoriella ingenua (Dufour) and Bradysia ocellaris (Comstock), in cultivated mushrooms. Pest Management Science 66: 1063-1074.

Shamshad A, Clift AD & Mansfield S (2008) Toxicity of six commercially formulated insecticides and biopesticides to third instar larvae of mushroom sciarid, Lycoriella ingenua Dufour (Diptera: Sciaridae), in New South Wales, Australia. Australian Journal of Entomology 47: 256-260.

Shamshad A, Clift AD & Mansfield S (2009) The effect of tibia morphology on vector competency of mushroom sciarid flies. Journal of Applied Entomology 133: 484-490.

Shirvani-Farsani N, Zamani AA, Abbasi S & Kheradmand K (2013) Toxicity of three insecticides and tobacco extract against the fungus gnat, Lycoriella auripila and the economic injury level of the gnat on button mushroom. Journal of Pest Science 86: 591-597.

Silva JB, Silva-Torres CSA, Moraes MCB, Torres JB, Laumann RA & Borges M (2015) Interaction of Anthonomus grandis and cotton genotypes: biological and behavioral responses. Entomologia Experimentalis et Applicata 156: 238-253.

Siyoum NA, Surridge K, van der Linde EJ & Korsten L (2016) Microbial succession in white button mushroom production systems from compost and casing to a marketable packed product. Annals of Microbiology 66: 151-164.

119

Smith JE, White PF, Edmondson RN & Chandler D (2006) Effect of different Agaricus species on the development of the mushroom sciarid fly Lycoriella ingenua. Entomologia Experimentalis et Applicata 120: 63-69.

Srivoramas T, Chaiwong T & Sanford MR (2012) Isolation of fungi from adult house fly; Musca domestica and the blow fly Chrysomya megacephala in Ubon Ratchathani province, Northeastern Thailand. International Journal of Parasitology Research 4: 53-56.

Stotefeld L, Holighaus G, Schutz S & Rohlfs M (2015) Volatile-mediated location of mutualist host and toxic non-host microfungi by Drosophila larvae. Chemoecology 25: 271-283.

Straatsma G, Gerrits JPG, Thissen JTNM, Amsing JGM, Loeffen H & Van Griensven LJLD (2000) Adjustment of the composting process for mushroom cultivation based on initial substrate composition. Bioresource Technology 72: 67-74.

Straatsma G, Samson RA, Olijnsma TW, Dencamp H, Gerrits JPG & Vangriensven L (1994) Ecology of thermophilic fungi in mushroom compost, with emphasis on Scytalidium thermophilum and growth-stimulation of Agaricus bisporus mycelium. Applied and Environmental Microbiology 60: 454-458.

Straatsma GO, Tineke W. Gerrits, Jan P. G. Amsing, Jos G. M. & Op Den Camp HJMVG, Leo J. L. D. (1994) Inoculation of Scytalidium thermophilum in button mushroom compost and its effect on yield. Applied and Environmental Microbiology 60: 3049-3054.

Su NY, Yu CJ, Shen HD, Pan FM & Chow LP (1999) Pen c 1, a novel enzymic allergen protein from Penicillium citrinum - Purification, characterization, cloning and expression. European Journal of Biochemistry 261: 115-123.

Tibbles LL, Chandler D, Mead A, Jervis M & Boddy L (2005) Evaluation of the behavioural response of the flies Megaselia halterata and Lycoriella castanescens to different mushroom cultivation materials. Entomologia Experimentalis et Applicata 116: 73-81.

Trienens M, Keller NP & Rohlfs M (2010) Fruit, flies and filamentous fungi - experimental analysis of animal-microbe competition using Drosophila melanogaster and Aspergillus mould as a model system. Oikos 119: 1765-1775.

Tung PS & Snetsinger R (1973) New technique for culturing a mushroom-infesting sciarid Lycoriella mali (Fitch). Melsheimer Entomology Series. No.11: 1-4.

USDA (2014) Specialty Crops: Mushroom Production. Agricultural Marketing Resource Center.

120

Wang Z-L, Lu J-d & Feng M-G (2012) Primary roles of two dehydrogenases in the mannitol metabolism and multi-stress tolerance of entomopathogenic fungus Beauveria bassiana. Environmental Microbiology 14: 2139-2150.

Wetzel HA, Wuest PJ, Rinker DL & Finely RJ (1982) Significant pests of the commercial mushroom. The Pennsylvania State University, College of Agriculture,, University Park, PA, USA.

White PF (1986) The effect of sciarid larvae (Lycoriella auripila) on cropping of the cultivated mushroom (Agaricus bisporus). Annals of Applied Biology 109: 11-17.

White PF (1997) The use of chemicals, antagonists, repellents and physical barriers for the control of Lycoriella auripila (Diptera: Sciaridae), a pest of the cultivated mushroom Agaricus bisporus. Annals of Applied Biology 131: 29-42.

White PF (1999) Comparative effects of three insect growth regulator insecticides and a dipteran-active strain of Bacillus thuringiensis on cropping of the cultivated mushroom Agaricus bisporus. Annals of Applied Biology 134: 35-43.

White PF & Gribben DA (1989a) Variation in resistance to diazinon by the mushroom sciarid Lycoriella auripila. Mushroom Science 12: 851-859.

White PF & Gribben DA (1989b) Variation in resistance to diazinon by the mushroom sciarid Lycoriella auripila, Vol. 2: Mushroom Science, Proceedings of XII International Congress on the Science and Cultivation of Edible Fungi (pp. 851–859).

Wicker-Thomas C (2007) Pheromonal communication involved in courtship behavior in Diptera. Journal of Insect Physiology 53: 1089-1100.

Wicklow DT & Yocom DH (1982) Effect of larval grazing by Lycoriella mali (Diptera: Sciaridae) on species abundance of coprophilous fungi. Transactions of the British Mycological Society 78: 29-32.

Witzgall P, Kirsch P & Cork A (2010) Sex pheromones and their impact on pest management. Journal of Chemical Ecology 36: 80-100.

Witzgall P, Proffit M, Rozpedowska E, Becher PG, Andreadis S, Coracini M, Lindblom TUT, Ream LJ, Hagman A, Bengtsson M, Kurtzman CP, Piskur J & Knight A (2012) "This is not an apple"-yeast mutualism in codling moth. Journal of Chemical Ecology 38: 949-957.

Yang Y, Zhang Y, Wang M, Li SS, Ma XY & Xu ZH (2015) Bioefficacy of entomopathogenic Aspergillus strains against the melon fly, Bactrocera cucurbitae (Diptera: Tephritidae). Applied Entomology and Zoology 50: 443-449.

121

Yi JH, Perumalsamy H, Sankarapandian K, Choi BR & Ahn YJ (2015) Fumigant toxicity of phenylpropanoids identified in Asarum sieboldii aerial parts to Lycoriella ingenua (Diptera: Sciaridae) and Coboldia fuscipes (Diptera: Scatopsidae). Journal of Economic Entomology 108: 1208-1214.

Yu F, Harada H, Yamasaki K, Okamoto S, Hirase S, Tanaka Y, Misawa N & Utsumi R (2008) Isolation and functional characterization of a β-eudesmol synthase, a new sesquiterpene synthase from Zingiber zerumbet Smith. FEBS Letters 582: 565-572.

Zhang L & Sun XY (2014) Changes in physical, chemical, and microbiological properties during the two-stage co-composting of green waste with spent mushroom compost and biochar. Bioresource Technology 171: 274-284.

Zhang X, Zhong YH, Yang SD, Zhang WX, Xu MQ, Ma AZ, Zhuang GQ, Chen GJ & Liu WF (2014) Diversity and dynamics of the microbial community on decomposing wheat straw during mushroom compost production. Bioresource Technology 170: 183-195.

VITA Kevin R. Cloonan

EDUCATION University of Idaho 2007-2011 B.S.: Entomology

University of California Davis 2011-2013 M.S.: Entomology

The Pennsylvania State University PhD: Entomology 2013-2017

PUBLICATIONS

Andreadis SS, Cloonan KR, Bellicanta GS, Paley K, Pecchia J & Jenkins NE (2016) Efficacy of Beauveria bassiana formulations against the fungus gnat Lycoriella ingenua. Biological Control 103: 165-171.

Andreadis SS, Cloonan KR, Myrick AJ, Chen HB & Baker TC (2015) Isolation of a female-emitted sex pheromone component of the fungus gnat, Lycoriella ingenua, attractive to males. Journal of Chemical Ecology 41: 1127-1136.

Cloonan K, Bedoukian RH & Leal W (2013) Quasi-Double-Blind Screening of Semiochemicals for Reducing Navel Orangeworm Oviposition on Almonds. PLoS One 8.

Cloonan KR, Andreadis SS & Baker TC (2016) Attraction of female fungus gnats, Lycoriella ingenua, to mushroom-growing substrates and the green mold Trichoderma aggressivum. Entomologia Experimentalis et Applicata 159: 298-304.

Cloonan KR, Andreadis SS, Chen HB, Jenkins NE & Baker TC (2016) Attraction, oviposition and larval survival of the fungus gnat, Lycoriella ingenua, on fungal species isolated from adults, larvae, and mushroom compost. PLoS One 11.