Corn Rootworms, Diabrotica Virgifera Virgifera Leconte (Coleoptera: Chrysomelidae),Transgenic Corn, Zea Mays L

Abstract GORSKI, STEPHANIE L. Multipartite Interactions among Western Corn Rootworms, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae),Transgenic Corn, Zea mays L. (Poales: Poaceae), and Soil Microbes. (Under the direction of Drs. Yasmin J. Cardoza and Fred L. Gould).

The western corn rootworm (Diabrotica virgifera virgifera LeConte, WCR) is one of the most serious corn (Zea mays L.) (Poales: Poaceae) pests in the Americas and is increasingly becoming a global problem. Moreover, WCR larval behavior is understudied because its cryptic, soil dwelling, larval habits makes it difficult to observe. Transgenic corn expressing Bt toxins, isolated from Bacillus thuringiensis bacteria, is a valuable tool for WCR management, but resistance development by WCR to transgenic Bt-expressing corn has already been reported. The factors or mechanisms involved in WCR resistance to Bt corn are unknown, but this resistance is likely due to a combination of environmental factors affecting both insects (such as behavior) and plants (such as toxin levels). Refuges, areas within or near fields of Bt corn that are planted with a non-Bt alternate host, are critical for resistance management, so we investigated WCR larval responses to non-Bt and Bt corn roots under scenarios mimicking the two types of refugia currently available, block/strip refuge and seed mix refuge. Our results show that rootworms were more likely to be found in contact with non-Bt corn than Bt corn under both refuge mimicking scenarios, which may lead to reduced larval toxin exposure. In addition, larvae were more often associated with root crowns than tips, regardless of plant genotype (Bt or non-Bt). We investigated expression of Bt toxin in corn roots grown under both refuge mimicking scenarios to determine if our observed larval responses could be explained by differences in toxin expression along the root. Larval preference for root crowns in our experiments could not be attributed to differential toxin expression along the root length, but may be due to other causes such as nutritional content of the root tissue. In fact, root crowns are known to be more nutritious than root tips. A particularly interesting finding made during these investigations is that when plants were infested with WCR, Bt corn roots expressed higher toxin levels when grown adjacent to non-Bt plants than when grown alone. Interestingly, however, the opposite was true when plants were not infested by WCR, where Bt corn roots expressed lower toxin levels when grown adjacent to non-Bt plants. Soil microbiota such as Metarhizium anisopliae (Metchnikoff) Sorokin (Hypocreales: Clavicitipitaceae) and Beauveria bassiana (Balsamo) Vuillemin (Hypocreales: Clavicitipitaceae) have proven effective at killing WCR in the laboratory, but have shown inconsistent results in the field. Although environmental factors such as humidity are likely factors reducing infection rates, lower effectivity may be additionally explained by evolved WCR behavioral non-preference to entomopathogens. To explore this possibility, we studied the behavioral responses of larval and gravid adult WCR towards Bt and non-Bt corn sprouts treated with various soil microbe species. We also investigated WCR’s time to emergence, survival, and sex ratios when feeding on plants produced from Bt and non-Bt corn seeds treated with the various microbes of interest, under greenhouse conditions. We found that treatment with several different species of soil microbes confounds WCR’s ability to avoid Bt-expressing transgenic seeds in both larvae and adults. We also found that rearing WCR on corn plants grown from microbe-treated seeds grown under greenhouse conditions reduced survival from egg to adulthood, regardless of microbe species, among those studied. Moreover, we found that responses to M. anisopliae were consistently significant, although disparate, as adults showed preference for M. anisopliae treated corn while larvae showed non-preference for the same corn treatment, suggesting that this may be a species of interest for further study with respect to its potential role in modulating WCR behavior. Plant-plant interactions have been described in many systems, and some prior research has suggested that transgenic plants may have decreased volatile emission relative to nontransgenic lines of similar genetic background. However, the possibility of plant-plant interactions affecting how transgenes are expressed in genetically modified plants has, to our knowledge, not been reported or explored. Our previous research on larval responses to Bt and non-Bt plants revealed that when Bt-expressing corn plants interact with non-Bt-expressing corn, this reduces Bt expression in uninfested, healthy plants. To explore this phenomenon further, we designed an experiment where we allowed transgenic Bt-expressing corn plants to interact with either Bt or non-Bt plants via aerial tissue, root tissue, both, or neither, and used qRT-PCR to quantify the amount of Bt toxin expressed in their root systems. We found that interactions between plants reduced the amount of the Bt gene that was expressed, regardless of corn genotype. Furthermore, we determined that plant aerial tissue interactions appear to be the drivers for Bt toxin downregulation previously observed.

Multipartite Interactions among Western Corn Rootworms, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae), Transgenic Corn, Zea mays L. (Poales: Poaceae), and Soil Microbes

by Stephanie L. Gorski

A dissertation submitted to the Graduate Faculty of North Carolina State University in partial fulfillment of the requirements for the degree of Doctor of Philosophy

Entomology

Raleigh, North Carolina 2015

APPROVED BY:

______Yasmin J. Cardoza Fred L. Gould Committee Co-Chair Committee Co-Chair

______Clyde Sorenson Dominic Reisig ii

Biography Stephanie Gorski (Steph) received her bachelor's in Biology and Philosophy at NC State, She received a Sigma Xi award for undergraduate research. A North Carolina native, she briefly considered moving to Massachusetts but would much prefer to stay down south where the weather is nice. It was during her BS program that Steph discovered that she enjoyed being an undergraduate Entomology laboratory assistant with Dr. Fred Gould much more than she enjoyed writing philosophy papers, which led her to seek a higher degree in Entomology. Steph joined the Entomology Department at NC State because of her interest in environmental sustainability and because she thinks insects are nicer looking and better smelling than large animals. Steph chose to pursue a PhD and to focuse her research on western corn rootworm behavioral and ecology and transgenic Bt corn. She considers herself lucky to have chosen a system of study that is so relevant, because corn rootworms are an important problem for farmers and our current knowledge and use of transgenic technology is constantly under flux, and also because she cares about reducing the environmental impact of human life on the planet! She is specifically interested in the ways in which complex external factors can influence the effectivity of transgenic Bt crops, how insect behavior, soil microbes, and the plant community all influence the interaction between the plant and insect. Learning about this system has taught her about not only ecology, genetics, biotechnology, and insect behavior, but also about the strange interactions between science and the humans it aims to serve. Steph’s research has led her to love outreach and teaching. She has done over 60 outreaches and spent countless hours explaining biotechnology to concerned friends. Her blog posts about biotechnology have been featured on large websites including the Genetic Literacy Project and BioFortified. During her time as a graduate student, she also participated in the Linnaean Games insect trivia competition, and her team won the national championship in 2014. In her free time, Steph enjoys being a jock and is currently organizing a women's arm wrestling team. iii

Table of Contents List of Tables ...... vi List of Figures ...... vii

Chapter 1 ...... 1

Introduction and literature review ...... 1

Western corn rootworm: History, Biology, and Ecology ...... 1 Western corn rootworm management ...... 3 Chemical Control...... 3 Crop Rotation ...... 4 Host Plant Resistance ...... 5 Bacillus thuringiensis (Bt), Bt Crops, and Bt resistance in the US ...... 5 Benefits of transgenic crops expressing Bt toxin ...... 8 Resistance management in Bt-expressing crops ...... 9 Resistance to Bt by WCR ...... 12 Impact of soil microbiota on rootworm behavior and survival ...... 15 Plant growth promoting organisms (PGPO) ...... 21 Plant defense and plant-plant interactions ...... 22 References ...... 27 Chapter 2 ...... 49

Transgenic Bt Toxin Expression in Corn Roots and Interactions with Western Corn Rootworm Under Two Refuge Mimicking Scenarios...... 49

Methods and Materials ...... 52 Insect Behavioral Responses to Non-Bt and Bt Corn under Block/strip and Seed mix Refuge Scenarios...... 53 Block/strip Mimicking (No Choice) Scenario ...... 53 Seed Mix Mimicking (Choice) Scenario ...... 54 Environmental Factors and Root Phenology and their Effects on Bt Toxin Expression...... 55 Effect of Plant Growth Substrate on Root Bt Expression ...... 55 Root Bt Toxin Expression Along the Root Length under Two Planting Scenarios and Insect Infestation Levels...... 55 iv

Results ...... 56 Insect Behavioral Responses to Non-Bt and Bt Corn under Block/strip and Seed Mix Planting Scenarios ...... 56 Block/strip Mimicking (No Choice) Scenario ...... 56 Environmental Factors and Root Phenology and their Effects on Bt Toxin Expression...... 57 Effect of Plant Growth Substrate on Root Bt Expression...... 57 Root Bt Toxin Expression Along the Root Length under Two Planting Scenarios and Insect Infestation Levels...... 57 Discussion...... 58 Acknowledgements ...... 62 References ...... 63 Figure Legends ...... 68 Chapter 3 ...... 73

Responses of Western corn rootworm, Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae), larvae to soil microbe seed treatments on Bt and non-Bt corn...... 73

Introduction ...... 74 Methods and Materials ...... 78 Experimental system ...... 78 Insects and corn seeds ...... 78 Microbial treatments ...... 79 Larval behavioral response to beneficial soil-borne microbes applied to non-Bt and Bt seeds in test tube arenas...... 80 Larval responses to microbe-treated non-Bt and Bt corn seeds in dual-choice test tube assays ...... 80 Larval behavioral response to microbe-treated non-Bt versus Bt seeds in dual- choice test tube arenas ...... 81 Larval response to interaction between seeds and microbes in dual-choice Petri dish arenas ...... 81 Effect of microbial seed treatments on adult oviposition preference ...... 83 Olfactory response of gravid females to microbe-treated corn seeds ...... 84 Insect survival, development, weight, and sex ratio under greenhouse mesocosms 85 Statistical Analyses...... 86 v

Results ...... 87 Discussion...... 89 Figure Legends ...... 100 Chapter 4 ...... 110

Plant-plant interactions on the expression of Bt in transgenic corn (Zea mays)...... 110

Introduction ...... 110 Methods and Materials ...... 113 Experimental system ...... 113 Planting scenarios ...... 113 Results ...... 115 Discussion...... 115 References ...... 119 Figure Legends ...... 123 Chapter 5 ...... 126

Conclusion ...... 126

List of Tables Table 1 For test tube assays, mean number of larvae in treated side, with standard errors for all values…………...……………………………………………………………….111

Table 2: Ct values for expression of Cry35 and Cry35 in root tissue, by interaction treatment and plant genotype (Bt or non-Bt)…………………………………………………..127

vii

List of Figures

Chapter 2

Figure 1: Experimental plexiglass arenas were constructed and were filled with a clear agar/MiracleGro growth medium to facilitate rootworm observations. Markings along the root denote assignment of root regions for insect behavior and toxin expression assessments.………………………………………………………………………….70 Figure 2: Block/strip (no choice) scenario. Mean number of 3rd instar WCR larvae in contact with corn root tissue per observation time, averaged over all observations in a 48-hour period for n=20. A) Number of larvae touching root tissue per segment per observation time on Bt and non-Bt corn roots. B) Number of larvae per observation time found in contact with different segments of root, from crown to tip. Numbers represent segment, with 1 representing the segment nearest the root crown and 4 representing the segment nearest the root tip. Different letters indicate significant difference at p<0.05……………………………………………………….…………71 Figure 3: Seed mix (choice) scenario. A) Mean number of 3rd instar WCR larvae in contact with corn root tissue over all observations in a 48-hour period for n=20. B) Number of larvae found associated with different regions of the root, from crown to tip. Numbers represent region, with 1 representing the region nearest the root crown and 4 representing the region nearest the root tip. Different letters indicate significant difference at p<0.0001.……………………………………………………………....72 Figure 4: Amount of Bt protein (ng/g dry weight) of root tissue. A) Cry34 and Cry35 protein in roots that were infested with 3rd instar WCR larvae vs. uninfested roots. B) Cry34 and Cry35 expression in corn roots grown next to non-Bt corn (choice) vs. corn roots grown alone (no choice) when agar arenas were used. * significant at p<0.05, ** significant at p<0.0001………...……..………………………………………...……73

viii

Chapter 3 Figure 1: Test tube arenas were constructed for our behavioral response assays. Five third- instar larvae were released in the center of an arena consisting of two test tubes affixed together. Alternate treatments were put in opposite ends of the arena, and insect choice was recorded by which side the insect was found in at 24 hours……104 Figure 2: Petri dish arenas were constructed for our plant/microbe interaction assays. Arenas were 10 cm Petri dishes with a layer of plain agar to retain moisture. Corn seeds treated for 48 hours or dipped for 10 seconds in microbial treatments were placed in the opposite sides of each arena, inside holes cut in agar. Third-instar larvae were released in the center and monitored for 15 minutes to determine choice of treatment. The location of the larvae was then checked at 6 hours and 24 hours…………...... 105 Figure 3: Y-tube olfactometer, used for adult choice experiments. A) Air tube where vacuum was attached for airflow B) Chamber where insects were introduced C) Acrylic cylinder body bifurcating into two arms, 60 cm long, 10cm wide each D) and E) Chambers where odor sources were introduced, 30cm long, 10cm wide F) and G) Flowmeter to measure rate of inflow airflow…………………………………………………………………………...... 106 Figure 4: Responses of 3rd instar WCR larvae to microbe treated versus untreated non-Bt and Bt corn seeds. A) Microbial non-preference scenario. Number of larvae on each side. Different letters indicate significant differences (p<0.05). The bars represent standard error. B) Larval responses to microbe-treated non-Bt versus Bt corn seeds. Number of WCR found in each side when given the choice between Bt and non-Bt corn treated with microbes within the same test tube arena……………………………..107 Figure 5: A) Number of female WCR observed within oviposition arenas containing corn seeds treated with water (control), Beauveria bassiana (BB), Metarhizium anisopliae (MA), or Bacillus subtilis (BS). Different letters indicate significant difference (p<0.05). B) Mean number of eggs laid by gravid adults in oviposition arena with Bt seeds. C) Mean number of eggs laid by gravid adults in oviposition arena with non- Bt seeds..…………………………………………………………………….…...... 108 ix

Figure 6: Mean number of eggs laid by gravid WCR within oviposition chambers containing either non-Bt or Bt seeds treated with microbes or soaked in plain water (control) for 48 hours……………………………………………………………………………109 Figure 7: Number of adult WCR emerging from greenhouse mesocosm plots. Different letters indicate significant differences (p<0.05). A) Number of adults emerging based on microbe NT= no treatment; BB= Beauveria bassiana; MA= Metarhizium anisopliae; BS= Bacillus subtilis; SM= Serratia marcescens; TH= Trichoderma harzianum. B) Number of adults emerging based on planting scenario (block strip, seed mix, or none); C) Number of adults emerging based on interaction between planting scenario and microbe…………………………………………….…….....110

Chapter 4

Figure 1: Ct values for expression of combined Cry34 and Cry35 toxins in root tissue of corn plants, shown by interaction treatment. F = Full (aerial and root tissue) interaction; Aerial= aerial tissue interaction only; Root= root interaction only; None= no communication (plant grown alone). Different letters indicate significant differences (p<0.0001)...... ……………………………………………………………………126

Chapter 1

Introduction and literature review

Corn rootworms, Diabrotica spp. (Chrysomelidae), are the most economically important pests of corn, causing yield losses and management costs estimated at around $2 billion per year in the United States alone (Paul Mitchell, personal communication). More insecticides in the US are used to manage rootworms than any other single pest (EPA 2005). Yet, their biology, ecology and behavior have been largely understudied due to the complexity of their habitat. Basic studies are necessary to help identify behavioral and /or biological weaknesses within these pest systems that can be used in management programs for these soil-dwelling species. Western corn rootworm: History, Biology, and Ecology Diabrotica virgifera virgifera LeConte, the Western corn rootworm (WCR), causes the most damage in the larval stage feeding on the roots of its host plant. This feeding can lead to corn lodging, secondary infection, reduced yield, and a reduction in drought tolerance (Bryson et al. 1953). Adult WCR beetles emerge when corn is in the flowering stage (Elliot et al. 1991), usually in July or August depending on location and weather (Pruess et al. 1968, Elliot et al. 1991). Adults may also cause minor damage to corn by feeding on silks, resulting in incompletely fertilized ears with only a few kernels, or ears that are not completely closed and are susceptible to destruction by other insects, fungi, birds, and weather (Bryson et al. 1953). Adult WCR then lay eggs in cracks in the soil, and the egg distribution is varied depending on microhabitat (Pruess et al. 1968). Oviposition occurs for a window lasting for 25 days (Elliot et al. 1991) to 38 days (Hein and Tollefson 1985). Western corn rootworms overwinter in the egg stage, where the soil insulates them from temperature fluctuations, although a very warm or very cold winter will cause mortality (Gustin 1981). Neonate WCR must find corn roots by navigating through pores in the soil (Gustin and Schumacher 1989). It is believed that CO2 released by corn roots is the major cue used

by larvae to trigger search behavior (Strnad et al. 1986), while other compounds such as DIMBOA serve as more specific larval cues (Johnson and Nielsen 2012, Robert et al. 2012). Root crowns are more nutritious, and WCR orient themselves towards the defensive chemicals emitted by root crowns (Robert et al. 2012). Over time, WCR will leave the root and move through the soil to search for roots that have not been fed upon (Strnad and Bergman 1987). Western corn rootworms are considered "the classic example of a man-made pest” (Metcalf 1986), because they may not have been a problem before the advent of modern farming techniques and because early incidents of pesticide resistance were brought on by overreliance on pesticides and single-pronged management techniques. Western corn rootworms' ecological role before monocropping is somewhat controversial. On the one hand, some authors believe that rootworm species, including WCR, have been a problem for corn production for approximately 5,000 years in Guatemala due to traditional cultural practices to prevent lodging and presence of Guatemalan strains of corn showing some resistance to rootworms (Melhus et al. 1954, Gray et al. 2009). Often, 1828 is cited as the first year Diabrotica were reported as pests because of their morphological and niche similarity with a certain “little white worm” attacking corn plants (see for example Melhus et al. 1954), although the author believed the “worm” to be a fly (Yancey 1828). Other authors (see for example Branson and Krysan 1981) claimed that WCR were probably polyphagous and nonpestiferous in Mesoamerica before they followed the introduction of corn north to the US. Reasons cited by the latter authors include the rarity of root damage on archaeological corn remains housed in the National Institute of Anthropology and History in Mexico, the inability of WCR larvae to distinguish host from nonhost plants over a distance as observed by the authors, lack of evolved specialized WCR defenses in corn, and the fact that beetle species in the same genus are polyphagous and multivoltine, while WCR’s univoltine life cycle implies that it is specialized to follow the life cycle of corn. Moeser and Hibbard (2005) add that WCR has not always been closely associated with corn because WCR was first described in Kansas in 1868 (LeConte 1868, Moeser and Hibbard 2005), although corn was not historically grown in Kansas (Weatherwax 1954, Moeser and Hibbard

2005); thus WCR was likely feeding on native grasses in this region at the time (Moeser and Hibbard 2005), and also the first documented damage to corn by WCR was not until decades later, in Colorado in 1909 (Gillette 1912, Moeser and Hibbard 2005). Because WCR feeds on corn, it is highly successful at invading new regions where corn is grown and has been reported as an invasive species in 22 European countries (EPPO 2011). Unlike larval WCR, which are, at most, oligophagous (Oyediran et al. 2004), adult WCR are polyphagous; in one study they had apparently consumed pollen from at least 19 different species of weeds when gut analyses were performed (Moeser and Vidal 2005). Adult polyphagy probably contributed greatly to WCR’s successful invasion of Europe, because it allows them to utilize weeds available in cornfields (Moeser and Vidal 2005). Western corn rootworms were first reported in Serbia in 1992, and Bayesian analysis indicates that there were at least three separate introductions into Europe (Miller et al. 2005). These insects are currently spreading through Europe at a rate of 60 to 100 km per year and 20 km per year without and with containment measures, respectively (Baufeld and Enzian 2001). Western corn rootworm management Because larval WCR are specialists, they are especially problematic in monocropped systems (O’Rourke and Jones 2011). Chemical control, cultural control including crop rotation, and host plant resistance have all been investigated for WCR control. Future control of WCR may include RNA interference, because WCR ingestion of double-stranded RNA triggers RNA interference (Baum et al. 2007); however, no such product is currently available. Chemical Control. Soil-based insecticides currently recommended as effective for rootworm control include Aztec (tebupirimphos/cyfluthrin), Counter (terbufos), Lorsban (chlorpyrifos), and Capture LFR (bifenthrin) (Gassmann and Weber 2012). However, rootworms have repeatedly shown ability to develop resistance to control methods. Researchers documented pesticide resistance in WCR as early as 1959, in both soil-based insecticides that targeted larvae (Ball and Weekman 1962) and aerial insecticides that targeted adults (Meinke et al. 1998). Insecticides to which WCR have developed resistance

include cyclodienes like aldrin (Ball and Weekman 1962), organophosphates like methyl parathion (Meinke et al. 1998), and carbamates like carbaryl (Meinke et al. 1998). This resistance can persist for years even after these pesticides are no longer used. For instance, there is resistance to aldrin in wild WCR populations in the United States although aldrin has not been used since 1972 (Wang et al. 2013) and among introduced WCR populations in Europe, though they are unlikely to have encountered any selection pressure from this insecticide in Europe, since aldrin is not regularly used there (Ciosi et al 2009). Methyl parathion and carbaryl resistance are likely caused by cytochrome P450-based oxidation (Scharf et al. 2001), while aldrin resistance appears to be linked to a gamma-aminobutyric acid receptor mutation (Wang et al. 2013). Crop Rotation. Since 1912, crop rotation has been a highly effective method of rootworm control (Gillette 1912), since rootworm larvae will die of starvation if they cannot locate a suitable host shortly after hatching. Mathematical models have shown that maize rotation need not occur every year to be effective; rotating after three consecutive years of maize seems to be sufficient (Szalai et al. 2014). Crop rotation also has additional benefits in terms of soil quality, including increased microbial biomass (McGill et al. 1986) and microbial diversity (Lupwayi et al. 1998). Unfortunately, rootworms have managed to overcome rotation in some areas. For instance, a close relative of WCR, the northern corn rootworm (Diabrotica barberi Smith and Lawrence), has adapted to two-year rotations by undergoing a two-year diapause in the egg stage (Krysan et al. 1984). Western corn rootworms have also developed resistance to crop rotation in some areas; however, this resistance is behavioral (Onstad et al. 2001). Reports of rotation- resistant WCR occurred in east central Illinois as early as 1987 (Levine et al. 2002). Soil sampling in soybean fields in 1995 led researchers to believe that damage to first-year corn was caused by oviposition in soybean fields (Levine et al. 2002). Further evidence was given by behavioral differences between the two populations; females from rotation-resistant regions tend to be more mobile (Knolhoff et al. 2006). Rotation-resistant WCR adults also have different gut bacterial profiles, which allow them to better utilize soybeans as a food source (Chu et al. 2013).

Host Plant Resistance. Types of plant resistance to insects can generally be categorized as antibiosis, non-preference/antixenosis, and tolerance (Painter 1958). To date, almost all reports of native resistance in maize can be attributed to a more tolerant root system (Riedell and Evenson 1993), which is not ideal because it may lead to WCR population increases in a given area; this concern has precluded development and commercialization of these resistant lines. According to El Khishen et al (2009), only three examples of non-transgenic corn that had antibiotic or antixenotic effects on WCR have ever been documented. Lines with high hydroxamic acid levels appeared to have reduced damage from WCR in Canadian fields (Assabgui et al. 1995), although they were found to be susceptible to WCR in Missouri (Moeser and Hibbard 2005). One other line found (Branson et al. 1983) appeared to have antibiotic or antixenotic effects and one line (El Khishen et al. 2009, Bernklau et al. 2010) had antixenotic effects, but no more specific information is available to date. Although Iowa State University, which developed our current system of root damage ratings, the University of Ottawa, which documented the high hydroxamic acid lines mentioned above (Assabgui et al. 1995), and the USDA-ARS, Brookings, SD, which evaluated many possible resistant lines, have all made useful contributions to our knowledge of WCR resistance (Moeser and Hibbard 2005), the difficulties of identifying an antibiotic or antixenotic WCR line mean that no one currently claims to have a commercially available non-transgenic corn line that is WCR resistant (Moeser and Hibbard 2005, El Khishen et al. 2009). Our struggle with controlling pests should not be viewed as a battle that can be won; at best we are on a metaphorical treadmill (Dover 1985). Because pest management is not one-size-fits-all and must rely on pluralist attitudes towards management approaches (Denholm and Rowland 1992), it is worth considering alternative, possibly additive or synergistic solutions. One solution to the problem of WCR damage has been suggested in the form of transgenic Bacillus thuringiensis. Bacillus thuringiensis (Bt), Bt Crops, and Bt resistance in the US. Bacillus thuringiensis Berliner (Bt) is a gram-positive, soil-dwelling bacterium that produces spores containing insecticidal crystal proteins (ICP) (Vilas-Bȏas et al. 2007), proteinaceous

parasporal crystals with insecticidal properties that are formed within the cell's cytoplasm (Bechtel and Bulla 1976). Bacillus thuringiensis is classified as part of the Bacillus cereus group, consisting of six closely related bacterial species: B. anthracis Cohn, B. thuringiensis Berliner, B. mycoides Flügge, B. pseudomycoides Nakamura, B. weihenstephanensis Lechner, and B. cereus (sensu stricto) Frankland & Frankland (Guinebretiere et al. 2008). Interestingly, B. thuringiensis, which are entomopathogenic bacteria, Bacillus anthracis, the anthrax bacterium, and Bacillus cereus, a usually harmless bacterium that may occasionally cause food poisoning, are often considered to be a single species because they are genetically similar, but they are phenotypically and ecologically very different (Vilas-Bȏas et al. 2007). Bacillus thuringiensis is global and ubiquitous, and can be isolated from such diverse habitats as mosquito larval breeding sites (Goldberg and Margalit 1977), agricultural and non- cultivated soils (Iriarte et al. 1998), stored grains (Iriarte et al. 1998), dead insects (Iriarte et al. 1998), live insects (Damgaard et al. 1997), and the leaves of crop plants (Damgaard et al. 1997) and trees (Smith and Couche 1991). More than 100 ICP genes have been isolated and sequenced from Bt strains; many have insecticidal properties that are active against Lepidoptera, Diptera, and Coleoptera, with a few active against other insect orders, nematodes, mites, and protozoa (Schnepf et al. 1998). These ICP are dimer proteins, with two sulfide-linked subunits, and most variants have a molecular weight of about 230,000 (Huber et al. 1981). These crystal proteins work by binding to brush border membrane vesicles in the insects’ midgut, disrupting ion transport in the midgut, and forming pores (Gill et al. 1992). Insect gut bacteria appear to facilitate Bt- mediated mortality as well, apparently because Bt induces otherwise benign gut bacteria to have pathogenic effects (Broderick et al. 2009). However, how gut bacteria affect Bt- mediated mortality differs greatly between species (Broderick et al. 2009, Raymond et al. 2010a). These interactions between Bt and gut bacteria are controversial; some authors maintain that Bt does not likely require facilitation by other bacteria to be effective, because it has an ICP, a modified cell surface to protect it from being destroyed in the insect gut, antimicrobial peptides that allow it to destroy competitors, mechansms allowing it to breach the insect gut membrane, and other complex adaptations that would be unlikely to develop in

a merely opportunistic pathogen (Raymond et al. 2010a). Up to 20-30% of a Bt cell’s dry mass can be ICP (Baum and Malvar 1995). However, many strains of Bt lack these crystal proteins (Stahly et al. 1978). These non-pathogenic isolates can reproduce faster than ICP- producing isolates, but ICP-producing isolates can suppress the reproductive rate of non-ICP- producing isolates, which is one reason these two types are kept in balance (Raymond et al. 2007). Bacillus thuringiensis isolates that produce ICP are common in nature and are more likely to proliferate in insect guts (Raymond et al. 2010b). These isolates, though common and insecticidal, do not cause constant epizootics (Raymond et al. 2010b), in part because of the balance between ICP-producing and non-ICP-producing isolates. In addition, Bt multiplies more rapidly in the frass of insects they have not killed than in the guts of insects, and when this frass falls to the ground, this allows Bt to colonize the phylloplane via newly emerged seedlings (Bizzarri and Bishop 2007). Bacillus thuringiensis’s ability to be taken up endophytically by plant tissues may also be responsible for its ability to enter the next generation of plants (Monnerat et al. 2009). The best-known of these crystal proteins are classified as Cry1 (lepidopteran-active), Cry2 (lepidopteran and dipteran-active), Cry3 (coleopteran-active) or Cry4 (dipteran-active) (Hofte and Whiteley 1989). In addition, there are vegetative insecticidal proteins (VIP) also expressed in Bt (Estruch et al. 1996). Insecticidal crystal proteins classified as Cry3 are active against Coleoptera such as WCR. Most ICP genes are sporulation-specific, but Cry3s are unique in that they are not (Baum and Malvar 1995). In Cry3 crystals, the ICP antigen and transcript increase during the vegetative growth phase of Bt and plateau during the stationary phase (Sekar 1988). Cry3 strains produce rhomboidal crystals with one major protein that is conserved across strains (Hofte and Whiteley 1989). These strains include Bt subsp. tenebrionis, Bt subsp. san diego, and Bt EG2158 (Hofte and Whiteley 1989). The extensive research available on the insecticidal bioactivity of Bt toxins is a major reason that these toxins are by far the most highly utilized microbially-derived insecticides (Beegle and Yamamoto 1992). Other reasons include these toxins’ high efficacy, specificity, and environmental friendliness (Sanchis 2011). Bt bacteria were first used for insect control

in the 1920s and were sold commercially in France in 1938, but were largely forgotten in the 1940s because of the wide availability, efficacy, and low cost of DDT and other synthetic insecticides (Sanchis 2011). Bacillus thuringiensis bacteria have been available in the US as a bioinsecticide, originally sold under the brand name Thuricide, labeled to control lepidopteran pests on food and forage crops since 1958 (Fisher and Rosner 1959). Given the specificity of Bt toxins, they are considered safe to non-target organisms, but high cost and variable efficacy kept their use restricted to a small niche market (Cannon 1993) until transgenic technology allowed insertion of chimeric Bt toxin-expressing genes into crop plants (Peferoen 1997). Transgenic Bt. Benefits of transgenic crops expressing Bt toxin. Transgenic Bt corn and cotton effective against Lepidopteran pests have been commercially available since 1996 (Carozzi and Koziel 1997), beginning in the US and Australia (Peferoen 1997). Decreasing pesticide use was cited in 42% of US farmers’ decisions to adopt Bt cotton, and adoption of Bt cotton has been shown to reduce insecticide use in the Southeastern US (Fernandez-Cornejo and McBride 2000). More recent reviews have shown that with adoption of Bt-expressing crops, pesticide use on corn has dropped from 0.21 pounds per acre to 0.02 pounds per acre, with only 9 percent of corn farmers in the US using insecticides in 2010 (Fernandez-Cornejo et al. 2014). Use of transgenic Bt-expressing crops can also lead to area-wide suppression of some pests such as European corn borer (Ostrinia nubilalis Hübner, Lepidoptera: Crambidae) (Hutchison et al. 2010) and pink bollworm (Pectinophora gossypiella Saunders, Lepidoptera: Gelechiidae) (Fernandez-Cornejo et al. 2014). Brookes and Barfoot (2010) estimated that the energy savings of transgenic crops in terms of reduced spraying and reduced tillage were potentially up to the equivalent of removing 6.9 million cars from the roads annually. In India, pesticide use on eggplants could also be reduced 42% with the introduction of Bt eggplants (Choudhary and Gaur 2009). In a review of economic studies of Bt crops (corn and cotton), the majority of research shows an economic benefit to Bt technology adoption by farmers in the US (Fernandez-Cornejo et al. 2014). Developing countries may also benefit economically from

transgenic Bt crops (see, for example, Ismael et al. 2002, Qaim and Janvry 2005, Choudhary and Gaur 2009, Subramanian and Qaim 2009, Pray et al. 2001, Morse and Mannion 2009; but see Dowd-Uribe 2014). Chinese farmers are also less likely to experience pesticide poisoning when they adopt transgenic Bt cotton (Hossain et al. 2004, Pray et al. 2001). Resistance management in Bt-expressing crops. The continued success of Bt crop use to date, and thus the continued access to benefits such as pesticide reduction and increased income to farmers, has been attributable to the rigorous resistance management plans required for their use (Bates et al. 2005). Thus, Bt crops must be coupled with refuges of non-transformed, susceptible plants of the same species (EPA 2014a). In the US, refuge requirements for Lepidopteran-targeting Bt crops range from no structured refuge at all for cotton (EPA 2014 b), to 5% of the crop for corn with pyramided traits (Gray 2011), to 50% for single-toxin corn grown in cotton-growing areas (EPA 2010 a). However, some authors estimate that about one in four farmers will not fully comply with refuge requirements (Jaffe 2009), though other estimates are somewhat lower (Gray 2011). In the United States, the Environmental Protection Agency (EPA) requires companies producing transgenic Bt seed to monitor for insect resistance development (EPA 2001), and has the authority to cancel pesticide registrations in the event of nocompliance by the companies (Taylor and Tick 2003). Resistance developed in one country can spread to another country (Okeke and Edelman 2001); thus, insects developing resistance to Bt crops in other parts of the world where refuge strategies are not well implemented present an imminent threat to US agriculture as resistant insect populations could be introduced and established. Of special concern are countries that share borders with the US, and those in Central America, South America, and the Caribbean. Mexico allows a 4% refuge in cotton when the refuge is not treated with pesticides (Traxler et al. 2001), but Bt corn has been indefinitely suspended because of concerns about contaminating native landraces of corn and also because some Mexican organizations such as Via Campesina are against the use of transgenic crops because of food-sovereignty issues (McAfee 2008). Argentina requires a 20% refuge with insecticide use in refugia (Qaim and de Janvry 2005).

Countries in other continents may also have lax refuge planting accountability. South Africa requires a minimum of 5%, but does not monitor compliance, which has been historically low (Kruger et al. 2009). In China, there is no refuge requirement, but plants that express Bt are normally intercropped with other plants that are not closely related but are hosts to the same insect pest; however, the effectiveness of this strategy is in question, because gene flow of old world cotton bollworm (Helicoverpa armigera Hübner, Lepidoptera: Noctuidae) among populations on Bt cotton and sesame seed or peanut is much lower than gene flow between populations from Bt and non-Bt crops of the same species (Tan et al. 2001). Brazil does not enforce refuges, (Dominic Reisig, personal communication), though they are strongly recommended (IRAC 2015). The refugium concept is based on the theory that since more insects should emerge from susceptible plant fields, any Bt-resistant insects that might emerge are more likely to mate with susceptible insects (Gould 1998), thus slowing the spread of resistance genes (Taylor and Georghiou

1979). Additionally, if the Bt expression rate is high enough relative to the insect’s LD50, then, provided that the resistance gene is not 100% dominant, the toxin should kill heterozygous resistant insects as well as homozygous susceptible insects, making resistance development more difficult (Gould 1998). Since it is difficult to determine the LD50 of a hypothetical heterozygote, the EPA has defined “high dose” as “25 times the toxin concentration needed to kill susceptible larvae” (EPA 1998). The success of this plan depends on the dispersal habits of the target species; if resistant insects don’t move into refuge fields, or susceptible insects into Bt crop fields to mate, resistant insects will breed only with other resistant insects and produce homozygous resistant offspring (Gould 1998). The latter phenomenon is difficult to document in the field, but some researchers have shown that this mating dispersal behavior does in fact occur in lepidopteran systems, supporting the effectiveness of currently mandated refuge strategies. For example, researchers examined behavior in the rice stem borers (Cuong and Cohen 2003) and gene flow in the European corn borer (ECB) Ostrinia nubilalis (Crambidae) (Bourguet 2004). Cuong and Cohen (2003) observed that most newly emerged male rice stem (Scirpophaga incertulas Walker and Chilo suppressalis Walker, Lepidoptera: Crambidae) borers would disperse rather than attempt to

mate with a caged female near the site of emergence. Bourguet (2004) reported that different races of European corn borer (Ostrinia nubilalis, Lepidoptera: Crambidae) appear in similar percentages across generations and across locations. This led the researchers to conclude that European corn borers disperse over large areas to mate and, therefore, refuges are an effective strategy for resistance prevention in this particular species. A study by Marquart and Krupke (2009) reported that, similar to Lepidopteran pests, rootworm males appeared to disperse readily, with refuge-produced susceptible beetles appearing as frequently 200 m deep into Bt fields as at the edge of Bt fields. However, they noted that this conclusion was preliminary because it was based on insect gut contents; thus, Bt males that had not fed were potentially misclassified as refuge males. Moreover, the same study found that female WCR did not disperse immediately and began calling for mates near where they emerged, but no females mated while being observed, and their subsequent behavior was not recorded (Marquart and Krupke 2009). This suspected difference in mating and dispersal behaviors between rootworms and lepidopteran pests is of great importance; since the introduction of Bt-expressing corn in the 1990s, refugia have been planted in single blocks or strips away from Bt corn. Another problem with this refuge design for rootworms, in addition to the mating and dispersal behaviors, is that resistant and susceptible beetles may not have an opportunity to mate because they are separated in time; beetles emerge from Cry34/35Ab corn later than from refuge corn (Storer et al. 2006). Males become less likely to mate as they age, even when virgin females are not given a choice of males (Kang and Krupke 2009), while females appear to mate only once, shortly after emergence (Hill 1975). For these reasons, seed-mix refugia with non-transgenic corn interspersed with Bt corn, are being deployed as an alternative for rootworms (EPA 2010 b). In seed-mix refugia, beetles emerge from Bt and non-Bt corn at about the same time, which is helpful for resistance management purposes (Murphy et al. 2010). Unfortunately, this may be because neonates are eating refuge corn and then moving to Bt plants as they get older, or showing a non- preference for Bt corn and delaying their development while they search for susceptible corn (Murphy et al. 2010). Alternately, if larvae move from refuge to Bt plants and die, this will reduce the effective size of the available refuge (Head and Greenplate 2012). Either way,

this may facilitate Bt resistance development, because second and third instars are much less susceptible to the toxin than neonates (EPA 2002). This loss of susceptibility may only be because larger insects require a higher dose of toxin, as shown in other Coleoptera, such as the Colorado potato beetle (Leptinotarsa decemlineata Say, Coleoptera: Chrysomelidae) (Ferro and Lyon 1991). The rarity of Bt resistance development by target pest insects thus far supports the idea that resistant pests are mating with susceptible pests and that the high dose and refugia management strategies are effective (Head and Greenplate 2012). Despite worldwide monitoring (Sivasupramaniam et al. 2007), only five incidents of Bt resistance have been reported (Liu et al. 2008, Tabashnik 2008, Gassmann et al. 2011) if WCR (discussed in the following sections) are included. In at least one case, in corn earworms (Helicoverpa zea Boddie, Lepidoptera: Noctuidae) on cotton in the southern US, lack of high dose Bt expression appears to be a major factor in resistance development (Tabashnik et al. 2008). The other three resistance incidences on record for Lepidoptera, fall armyworms (Spodoptera frugiperda Smith, Lepidoptera: Noctuidae) on corn in Puerto Rico, cotton bollworms (Helicoverpa armigera Hübner, Lepidoptera: Noctuidae) on cotton in China, and stem borers (Busseola fusca Fuller, Lepidoptera: Noctuidae) on corn in South Africa, are suspected to be caused by suboptimal refuge planting (Matten et al. 2008, Liu et al. 2008, van Rensburg 2007). Van Rensburg (2007) noted that South African farmers historically have not irrigated their refuge corn crops, which may make them less attractive to pests. China does not generally use refuges, and some farmers use off-market seed that may not express a high dose of Bt toxin in the plant tissue (Liu et al. 2008). Since S. frugiperda is multivoltine in Puerto Rico and multiple crops may be planted in the region, more generations of the pest may occur per year than on the mainland U.S., speeding resistance development, while Puerto Rico’s island habitat reduces gene flow compared to mainland regions, and this may also speed resistance development (Matten et al. 2008). Resistance to Bt by WCR. Adding to the threat of resistance development in rootworm-resistant Bt crops is the fact that event 59122 (Cry34/35Ab, Herculex, DuPont Pioneer, Des Moines, IA), unlike lepidopteran-active events, does not meet the necessary

standards to qualify as high-dose by EPA standards (Binning et al. 2010). Therefore, the effects refuges might have on resistance development in these cases needs to be reexamined (Binning et al. 2010). Exposure to low Bt expression in plants or target tissues may be another mechanism by which rootworms develop resistance. For instance, volunteer Bt corn regularly contaminates corn fields rotated to soybean. Volunteer Cry3Bb corn is known to produce lower levels of toxin than cropped Cry3Bb corn, presumably because soybean fields are not fertilized with nitrogen (Krupke 2009) or because of cross-pollination between Bt and non-Bt plants (Chilcutt and Tabashnik 2004). Older corn plants in the V9 stage (plants in the vegetative stage with nine leaves) also express significantly lower levels of Cry3Bb than younger (V4) plants (Vaughn et al. 2005). Gray et al (2007) reported increased late-season damage in Bt corn in Illinois. Hibbard et al. (2009) suggested the possibility that late- hatching WCR might be exposed to lower toxin levels expressed by older plants and therefore survive, but field tests in Missouri did not show significant variation in survival between early and late hatching WCR on Bt corn. Altogether, the above-mentioned reasons made Bt crop resistance development in rootworm populations likely. Thus not surprisingly, the first confirmed case of rootworm resistance to transgenic Bt corn in the field occurred in Iowa in 2009 (Gassmann et al. 2011). Incidents of potential resistance in rootworms were correlated with a lack of crop rotation by farmers, indicating that the cultural practices of farmers are linked to the potential for resistance development in insects (Gassmann et al. 2011). Refuge compliance among farmers in the problem fields was not known, but other research has shown compliance to be around 63% (Jaffe 2009) to 82% (Gray 2011). Since then, eleven WCR populations in Iowa have confirmed resistance to Bt corn (Cullen et al. 2013), and reduced susceptibility to Bt (which may indicate resistance but has a lower standard of evidence) has also been confirmed in Colorado, Minnesota, Nebraska, and South Dakota (EPA 2014), indicating the phenomenon is widespread among US WCR populations. The Western corn rootworm’s ability to develop resistance to other forms of control is a call for caution. A WCR-active Bt protein variant, Cry3Bb, has been available in YieldGard Rootworm ® corn seed since 2003. YieldGard has been shown to consistently

reduce root damage by these insects (Vaughn et al. 2005). Other Bt proteins, Cry34Ab1 and Cry35Ab1, have been commercialized for rootworm protection as Herculex ® RW. Under constant exposure in both greenhouse conditions with corn, and under laboratory conditions with artificial diet, WCR have been shown to develop resistance to Cry3Bb corn in as little as 3 generations (Miehls et al. 2008). Field-collected WCR fed

Cry3Bb-incorporated diet showed a twelvefold natural variation in LC50 (Siegfried et al. 2005). Field studies have shown that the rootworm protection offered by Bt varies with location, with all of nine hybrids tested showing less protection in areas of Illinois known to have rotation-resistant variant rootworms (Gray et al. 2007). Values for EC50 and LC50 were also much higher for larvae collected in rotation-resistant counties, so some cross-resistance may be occurring in those areas (Gray et al. 2007). In order to slow the spread of WCR Bt resistance among US rootworm populations, it is therefore necessary to maximize the chances of rootworm larvae receiving a lethal dose of the toxin even under ever-changing field conditions. For example, nitrogen availability has a partial ability to increase Bt toxin expression in plants, although Bt toxin expression varies widely among plants (Marquardt et al. 2014). Although seed-mix refuges (“Refuge in a bag”) may address concerns over rootworm adult dispersal and mating behaviors, some scientists have voiced the concern that if resistant and susceptible plants are close enough that larvae can move from a susceptible to a resistant plant during the course of their development, resistance evolution may be exacerbated (Murphy et al. 2010). This would be the result of fostering insect behavioral non-preference, by favoring WCR that can tolerate lower Bt doses, or by affecting exposure time/doses for larvae as they move back and forth from susceptible to Bt corn roots. Under such scenarios, mathematical models show that resistance may evolve much faster than with a traditional block refuge (Mallet and Porter 1992). Alternatively, rootworms moving away from non-Bt refuge plants may shrink the effective size of the refuge (Head and Greenplate 2012). Cry3Bb corn in seed mixes showed more root damage than when an equivalent percent of refuge was planted as a block or strip (Murphy et al. 2010), indicating that WCR might be feeding on susceptible plants as neonates and then moving to Bt plants as they aged, providing them with the dose of toxin

necessary to exert selection pressure. Diamondback moth larvae (Plutella xylostella, Lepidoptera: Plutellidae) showed a higher number of larvae on Bt –expressing broccoli plants when these plants were interspersed with, rather than separated from, refuge plants, although these larvae are more mobile than WCR larvae (Tang et al. 2001). Bacillus thuringiensis toxin expression in corn roots varies with distance from the root crown, with a tendency towards higher expression in root tips, although there is significant interaction between distance and plant age (Meissle et al. 2009). This could potentially increase the chance that a neonate would ingest a lethal dose of toxin; however, Clark et al. (2006) observed divergent behavior in neonate WCR offered Bt corn roots. In this study, some larvae fed on Bt corn roots and then appeared lethargic and flattened until they died, presumably from toxin ingestion, whereas other larvae actively moved up and down root tissue, appearing to sample from root hairs or root tissue. These larvae had visibly empty midguts and also died, presumably from starvation. The researchers suggest that since the larvae that starved appear to detect toxin expression, if transgenic corn had some roots that expressed Bt at a lower rate than others, this might provide a possible route of resistance development. For all these reasons, in Chapter 2 we investigated whether rootworms preferentially feed on different portions of roots in transformed and untransformed corn, and whether this preference is correlated with differential Bt toxin expression levels. We believe this information would help assess the potential for Bt toxin behavioral escape by rootworms, which could lead to faster resistance development. Impact of soil microbiota on rootworm behavior and survival Soil microbes are more diverse than aquatic microbes, and far more diverse than eukaryotic organisms (Torsvik and Øvreas 2002). Dangar et al. (2008) found 0.01 – 0.05 x 105 CFUs fungi, 0.1x104 CFUs actinomycetes, 0.13-0.25x106 CFUs Bt, and other bacteria in the range of 106 CFUs per gram of soil in various soil samples in India. Bacillus thuringiensis was found in 100% of Indian soil samples in Dangar et al. (2008) and 6% of Iranian soil samples in Salehi et al. (2008). Corn rootworm neonates search for a host by navigating through pores in soil (Gustin and Schumacher 1989), where they are surrounded

by microbial life, such that one might expect these insects to have complex interactions with these soil microbes. Numerous species of fungi and bacteria have been reported in association with various species of rootworms. For example, an entomophthoralean fungus was found naturally infecting northern corn rootworms (Diabrotica barberi Smith and Lawrence) in New York State (Naranjo and Steinkraus 1988), but little work appears to have been done on this fungus since then. Chromobacterium subtsugae Martin (Neisseriales: Neisseriaceae), a Gram-rod soil bacterium lethal to Colorado potato beetle (Leptinotarsa decemlineata, Coleoptera: Chrysomelidae), was recently found to be toxic to both adult and larval stages of WCR and the Southern corn rootworm (Diabrotica undecimpunctata howardi L., Coleoptera: Chrysomelidae or SCR) (Martin et al. 2007). The much lower mortality rates in larval rootworms were attributed to the fact that the larvae did not appear to feed on infected diet (Martin et al. 2007). A Wolbachia bacterium species is believed to be responsible for the reproductive incompatibility between WCR and another subspecies of D. virgifera, the Mexican corn rootworm (D. v. zeae) (Giordano et al. 1997). A novel spiroplasma was isolated from SCR. This microbe species is not believed to be pathogenic, but may be an ideal vector for transgenic toxins (Carle et al. 1997). Prischmann et al. (2008) found two Serratia spp. isolates that appear to persist in corn roots and were also present in discolored (presumed infected) Diabrotica spp. However, most work on rootworm pathogens has focused on Beauveria bassiana (Hypocreales: Clavicipitaceae) and Metarhizium anisopliae (Hypocreales: Clavicipitaceae). The first recorded use of B. bassiana for pest control was in 1839, when it was observed that emptying contaminated pans used for rearing silkworms (Bombyx mori L., Lepidoptera: Bombycidae) on unidentified pest larvae in a tree caused the infestation to disappear (Steinhaus 1956). Much research has centered on B. bassiana's applications, particularly for management of European corn borer (Ostrinia nubilalis) (see for example Lefebvre 1934, Bing and Lewis 1991, 1993). Infections from B. bassiana render insects sluggish and sometimes discolored; eventually the insect is overtaken with hyphae and mummified (Lefebvre 1934). Indigenous to corn fields regardless of tillage (Bing and Lewis 1993), B.

bassiana maintains an endophytic relationship with corn (Bing and Lewis 1991). Ubiquitous in the environment, B. bassiana and can be found in tree bark and in soil (Doberski and Tribe 1980). When applied to adult WCR females, B. bassiana reduces fecundity, while leaving egg viability apparently unaffected (Mulock and Chandler 2001). This reduction of total number of eggs appears most dramatic when the fungus is applied to 10-day-old adult females, an age when WCR eggs are being produced inside the ovary, probably because the physiological stress interferes with egg development (Mulock and Chandler 2001). However, Bruck and Lewis (2002) found no correlation between increased numbers of B. bassiana CFUs applied on leaf collars and increased infection rates of confined WCR adults, possibly because the collar region they treated was difficult for beetles to access for feeding. Consolo et al. (2003) found mortality rates of up to 70% on third-instar Diabrotica speciosa Germar (Coleoptera: Chrysomelidae) larvae due to a strain of B. bassiana isolated from a different leaf beetle (Maecolaspis bridarolli Cabrera, Coleoptera: Chrysomelidae). Some Diabrotica spp. adults emerge with natural infections of B. bassiana, and although infection rates were not observed to exceed 3.17%, this translates into tens of thousands of diseased individuals in a field situation (Bruck and Lewis 2001). Metarhizium anisopliae, a fungus which causes symptoms similar to B. bassiana, was first researched as a control method in 1879 on the wheat cockchafer (Anisoplia austriaca Hrbst., Coleoptera: Scarabaeidae) (Steinhaus 1956). Both M. anisopliae and B. bassiana decrease corn root damage caused by corn rootworm (Krueger and Roberts 1997). In field studies, M. anisopliae decreases goosenecking (curvature of corn stems caused by root damage) and decreases adult SCR emergence in SCR-infested corn at a rate of 9.3 g of mycelial particles/ 6.1 m row (Krueger and Roberts 1997). Importantly from a resistance management standpoint, when combined with Bt toxin from Bt-expressing transgenic corn, M. anisopliae appears to have additive, rather than antagonistic or synergistic, effects on WCR larval development and adult mortality (Meissle et al. 2009). Ideally, a strain of M. anisopliae intended for rootworm control would be one that was isolated from a rootworm or closely related species of insect; strains of M. anisopliae isolated from rootworms have a

higher virulence to rootworm species, and adults are more susceptible to infection than larvae (Pilz et al. 2007). Understanding microbiota's role in insect mortality depends on our knowledge of insect behavioral responses to these microorganisms (Baverstock et al. 2010). The behavioral defenses of various ant species against entomopathogens, including grooming, antibiotic secretions, nest hygiene, avoidance, dispersal, and colony movement, were reviewed by Oi and Pereira (1993). Repellence is an important defense; for instance, beneficial predatory mites (Phytoseiulus persimilis Evans) (Mesostigmata: Phytoseiidae) were repelled by prey two-spotted spider mites (Tetranychus urticae Koch, Trimbidiformes: Tetranychidae) infected with entomopathogenic B. bassiana, but only if the fungi were allowed to contact the prey items for 48 hours (Seiedy et al. 2013). Some insects have behavioral responses that help them recover from existing infections. For instance, behavioral fevers occur when an insect attempts to move to a hotter or sunnier location. Ouedraogo et al. (2003) showed that M. anisopliae-infected locusts (Locusta migratoria L., Orthoptera: Acrididae) that were allowed to thermoregulate near a light bulb were better able to keep fungal growth in check, as measured by presence of hemocytes and blastospores. Adaptations to fungal pathogens are particularly important to soil-dwelling insects, because of the large naturally occurring fungus population in soils. For instance, the high density and high humidity in termite colonies make them an ideal environment for fungal infections, but fungal epizootics in termites are rarely observed. Yanagawa and Shimizu (2007) explain this by showing that grooming behavior in Coptotermes formosanus Shiraki (Blattodea: Rhinotermitidae) is apparently effective in removing M. anisopliae. To explain low mortality rates of M. anisopliae on Japanese beetles (Popillia japonica Newman, Coleoptera: Scarabaeidae), Villani et al. (1994) showed that the beetle grubs would not feed on grass infected with M. anisopliae in choice arenas. Japanese beetle grubs placed in test tubes with infected and uninfected soil will move away from the infected soil, though it may take grubs 48 to 96 hours to show a significant difference in choice (Fry et al. 1997). Mole crickets, especially the southern mole cricket (Scapteriscus borellii Giglio-Tos, Orthoptera: Gryllotalpidae), spend less time and make fewer tunnels in B. bassiana- treated soil, although

they cannot detect it from a distance (Thompson et al 2007). Wireworms (Agriotes obscurus L., Coleoptera: Elateridae), migrate to avoid M. anisopliae, although significantly fewer of them migrate when food is present (Kabaluk and Ericsson 2007). However, some species of entomopathogenic fungi may have developed counter- adaptations to attract their insect hosts. Black vine weevil larvae (Otiorhynchus sulcatus Fabricius, Coleoptera: Curculionidae) orient towards plants whose soil has been incorporated with M. anisopliae over untreated or insecticide-treated plants, though they do not orient towards M. anisopliae in the absence of host plants (Kepler and Bruck 2006). Japanese beetles oviposit more in M. anisopliae- treated soil, perhaps because CO2 is produced both by growing roots and by the fungus, and may act as a stimulus, although M. anisopliae reduces larval survival (Villani et al 1994). Interactions between entomopathogens and the plants and animals in their environment are also crucial modulators of insect behavior. Berdegué and Trumble (1997) found that celery (Apium graveolens L., Apialis: Apiaceae) plant defensive compounds reduce beet armyworms’ (Spodoptera exigua Hübner, Lepidoptera: Noctuidae) non- preference of commercially prepared non-transgenic Bt toxin with respect to diet consumption by beet armyworms. The authors noted that the deterrent properties of the plant defense compounds lead to one of two conclusions: 1) non-preference increases survival rates of larvae by reducing Bt intake, or 2) non-preference decreases survival rates of larvae by reducing food intake and therefore developmental rates, leaving larvae more susceptible to mortality from environmental factors or biocontrol agents. Green June beetle grubs (Cotinis nitida L., Coleoptera: Scarabaeidae) are more susceptible to nematode infections at higher insect population densities, presumably because of increased activity and intraspecies biting wounds (Townsend et al. 1998). The three-way interactions among herbivores, plants, and microbes have been described on many levels, but rarely in the context of entomopathogens as bodyguards actively recruited by the plant (Elliot et al. 2000). Yet Smith and Couche (1991) found significant numbers of Bacillus thuringiensis CFUs on tree leaf surfaces and suggested a mutualistic or commensalistic relationship between the plant and bacteria. Because fungal entomopathogens tend to have a wide host range making

them suitable for plant defense, and because they rely on plant phylloplane conditions to survive and reproduce, fungi may be the most likely entomopathogens to engage in these tritrophic interactions (Cory and Hoover 2006). Volatiles from damaged cassava (Manihot esculenta Crantz, Malpighiales: Euphorbiaceae) plants increase sporulation of at least one isolate of the acaropathogenic fungus Neozygites tanajoae Delalibera Jr. (Neozygitales: Neozygitaceae) (Hountondji et al. 2005). The use of generalist entomopathogenic fungi, such as B. bassiana and M. anisopliae, may also impact effectiveness of integrated pest management strategies through negative impacts on natural enemies. For instance, fungi used to control whiteflies (Bemisia tabaci Gennadius, Hemiptera: Aleyrodidae) have an antagonistic effect on beneficial predation by reducing the number of whiteflies eaten by lady beetles (Delphastus catalinae Horn,

Coleoptera: Coccinellidae), although fungi consumed should be far below the LC50 for lady beetles (Wang et al. 2005). The foliage-dwelling generalist predator Anthocoris nemorum L. (Hemiptera: Anthocoridae) is able to detect and avoid nettle leaves treated with B. bassiana when it directly contacts them (Meyling and Pell 2006). Low rates of field success for entomopathogenic fungi used to control pests may be a result of behavioral adaptations. Pilz et al. (2008) found up to 6,046 CFUs of fungi per gram of soil in Hungarian fields, but very low rates of WCR fungal infection, 1.4% of larvae and 0.05% of adults. Insects may use plant defenses to their own advantage, in a phenomenon termed pharmacophagy. By definition, “insects are pharmacophagous if they search for certain secondary plant substances directly, take them up, and utilize them for specific purpose other than primary metabolism or (merely) foodplant recognition” (Boppré 1984). Indeed, it has been suggested that the reason cucurbitacins, the bitter compounds found in Cucurbitaceae (Enslin 1954), repellent to most insects, are feeding stimulants to rootworms is that they inhibit the growth of entomopathogenic fungi. Tallamy et al. (1998) found that SCR transferred cucurbitacins transovarially and that cucurbitacin-containing eggs were less susceptible to Metarhizium anisopliae. Unsterile bitter Hawkesbury watermelon juice (Citrullus vulgaris Schrad) apparently contains a strain of Bacillus subtilis Ehrenberg

(Bacillales: Bacillaceae) bacteria that inhibits Metarhizium anisopliae and Beauveria bassiana, which may be why it is beneficial to SCR (Martin and Schroder 2000). Species of soil-borne microbes that form tight associations with the rhizosphere and promote plant growth have been widely researched with an eye to practical and market applications (see for example Varma et al. 1999), but few researchers have investigated insect behavioral response to these microbes (but see Villani et al. 1994, Fry et al. 1997, Thompson et al. 2007). A commercial soil additive consisting of a proprietary combination of microorganisms (EM-1) reduced WCR damage in Hungarian field corn as much as the chemical pesticide tefluthrin in a wet year (Németh et al. 2008). Few other peer-reviewed papers are available on this product, although it can be purchased easily online. Plant growth promoting organisms (PGPO) Plant growth promoting rhizobacteria, or PGPR, are beneficial soil bacteria (Kloepper et al. 1989). This term was first coined in 1978 when nine bacterial strains were shown to increase radish (Raphanus sativus L., Brassicales: Brassicaceae) growth (Kloepper and Schroth 1978). However, as other organisms such as fungus can also have similar functions, the inclusive term plant growth promoting organism (PGPO) is now being used (see, for example, van Gerrewey 2015). These PGPO can affect plant growth directly, such as by facilitating nutrient uptake or synthesizing beneficial compounds for the plant, or indirectly, such as by lessening or preventing the deleterious effects of phytopathogens (Glick 1995). Some, like the bacterial species Bacillus subtilis (Bhattacharya and Pramanik 2003) and the fungus Trichoderma harzianum Rifai (Hypocreales: Hypocreaceae) (Shakeri and Foster 2007), can also be entomopathogenic. Bacillus subtilis is known to kill the moth Diacrisia obliqua Walker (Lepidoptera: Arctiidae) (Bhattacharya and Pramanik 2003). However, B. subtilis is better known as a PGPO. Bacillus subtilis is known to contain lipopeptide antibiotics that inhibit plant pathogenic fungi (Yamada et al. 1990). Bacillus subtilis’s practical applications include preventing mummy berry disease, caused by Monilinia vaccinii-corymbosi Reade (Helotiales: Sclerotiniaceae) in blueberries and dieback, caused by Eutypa lata Pers. (Xylariales: Diatrypaceae) in grapevines (Ferreira et al. 1991) (Scherm et al. 2002) and

slowing the spread of brown rot, caused by Monilinia fructicola Winter (Helotiales: Sclerotiniaceae) of peaches (McKeen et al. 1986). Bacillus subtilis is sold as a plant disease control agent under the trade names Serenade (AgraQuest, Davis, CA), Plantacillin (Future Harvest, Kelowna, BC), and Companion (Growth Products, White Plains, NY). Trichoderma harzianum is also predominantly known as a PGPO (Harman et al. 2004a). Trichoderma spp. are capable of reproducing in soil, but have evolved as opportunistic plant symbionts (Harman et al. 2004a). Nearly all soil in the temperate and Neotropical regions contains between 101 and 103 CFUs per gram (Harman et al. 2004a). Inoculating maize seedlings with T. harzianum causes them to grow more quickly and reduces symptoms of disease including anthracnose, caused by Colletotrichum graminicola Politis (Glomerellales: Clomerellaceae) (Harman et al. 2004b). Trichoderma harzianum has some known entomopathogenic properties; it is known to kill mealworms (Tenebrio molitor L., Coleoptera: Tenebrionidae) (Shakeri and Foster 2007). In the interest of better understanding WCR microbe interactions, in chapter three we therefore assessed WCR larval responses to non-Bt and Bt corn seeds treated with commercially-available fungal and bacterial microbial single-species formulations of entomopathogens (B. bassiana, M. anisopliae, and S. marcescens) as well as PGPO (B. subtilis and T. harzianum). We followed laboratory behavioral response assays with greenhouse mesocosm experiments to assess any potential impact of microbial seed treatments on insect development, survival, weight, and sex ratio. Plant defense and plant-plant interactions Corn is known to use both above and below ground signaling; for instance, volatiles are more attractive to the parasitoid wasp Cotesia marginiventris Cresson (Hymenoptera: Braconidae) after corn has been treated with caterpillar (Spodoptera exigua Hübner, Lepidoptera: Noctuidae) regurgitant (Turlings and Tumlinson 1992), while the green leafy volatiles of corn plants are more attractive to the parasitoid wasp Cotesia sesamidae Cameron (Hymenoptera: Braconidae) after pest lepidoptera (Chilo partellus Swinhoe, Lepidoptera: Crambidae) have oviposited on it (Tamiru et al. 2012). These volatiles can be either attractive or repellent to pest insects; for instance, WCR are attracted to defensive corn

root volatiles (Johnson and Nielsen 2012), while corn leaf aphids (Rhopalosiphum maidis Fitch, Hemiptera: Aphididae) are repelled by volatiles emitted by damaged corn (Bernasconi et al. 1998). In the same manner, volatile cues are also important for mediating tri-trophic interactions below ground as well. For example, E-β-caryophyllene emitted by corn rootworm spp.-damaged roots attracts entomopathogenic nematodes (Rasmann et al. 2005). However, most commercial corn in North America has lost the ability to produce E-β- caryophyllene (Kollner et al. 2008). The fact that we cannot currently generate an effective response by WCR to synthetic E-β-caryophyllene in the laboratory indicates that there may be subtler cues occurring that have not yet been identified (Anbesse and Ehlers 2013). For instance, in another system, geijerines were shown to be the cues responsible for beneficial nematode (Steinernema diaprepesi Nguyen & Duncan, Rhabdita: Steinernematidae) recruitment on citrus (Citrus paradisi Macf. × Poncirus trifoliata L. Raf.) against root weevil (Diaprepes abbreviates L., Coleoptera: Curculionidae) pests (Ali et al. 2010). Interplant competition determines how well a plant will fare (Muller 1966). Corn, and in particular its root system, competes poorly with other plants such as weeds (Rajcan and Swanton 2001). Corn has about a 37% potential yield loss to weeds, approximately equal to that of soybeans but higher than many other major crops such as potatoes, barley, and wheat (Oerke and Dehne 2004). Interactions with other plants have mixed results; corn tends to suffer from delayed growth when intercropped with wheat (Triticum spp., Poales: Poaceae), for instance (Gao et al. 2014), although Li et al. (1999) noticed yield advantages when corn was intercropped with faba beans (Vicia faba L., Fabales: Fabaceae), likely because of the increased nitrogen content of the soil. Competition can cause reduced yield when it occurs before corn has reached the 14-leaf stage (Hall et al. 1992). Intraspecies competition is also a problem in corn (Andrade et al. 1993). Pagano and Maddoni (2007) found that tolerance to crowding varies with corn line and with environmental conditions. Faria et al. (2014) investigated a transgenic Bt-expressing corn hybrid (DKB 390 PRO 2) and found that it competed well with five common weed species. However, little other work has been done on transgenic corn competition.

Mounting defenses against natural enemies can be costly for many reasons: allocation of resources, autotoxicity, opportunity costs, or indirect costs which influence community interactions (Strauss et al. 2002). Defensive compounds can be costly for plants; therefore, releasing them selectively and only when an herbivore threat is imminent could provide a fitness benefit (Baldwin et al. 2006). Volatiles released from damaged plants can communicate the presence of herbivores to neighboring plants, allowing these plants to activate their defenses (Dicke et al. 1990). Communication is known to occur in plants, including communication that is conditional on the kin status of neighboring plants (Poelman 2013). This communication can take place both above and below ground (Baldwin et al. 2002), via volatile organic compounds. Volatile organic compounds (VOCs) are defined by the EPA as any compound of carbon which participates in atmospheric photochemical reactions, with exceptions for a few compounds such as metallic carbides or carbon dioxide (EPA 1986). Volatile organic compounds emitted by plants can mediate the transmission of information from plants to pollinators and other organisms that are mutualistic with plants, as well as transmitting information to other plants (Paschold et al. 2006). In the above-ground sphere, VOCs emitted by beetle-damaged (Gynandrobrotica guerreroensis Jacoby and Cerotoma ruficornis Horn, Coleoptera: Chrysomelidae) lima bean plants (Phaseolus lunatus L., Fabales: Fabaceae) caused undamaged lima bean plants to secrete extrafloral nectaries as a defense (Heil and Bueno 2006). Sometimes, interplant signals that warn plants of imminent herbivores are referred to as "eavesdropping", because the benefit to the receiver is very clear, while the benefit to the sender is not (Baldwin et al. 2002). Below-ground communication is believed to take place primarily through root exudates. Root exudates are understudied but have many proposed uses, including not only communication between plants, microbes, and insects, but also alteration of the characteristics of surrounding soil (Walker et al. 2003). Interplant communication can convey specific and detailed information; for example, Semchenko et al. (2014) showed that tussock grass, Deschampsia caespitosa L. (Poales: Poaceae), could determine kinship status from root exudate and use this information to limit selfish resource

allocation to root production when neighboring plants were siblings. Zakir et al. (2013) noted that pest Spodoptera littoralis Boisduval (Lepidoptera: Noctuidae) was less likely to oviposit on a plant if its neighbor was a damaged cotton plant (Gossypium hirsutum L., Malvales: Malvaceae). They found that cotton was the best signaling plant out of several plants they tested, regardless of the species of the receiving plant. The unavoidable difficulties of experimental design with respect to root plant-plant communication (Barto et al. 2012) has left many questions about root communication unanswered, but some researchers believe that common mycorrhizal networks are responsible for facilitating these interactions (Barto et al. 2012). Allelopathy can also occur in aerial tissues in some plants such as Rapanea umbellata Martius (Ericales: Primulaceae) (see, for example, Novaes et al. 2013) In addition to beneficial interactions, chemical inhibition, or allelopathy (Muller 1966), also exists in plants. For example, the invasive weed Centaurea maculosa Lam. (Asterales: Asteraceae) uses allelopathic root exudates to inhibit native grasses (Ridenour and Callaway 2001). Extracts of Mexican sunflower (Tithonia diversifolia Gray, Asterales: Asteraceae) have an inhibitory effect on the growth of young corn plants (Oyerinde et al. 2009), though not on older plants. Autotoxicity, a kind of allelopathy in which allelochemicals from crop residue have a negative effect on the next season’s crop of the same species, is a problem in some crops, including corn (Singh et al. 2010). Interplant signals, whether beneficial or allelopathic, can change plants’ resource allocations (see, for example, Baldwin et al. 2002, Oyerinde et al. 2009). Above-ground and below-ground infochemicals that can mediate such interactions have been identified in corn (see, for example, Baldwin et al. 2002). Levels of Bt toxin expression are important to the efficacy and sustained use of Bt corn to manage insect pests. As we have discussed above, environmental conditions can alter the expression of Bt toxin (see for example Hibbard et al. 2009). We have also discussed that neighboring plants can affect plants’ gene expression. However, there has been little attention paid to how corn plants’ neighbors may affect Bt toxin expression.

For all these reasons, in Chapter 4 we investigated potential effects of interactions between corn plants on the expression of Bt toxin in transgenic corn roots. We believe this work will be a novel research approach to an understudied but important aspect of transgenic plant ecology with potential impact on WCR management.

References

Ali, J., Alborn, H., and Stelinski, L. 2010. Subterranean herbivore-induced volatiles released by citrus roots upon feeding by Diaprepes abbreviates recruit entomopathogenic nematodes. Journal of Chemical Ecology 36(4): 361-368. Anbesse, S., and R. Ehlers. 2013. Heterorhabditis sp. not attracted to synthetic (E)-beta- caryophyllene, a volatile emitted by roots upon feeding by corn rootworm. Journal of Applied Entomology 137(1-2): 88-96. Andrade, F., S. Uhart, and M. Frugone. 1993. Intercepted radiation at flowering and kernel number in maize- shade versus plant-density effects. Crop Science 33(3): 482-485. Assabgui, R., R. Hamilton, and J. Arnason. 1995. Hydroxamic acid content and plant development of maize (Zea mays L.) in relation to damage by the western corn rootworm, Diabrotica virgifera virgifera LeConte. Canadian Journal of Plant Science 75(4): 851-856. Baldwin, I., R. Halitschke, A. Paschold, C. von Dahl, and C. Preston. 2006. Volatile signaling in plant-plant interactions: “talking trees” in the genomics era. Science 311(5762): 812-815. Baldwin, I., A. Kessler, and R. Halitschke. 2002. Volatile signaling in plant-plant-herbivore interactions: What is real? Current Opinion in Plant Biology 5(4): 351-354. Ball, H., and G. Weekman. 1962. Insecticide resistance in adult western corn rootworm in Nebraska. Journal of Economic Entomology 55(4): 439-441. Barto, E., J. Weidenhamer, D. Cipollini, and M. Rillig. 2012. Fungal superhighways: do common mycorrhizal networks enhance below ground communication? Trends in Plant Science 17(11): 633-637. Bates, S., J. Zhao, R. Roush, and A. Shelton. 2005. Insect resistance management in GM crops: past, present, and future. Nature Biotechnology 23(1): 57-62. Baufeld, P., and S. Enzian. 2001. Simulations model for spreading scenarios of western corn rootworm (Diabrotica virgifera virgifera) in case of Germany. Proceedings of the XXI IWGO Conference, Legnaro, Italy.

Baum, J., and T. Malvar. 1995. Regulation of insecticidal crystal protein production in Bacillus thuringiensis. Molecular Microbiology 18(1): 1-12. Baum, J., T. Bogaert, W. Clinton, G. Heck, P. Feldmann, O. Ilagan, S. Johnson, G. Plaetinck, T. Munyikwa, M. Pleau, T. Vaughn, and J. Roberts. 2007. Control of coleopteran insect pests through RNA interference. Nature Biotechnology 25(11): 732-737. Baverstock, J., H. Roy, and J. Pell. 2010. Entomopathogenic fungi and insect behaviour: from unsuspecting hosts to targeted vectors. BioControl 55(1): 89-102. Bechtel, D., and L. Bulla. Electron-microscope study of sporulation and parasporal crystal- formation in Bacillus-thuringiensis. Journal of Bacteriology 127(3): 1472-1481. Beegle, C., and T. Yamamoto. 1992. Invitation paper (C.P. Alexander Fund): History of Bacillus thuringiensis Berliner research and development. The Canadian Entomologist 124(4): 587-616, Berdegué, M., and J. Trumble. 1997. Interaction between linear furanocoumarins found in celery and a commercial Bacillus thuringiensis formulation on Spodoptera exigua (Lepidoptera: Noctuidae) larval feeding behavior. Journal of Economic Entomology 90(4): 961-966. Bernasconi, M., T. Turlings, L. Ambrosetti, P. Bassetti, and S. Dorn. 1998. Herbivore- induced emissions of maize volatiles repel the corn leaf aphid, Rhopalosiphum maidis. Entomologia Experimentalis et Applicata 87(2): 133-142. Bernklau, E., B. Hibbard, and L. Bjostad. 2010. Antixenosis in maize reduces feeding by Western corn rootworm larvae (Coleoptera: Chrysomelidae). Journal of Economic Entomology 103(6): 2052-2060. Bhattacharya, S., and A. Pramanik. 2003. In vitro sensitivity and bio-efficacy of Bacillus subtilis as entomopathogen against lepidopteran larvae, Diacrisia obliqua. Crop Research (Hisar): 25 (1): 119-126. Bing, L., and L. Lewis. 1991. Suppression of Ostrinia nubilalis (Hübner) (Lepidoptera: Pyralidae) by endophytic Beauveria bassiana (Balsamo) Vuillemin. Environmental Entomology 20(4): 1207-1211.

Bing, L., and L. Lewis. 1993. Occurrence of the entomopathogen Beauveria bassiana (Balsamo) Vuillemin in different tillage regimes and in Zea mays L. and virulence towards Ostrinia nubilalis (Hübner). Agriculture, Ecosystems, and Environment 45(1-2): 147-156. Binning, R., S. Lefko, A. Millsap, S. Thompson, and T. Nowatzki. 2010. Estimating western corn rootworm (Coleoptera: Chrysomelidae) larval susceptibility to event DAS- 59122-7 maize. Journal of Applied Entomology 134(7): 551-561. Bizzarri, M., and A. Bishop. 2007. The ecology of Bacillus thuringiensis on the phylloplane: Colonization from soil, plasmid transfer, and interaction with larvae of Pieris brassicae. Microbial Ecology 56(1): 133-139. Boppré, M. 1984. Redefining “pharmacophagy”. Journal of Chemical Ecology 10(7): 1151-1154. Bourguet, D. 2004. Resistance to Bacillus thuringiensis toxins in the European corn borer: what chance for Bt maize? Physiological Entomology 29(3): 251-256. Branson, T., and J. Krysan. 1981. Feeding and oviposition behavior and life-cycle strategies of Diabrotica- an evolutionary view with implications for pest-management (Coleoptera, Chrysomelidae). Environmental Entomology 10(6): 826-831. Branson, T., V. Welch, G. Sutter, and J. Fisher. 1983. Resistance to larvae of Diabrotica virgifera virgifera in three experimental maize hybrids. Environmental Entomology 12(5): 1509-1512. Broderick, N., C. Robinson, M. McMahon, J. Holt, J. Handelsman, and K. Raffa. 2009. Contributions of gut bacteria to Bacillus thuringiensis-induced mortality vary across a range of Lepidoptera. BMC Biology 7(11). Brookes, G., and P. Barfoot. 2010. Global impact of biotech crops: Environmental effects, 1996-2008. AgBioForum 13(1). Available at: http://www.agbioforum.org/v13n1/v13n1a06- brookes.htm Bruck, D., and L. Lewis. 2002. Whorl and pollen-shed stage application of Beauveria bassiana for suppression of adult western corn rootworm. Entomologia Experimentalis et Applicata 103(2): 161-169.

Bruck, D., and L. Lewis. 2001. Adult Diabrotica spp. (Coleoptera: Chrysomelidae) infection at emergence with indigenous Beauveria bassiana (Deuteromycotina: Hyphomycetes). Journal of Invertebrate Pathology 77(4): 288-289. Bryson, H., D. Wilbur, and C. Burkhardt. 1953. The western corn rootworm, Diabrotica virgifera Lec. in Kansas. Journal of Economic Entomology 46(6): 995-999. Cannon, R. 1993. Prospects and progress for Bacillus thuringiensis-based pesticides. Pesticide Science 37(4): 331-335. Carle, P., R. Whitcomb, K. Hackett, J. Tully, D. Rose, J. Bove, R. Henegar, M. Konai, and D. Williamson. 1997. Spiroplasma diabroticae sp nov, from the Southern corn rootworm beetle, Diabrotica undecimpunctata (Coleoptera: Chrysomelidae). International Journal of Systematic Bacteriology 47(1): 78-80. Carozzi, N., and M. Koziel, eds. 1997. Advances in Insect Control: The Role of Transgenic Plants. London: Taylor & Francis. Chilcutt, C., and B. Tabashnik. 2004. Contamination of refuges by Bacillus thuringiensis toxin genes from transgenic maize. Proceedings of the National Academy of Sciences of the United States of America 101(20): 7526-7529. Choudhary, B., and K. Gaur. 2009. The development and regulation of Bt brinjal in India. ISAAA Brief No. 38, International Service for Acquisition of Agri-Biotech Applications, Ithaca, NY. Chu, C., J. Spencer, M. Curzi, J. Zavala, and M. Seufferheld. 2013. Gut bacteria facilitate adaptation to crop rotation in the western corn rootworm. Proceedings of the National Academy of Sciences 110(29): 11917-11922. Ciosi, M., S. Toepfer, H. Li, et al. 2009. European populations of Diabrotica virgifera virgifera are resistant to aldrin, but not to methyl-parathion. Journal of Applied Entomology 133(4): 307-314. Clark, P., T. Vaughn, L. Meinke, J. Molina-Ochoa, and J. Foster. 2006. Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) larval feeding behavior on transgenic maize (MON 863) and its isoline. Journal of Economic Entomology 99(3): 722-727.

Cory, J., and K. Hoover. 2006. Plant-mediated effects in insect-pathogen interactions. Trends in Ecology and Evolution 21(5): 278-286. Cullen E, Gray M, Gassmann A, Hibbard B (2013) Resistance to Bt corn by western corn rootworm (Coleoptera: Chrysomelidae) in the U.S. Corn Belt. Journal of Integrated Pest Management 4(3):D1-D6. Cuong, N., and M. Cohen. 2003. Mating and dispersal behavior of Scirpophaga incertulas and Chilo suppressalis (Lepidoptera: Pyralidae) in relation to resistance management for rice transformed with Bacillus thuringiensis toxin genes. International Journal of Pest Management 49(4): 275-279. Damgaard, P., B. Hansen, J. Pedersen, and J. Eilenberg. 1997. Natural occurrence of Bacillus thuringiensis on cabbage foliage and in insects associated with cabbage crops. Journal of Applied Microbiology 82(2): 253-258. Dangar, T., Y. Babu, and J. Das. 2010. Population dynamics of soil microbes and diversity of Bacillus thuringiensis in agricultural and botanic garden soils of India. African Journal of Biotechnology 9(4): 496-501. Denholm, I., and M. Rowland. 1992. Tactics for managing pesticide resistance in arthropods- theory and practice. Annual Review of Entomology 37: 91-112. Dicke, M., M. Sabelis, J. Takabayashi, J. Bruin, and M. Posthumus. 1990. Plant strategies of manipulating predator-prey interactions through allelochemicals- prospects for application in pest-control. 1990. Journal of Chemical Ecology 16(11): 3091-3118. Doberski, J., and H. Tribe. 1980. Isolation of entomogenous fungi from elm bark and soil with reference to ecology of Beauveria bassiana and Metarhizium anisopliae. Transactions of the British Mycological Society 74(1): 95-100. Dover, M. 1985. Getting off the pesticide treadmill. Technology Review 88: 52-63. Dowd-Uribe, B. 2014. Engineering yields and inequality? How institutions and agro- ecology shape Bt cotton outcomes in Burkina Faso. Geoforum 53: 161-171. El Khishen, A., M. Bohn, D. Prischmann-Voldseth, K. Dashiell, B. French, and B. Hibbard. 2009. Native resistance to western corn rootworm (Coleoptera: Chrysomelidae) larval

feeding: characterization and mechanisms. Journal of Economic Entomology 102(6): 2350- 2359. Elliot, N., J. Jackson, G. Sutter, and D. Beck. 1991. A descriptive study of the population dynamics of adult Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) in artificially infested corn fields. Great Lakes Entomologist 24(3): 159-167. Elliot, S., M. Sabelis, A. Janssen, van der Geest, Beerling, and Fransen. 2000. Can plants use entomopathogens as bodyguards? Ecology Letters 3(3): 228-235. Enslin, P. 1954. Bitter principles of the Cucurbitaceae. I- Observations on the chemistry of Cucurbitacin-A. Journal of the Science of Food and Agriculture 5(9): 410-416. [EPA] Environmental Protection Agency. 1986. Protection of environment: Requirements for preparation, adoption, and submittal of implementation plans- procedural requirements. Electronic Code of Federal Regulations 40(51.100). Available from http://www.ecfr.gov/cgi-bin/text-idx?rgn=div8&node=40:2.0.1.1.2.3.8.1 [EPA] Environmental Protection Agency. 1998. Transmittal of the final report of the FIFRA Scientific Advisory Panel Subpanel on Bacillus thuringiensis (Bt) plant-pesticides and resistance management, meeting held on February 9 an 10, 1998. Available from http://www.epa.gov/scipoly/sap/meetings/1998/february/finalfeb.pdf [EPA] Environmental Protection Agency. 2001. Bt plant-incorporated protectants biopesticides registration action document. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_brad2/4-irm.pdf [EPA] Environmental Protection Agency. 2002. Transmittal of meeting minutes of the FIFRA Scientific Advisory Panel meeting held August 27-29, 2002. [EPA] Environmental Protection Agency. 2005. EPA annual pesticide report. Available from [EPA] Environmental Protection Agency. 2010. Introduction to biotechnology regulation for pesticides. Available from http://www.epa.gov/pesticides/biopesticides/regtools/biotech- reg-prod.htm#crops [EPA] Environmental Protection Agency (2010) Scientific Issues Related to Insect Resistance Management for SmartStax™ Refuge-in-the-Bag, a Plant-Incorporated Protectant

(PIP) Corn Seed Blend. Available from http://www.epa.gov/scipoly/sap/meetings/2010/december/120810agenda.pdf [EPA] Environmental Protection Agency (2014) Insect resistance management fact sheet for Bacillus thuringiensis (Bt) corn products. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_corn_refuge_2006.htm [EPA] Environmental Protection Agency (2014) White Paper on Corn Rootworm Resistance Monitoring for Bt Plant-Incorporated Protectants. Available from http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPP-2013-0490-0008 [EPA] Environmental Protection Agency (2014) Insect resistance management fact sheet for Bacillus thuringiensis (Bt) cotton products. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_cotton_refuge_2006.htm [EPPO] European and Mediterranean Plant Protection Organization. 2011. Diabrotica virgifera virgifera. Available from http://www.eppo.int/QUARANTINE/special_topics/Diabrotica_virgifera/diabrotica_virgifera .htm Estruch, J., G. Warren, M. Mullins, G. Nye, J. Craig, and M. Koziel. 1996. Vip3A, a novel Bacillus thuringiensis vegetative insecticidal protein with a wide spectrum of activities against lepidopteran insects. Proceedings of the National Academy of Sciences 93:5389- 5394. Faria, R., R. Barros, and L. Santos. Weed interference on growth and yield of transgenic maize. Planta Daninha 32(3): 515-520. Fernandez-Cornejo, Jorge, Seth Wechsler, Mike Livingston, and Lorraine Mitchell. Genetically Engineered Crops in the United States, ERR-162 U.S. Department of Agriculture, Economic Research Service, February 2014. Fernandez-Cornejo, J., and W. McBride. 2000. Genetically engineered crops for pest management in US agriculture: farm-level effects. USDA. Available from http://ddr.nal.usda.gov/bitstream/10113/38264/1/CAT11079167.pdf Ferreira, J., F. Matthee, and A. Thomas. 1991. Biological control of Eutypa lata on grapevine by an antagonistic strain of Bacillus subtilis. Phytopathology 81(3): 283-287.

Ferro, D., and S. Lyon. 1991. Colorado potato beetle (Coleoptera: Chrysomelidae) larval mortality: operative effects of Bacillus thuringiensis subsp. san diego. Journal of Economic Entomology 84(3): 806-809. Fisher, R., and L. Rosner. 1959. Toxicology of the microbial insecticide, Thuricide. Agricultural and Food Chemistry 7(10): 686-688. Fry, R. 1997. Radiographic study of the response of Japanese beetle larvae (Coleoptera: Scarabaeidae) to soil-incorporated mycelia particles of Metarhizium anisopliae (Deuteromycetes). Journal of the New York Entomological Society 105(1-2): 113-120. Gao, Y., P. Wu, X. Zhao, and Z. Wang. 2014. Growth, yield, and nitrogen use in the wheat/maize intercropping system in an arid region of northwestern China. Field Crops Research 167: 19-30. Gassmann, A., J. Petzold-Maxwell, R. Keweshan, and M. Dunbar. 2011. Field-evolved resistance to Bt maize by Western corn rootworm. PLoS one 6(7): e22629. Gassmann, A., and P. Weber. 2012. Evaluation of corn rootworm management practices in northeast Iowa. Iowa State Research Farm Progress Reports. Paper 1945. Gill, S., E. Cowles, and P. Pietrantonio. 1992. The mode of action of Bacillus thuringiensis endotoxins. Annual Review of Entomology 37: 615-36. Gillette, C. 1912. Diabrotica virgifera Leconte as a corn root-worm. Journal of Economic Entomology 5(4): 364-366. Giordano, R., J. Jackson, and H. Robertson. 1997. The role of Wolbachia bacteria in reproductive incompatibilities and hybrid zones of Diabrotica beetles and Gryllus crickets. Proceedings of the National Academy of Sciences of the United States of America 94(21): 11439-11444. Glick, B. 1995. The enhancement of plant-growth by free-living bacteria. Canadian Journal of Microbiology 41(2): 109-117. Goldberg, L., and J. Margalit. 1977. A bacterial spore demonstrating rapid larvicidal activity against Anopheles sergentii, Uranotaenia unguiculata, Culex univitattus, Aedes aegypti, and Culex pipiens. Mosquito News 37(3): 355-358.

Gould, F. 1998. Sustainability of transgenic insecticidal cultivars: Integrating pest genetics and ecology. Annual Review of Entomology 43: 701-726. Gray, M. 2011. Relevance of traditional integrated pest management (IPM) strategies for commercial corn producers in a transgenic agroecosystem: A bygone era? Journal of Agricultural and Food Chemistry 59(11): 5852-5858. Gray, M., T. Sappington, N. Miller, J. Moeser, and M. Bohn. 2009. Adaptation and invasiveness of western corn rootworm: intensifying research on a worsening pest. Annual Review of Entomology 54: 303-321. Gray, M., K. Steffey, R. Estes, and J. Schroeder. 2007. Responses of transgenic maize hybrids to variant western corn rootworm larval injury. Journal of Applied Entomology 131(6): 386-390. Guinebretiere, M., F. Thompson, A. Sorokin, P. Normand, P. Dawyndt, M. Ehling-Schulz, B. Svensson, V. Sanchis, C. Nguyen-The, M. Heyndrickx, and P. de Vos. 2008. Ecological diversification in the Bacillus cereus group. Environmental Microbiology 10(4): 851-865. Gustin, R. 1981. Soil temperature environment of overwintering Western corn rootworm eggs. Environmental Entomology 10(4): 483-487. Gustin, R., and T. Schumacher. 1989. Relationship of some soil pore parameters to movement of first-instar Western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 18(3): 343-346. Hall, M., C. Swanton, and G. Anderson. 1992. The critical period of weed control in grain corn (Zea mays). Weed Science 40(3): 441-447. Harman, G., C. Howell, A. Viterbo, I. Chet, and M. Lorito. 2004. Trichoderma species- opportunistic, avirulent plant symbionts. Nature Reviews Microbiology 2(1): 43-56. Harman, G., R. Petzoldt, A. Comis, and J. Chen. 2004. Interactions between Trichoderma harzianum strain T22 and maize inbred line Mo17 and the effects of these interactions on diseases caused by Pythium ultimum and Colletotrichum graminicola. Phytopathology 94(2): 147-153.

Head, G., and J. Greenplate. 2012. The design and implementation of insect resistance management programs for Bt crops. GM Crops and Food: Biotechnology in Agriculture and the Food Chain 3(3): 144-153. Heil, M., and J. Bueno. 2006. Within-plant signaling by volatiles leads to induction and priming of an indirect plant defense in nature. Proceedings of the National Academy of Sciences of the United States of America 104(13): 5467-5472. Hein, G., and J. Tollefson. 1985. Seasonal oviposition of northern and western corn rootworms (Coleoptera: Chrysomelidae) in continuous cornfields. Journal of Economic Entomology 78(6): 1238-1241. Hibbard, B., A. Khishen, and T. Vaughn. 2009. Impact of MON863 transgenic roots is equivalent on western corn rootworm larvae for a wide range of maize phenologies. Journal of Economic Entomology 102(4): 1607-1613. Hill, R. 1975. Mating, oviposition patterns, fecundity and longevity of western corn rootworm. Journal of Economic Entomology 68(3): 311-315. Hofte, H., and H. Whiteley. Insecticidal crystal proteins of Bacillus-thuringiensis. Microbiological Reviews 53(2): 242-255. Hossain, F., C. Pray, Y. Lu, J. Huang, C. Fan, and R. Hu. 2004. Genetically modified cotton and farmers’ health in China. International Journal of Occupational and Environmental Healt 10(3): 296-303. Hountondji, F., M. Sabelis, R. Hanna, and A. Janssen. 2005. Herbivore-induced plant volatiles trigger sporulation in entomopathogenic fungi: the case of Neozygites tanajoae infecting the cassava green mite. Journal of Chemical Ecology 31(5): 1003-1021. Huber, H., P. Lüthy, H. Ebersold, and J. Cordier. 1981. The subunits of the parasporal crystal of Bacillus thuringiensis: Size, linkage, and toxicity. Archives of Microbiology 129(1): 14-18. Hutchison, W., E. Burkness, P. Mitchell, R. Moon, T. Leslie, S. Fleischer, M. Abrahamson, K. Hamilton, K. Steffey, M. Gray, R. Hellmich, L. Kaster, T. Hunt, R. Wright, K. Pecinovsky, T. Rabaey, B. Flood, and E. Raun. 2010. Areawide suppression of European

corn borer with Bt maize reaps savings to non-Bt maize growers. Science 330(6001): 222- 225. [IRAC] Insecticide Resistance Action Committee Brazil. 2015. Insecticide resistance & pest management recommendations in Brazilian soybean, cotton, and corn. IRAC Brazil and IRAC International. Jaffe, G. 2009. Complacency on the farm. Center for Science in the Public Interest. Available from http://cspinet.org/new/pdf/complacencyonthefarm.pdf Johnson, S., and U. Nielsen. 2012. Foraging in the dark- chemically mediated host plant location by belowground insect herbivores. Journal of Chemical Ecology 38(6): 604-614. Iriarte, J., Y. Bel, M. Ferrandis, R. Andrew, J. Murillo, J. Ferré, and P. Caballero. 1998. Environmental distribution and diversity of Bacillus thuringiensis in Spain. Systematic and Applied Microbiology 21(1): 97-106. Ismael, Y., R. Bennett, and S. Morse. Farm-level economic impact of biotechnology: smallholder Bt cotton farmers in South Africa. Outlook on Agriculture 31(2): 107-111. Kabaluk, J., and J. Ericsson. Environmental and behavioral constraints on the infection of wireworms by Metarhizium anisopliae. Environmental Entomology 36(6): 1415-1420. Kang, J., and Krupke, C. 2009. Likelihood of multiple mating in Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). Journal of Economic Entomology 102(6): 2096- 2100. Kepler, R., and D. Bruck. 2006. Examination of the interaction between the black vine weevil (Coleoptera: Curculionidae) and an entomopathogenic fungus reveals a new tritrophic interaction. Environmental Entomology 35(4): 1021-1029. Kloepper, J., R. Lifshitz, and R. Zablotowicz. 1989. Free-living bacterial inocula for enhancing crop productivity. Trends in Biotechnology 7(2): 39-44. Kloepper, J., and M. Schroth. 1978. Plant growth promoting rhizobacteria on radishes. Proceedings of the IVth International Conference on Plant Pathogenic Bacteria, Angers, August 27- September 2, 1978. 2: 879-882. Knolhoff, L., D. Onstad, J. Spencer, and E. Levine. 2006. Environmental Entomology 35(4): 1049-1057.

Kollner, T., M. Held, C. Lenk, I. Hiltpold, T. Turlings, J. Gershenzon, and J. Degenhardt. 2008. A maize (E)-beta-caryophyllene synthase implicated in indirect defense responses against herbivores is not expressed in most American maize varieties. Plant Cell 20(2): 482- 494. Kruger, M., J. van Rensburg, and J. van den Berg. 2009. Perspective on the development of stem borer resistance to Bt maize and refuge compliance at the Vaalharts irrigation scheme in South Africa. Crop Protection 28(8): 684-689. Krueger, S., and D. Roberts. 1997. Soil treatment with entomopathogenic fungi for corn rootworm (Diabrotica spp.) larval control. Biological Control 9(1): 67-74. Krupke, C., P. Marquardt, W. Johnson, S.Weller, and S. Conle. 2009. Volunteer corn presents new challenges for insect resistance management. Agronomy Journal 101(4): 797- 799. Krysan, J., J. Jackson, and A. Lew. 1984. Field termination of egg diapause in Diabrotica with new evidence of extended diapauses in D. barberi (Coleoptera: Chrysomelidae). Environmental Entomology 13(5): 1237-1240. LeConte, J. 1868. New Coleoptera collected on the survey for the extension of the Union Pacific Railway, E.D. from Kansas to Fort Craig, New Mexico. Transactions of the American Entomological Society 2: 49-59. Lefebvre, C. 1934. Penetration and development of the fungus, Beauveria Bassiana, in the tissues of the corn borer. Annals of Botany 48(2): 441-452. Levine, E., J. L. Spencer, S. A. Isard, D. W. Onstad, and M. E. Gray. 2002. Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomologist 48(2): 94-107. Li, H., S. Toepfer, and U. Kuhlmann. Relationship between phenotypic traits and selected fitness components of Diabrotica virgifera virgifera. Entomologia Experimentalis et Applicata 131(3), 254-263. 2009. Li, L., S. Yang., X. Li, F. Zhang, and P. Christie. 1999. Interspecific complementary and competitive interactions between intercropped maize and faba vean. Plant and Soil 212(2): 105-114.

Liu, F., Z. Xu, J. Chang, J. Chen, F. Meng, Y. Zhu, and J. Shen. 2008. Resistance allele frequency to Bt cotton in field populations of Helicoverpa armígera (Lepidoptera: Noctuidae) in China. Journal of Economic Entomology 101(3): 933-943. Lupwayi, N., W. Rice, and G. Clayton. 1998. Soil microbial diversity and community structure under wheat as influenced by tillage and crop rotation. Soil Biology and Biochemistry 30(13): 1733-1741. Mallet, J., and P. Porter. 1992. Preventing insect adaptation to insect-resistant crops: are seed mixtures or refugia the best strategy? Proceedings of the Royal Society, Biological Sciences 250(1328): 165-169. Marquardt, P., C. Krupke, J. Camberato, and W. Johnson. 2014. The effect of nitrogen rate on transgenic corn Cry3Bb1 protein expression. Pest Management Science 70(5): 763-770. Marquardt, P., and C. Krupke. 2009. Dispersal and mating behavior of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) in Bt cornfields. Environmental Entomology 38(1): 176-182. Martin, P., E. Hirose, and J. Aldrich. 2007. Toxicity of Chromobacterium subtsugae to southern green stink bug (Heteroptera: Pentatomidae) and corn rootworm (Coleoptera: Chrysomelidae). Journal of Economic Entomology 100(3): 680-684. Martin, P., and R. Schroder. 2000. The effect of cucurbitacin E glycoside, a feeding stimulant for corn rootworm, on biocontrol fungi: Beauveria bassiana and Metarhizium anisopliae. Biocontrol Science and Technology 10(3): 315-320. Matten, S., G. Head, and H. Quemada. 2008. in J. Romeis, A. Shelton, and G. Kennedy, [eds.], Integration of Insect-Resistant Genetically Modified Crops within IPM Programs, Springer, New York, 27-39. McAfee, K. 2008. Beyond techno-science: transgenic maize in the fight over Mexico’s future. Environmental Economic Geography 39(1): 148-160. McGill, W., K. Cannon, J. Robertson, and F. Cook. 1986. Dynamics of soil microbial biomass and water-soluble organic C in Breton L after 50 years of cropping to two rotations. Canadian Journal of Soil Science 66(1): 1-19.

McKeen, C., C. Reilly, and P. Pusey. 1986. Production and partial characterization of antifungal substances antagonistic to Monilinia fructicola from Bacillus subtilis. Phytopathology 76(2): 136-139. Meihls, L., M. Higdon, B. Siegfried, N. Miller, T. Sappington, M. Ellersieck, T. Spencer, and B. Hibbard. 2008. Increased survival of western corn rootworm on transgenic corn within three generations of on-plant greenhouse selection. Proceedings of the National Academy of Sciences of the United States of America 105(49): 19177-19182. Meinke, L., B. Siegfried, R. Wright, and L. Chandler. 1998. Adult susceptibility of Nebraska western corn rootworm (Coleoptera: Chrysomelidae) populations to selected insecticides. Journal of Economic Entomology 91(3): 594-600. Meissle, M., C. Pilz, and J. Romeis. 2009. Susceptibility of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) to the entomopathogenic fungus Metarhizium anisopliae when feeding on Bacillus thuringiensis Cry3Bb1-expressing maize. Applied and Environmental Microbiology 75(12): 3937-3943. Melhus, I., R. Painter, and F. Smith. 1954. A search for resistance to the injury caused by species of Diabrotica in the corns of Guatemala. Iowa State College Journal of Science 29(1): 75-94. Metcalf, RJ. in JL Krysan and TA Miller [eds.], Methods for the study of pest Diabrotica. Springer, New York. pp. vii-xv. 1986. Meyling, N., and J. Pell. 2006. Detection and avoidance of an entomopathogenic fungus by a generalist insect predator. Ecological Entomology 31(2): 162-171. Miller, N., A. Estoup, S. Toepfer, D. Bourguet, L. Lapchin, S. Derridj, K.S. Kim, P. Reynaud, L. Furlan, and T. Guillemaud. 2005. Multiple transatlantic introductions of the western corn rootworm. Science 310(5750): 992. Moeser, J., and B. Hibbard. in S. Vidal, U. Kuhlmann, and R. Edwards [eds.], Western corn rootworm: ecology and management. CABI Publishers, Wallingford, UK. pp. 41-65. 2005. Moeser, J., and S. Vidal. 2005. Nutritional resources used by the invasive maize pest Diabrotica virgifera virgifera in its new south-east-European distribution range. Entomologia Experimentalis et Applicata 114(1): 55-63.

Monnerat, R., C. Soares, G. Capdeville, G. Jones, E. Martins, L. Praҫa, B. Cordeiro, S. Braz, R. Dos Santos, and C. Berry. 2009. Translocation and insecticidal activity of Bacillus thuringiensis living inside of plants. Microbial Biotechnology 2(4): 512-520. Morse, S., and A. Mannion. 2009. Can genetically modified cotton contribute to sustainable development in Africa? Progress in Development Studies 9(3): 225-247. Muller, C. Role of chemical inhibition (allelopathy) in vegetational composition. Bulletin of the Torrey Botanical Club 93(5): 332-351. Mulock, B., and L. Chandler. 2001. Effect of Beauveria bassiana on the fecundity of western corn rootworm, Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). Biological Control 22(1): 16-21. Murphy, A., M. Ginzel, and C. Krupke. 2010. Evaluating western corn rootworm (Coleoptera: Chrysomelidae) emergence and root damage in a seed mix refuge. Journal of Economic Entomology 103(1): 147-157. Naranjo, S., and S. Steinkraus. 1988. Discovery of an entomophthoralean fungus (Zygomycetes: Entomophthorales) infecting northern corn rootworm, Diabrotica barberi (Coleoptera: Chrysomelidae). Journal of Invertebrate Pathology 51(3): 298-300. Németh, T., M. Nádasy, Z. Marczali, F. Simon, E. Nádasyné, and J. Takács. 2008. Effect of a soil microbe preparation (EM-1) on the damage of western corn rootworm (Diabrotica virgifera virgifera LeConte). Proceedings of the VII. Alps-Adria Scientific Workshop, 28 April-2 May, 2008, Stara Lesna, Slovakia 36(5): 843-846. Novaes, P., Imatomi, M., Miranda, M., and Gualtieri, S. 2013. Phytotoxicity of leaf aqueous extract of Rapanea umbellata (Mart.) Mez (Primulaceae) on weeds. Acta Scientarum- Agronomy 35(2): 231-239. Oerke, E., and H. Dehne. 2004. Safeguarding production- losses in major crops and the role of crop protection. Crop Protection 23(4): 275-285. Oi, D., and R. Pereira. 1993. Ant behavior and microbial pathogens (Hymenoptera: Formicidae). The Florida Entomologist 76(1): 63-74. Okeke, I., and R. Edelman. 2001. Dissemination of antibiotic-resistant bacteria across geographic borders. Clinical Infectious Diseases 33(3): 364-369.

Onstad, D., J. Spencer, C. Guse, E. Levine, and S. Isard. Modeling evolution of behavioral resistance by an insect to crop rotation. Entomologia Experimentalis et Applicata 100(2): 195-201. O’Rourke, M., and L. Jones. 2011. Analysis of landscape-scale insect pest dynamics and pesticide use: an empirical and modeling study. Ecological Applications 21(8): 3199-3210. Ouedraogo, R., M. Cusson, M. Goettel, and J. Brodeur. 2003. Inhibition of fungal growth in thermoregulating locusts, Locusta migratoria, infected by the fungus Metarhizium anisopliae var acridum. Journal of Invertebrate Pathology 82(2): 103-109. Oyediran, I., B. Hibbard, and T. Clark. 2004. Prairie grasses as hosts of the western corn rootworm (Coleoptera: Chrysomelidae). Environmental Entomology 33(3): 740-747. Oyerinde, R., O. Otusanya, and O. Akpor. 2009. Allelopathic effect of Tithonia diversifolia on the germination, growth and chlorophyll contents of maize (Zea mays L.). Scientific Research and Essays 4(12): 1553-1558. Paschold, A., R. Halitschke, and A. Baldwin. 2006. Using ‘mute’ plants to translate volatile signals. The Plant Journal 45(2): 275-291. Pagano, E., and G. Maddoni. Intra-specific competition in maize: Early established hierarchies differ in plant growth and biomass partitioning to the ear around silking. Field Crops Research 101(3): 306-320. Painter, R. 1958. Resistance of plants to insects. Annual Review of Entomology 3:267-290. Peferoen, M. 1997. Progress and prospects for field use of Bt genes in crops. Trends in Biotechnology 15(5): 173-177. Pilz, C., R. Wegensteiner, and S. Keller. 2007. Selection of entomopathogenic fungi for the control of the western corn rootworm Diabrotica virgifera virgifera. Journal of Applied Entomology 131(6): 426-431. Pilz, C., R. Wegensteiner, and S. Keller. 2008. Natural occurrence of insect pathogenic fungi and insect parasitic nematodes in Diabrotica virgifera virgifera populations. Biocontrol 53(2): 353-359. Poelman, E. 2013. New synthesis: Volatiles bring out the animal in plants. Journal of Chemical Ecology 39(8): 1055.

Pray, C., D. Ma, J. Huang, and F. Qiao. 2001. Impact of Bt cotton in China. World Development 29(5): 813-825. Prischmann, D., R. Lehman, A. Christie, and K. Dashiell. 2008. Characterization of bacteria isolated from maize roots: Emphasis on Serratia and infestation with corn rootworms (Chrysomelidae: Diabrotica). Applied Soil Ecology 40(3): 417-431. Pruess, K., G. Weekman, and B. Somerhalder. 1968. Western corn rootworm egg distribution and adult emergence under two corn tillage systems. Journal of Economic Entomology 61(5): 1424-1427. Qaim, M., and A. de Janvry. 2005. Bt cotton and pesticide use in Argentina: economic and environmental effects. Environment and Development Economics 10(2): 179-200. Rajcan, I., and C. Swanton. 2001. Understanding maize-weed competition: resource competition, light quality and the whole plant. Field Crops Research 71(2): 139-150. Rasmann, S., T. Köllner, J. Degenhardt, I. Hiltpold, S. Toepfer, U. Kuhlmann, J. Gershenzon, and T. Turlings. 2005. Recruitment of entomopathogenic nematodes by insect-damaged maize roots. Nature 434(7034): 732-737. Raymond, B., P. Johnston, C. Nielsen-LeRoux, D. Lereclus, and N. Crickmore. 2010. Bacillus thuringiensis: an impotent pathogen? Trends in Microbiology 18(5): 189-194. Raymond, B., K. Wyres, S. Sheppard, R. Ellis, and M. Bonsall. 2010. Environmental factors in determining the epidemiology and population genetic structure of the Bacillus cereus group in the field. Public Library of Science Pathogens 6(5): e1000905. Raymond, B., D. Davis, and M. Bonsall. 2007. Competition and reproduction in mixed infections of pathogenic and non-pathogenic Bacillus spp. Journal of Invertebrate Pathology 96: 151-155. Ridenour, W., and R. Callaway. 2001. The relative importance of allelopathy in interference: the effects of an invasive weed on a native bunchgrass. Oecologia 126(3): 444- 450. Riedell, W., and P. Evenson. 1993. Rootworm feeding tolerance in single-cross maize hybrids from different eras. Crop Science 33(5): 951-955.

Robert, C., N. Veyrat, G. Glauser, G. Marti, G. Doyen, N. Villard, M. Gaillard, T. Köllner, D. Giron, M. Body, B. Babst, R. Ferrieri, T. Turlings, and M. Erb. 2012. A specialist root herbivore exploits defensive metabolites to locate nutritious tissues. Ecology Letters 15(1): 55-64. Salehi, J., A. Abad, A. Seifinejad, R. Marzban, K. Kariman, and B. Maleki. 2008. Distribution and diversity of Dipteran-specific cry and cyt genes in native Bacillus thuringiensis strains obtained from different ecosystems of Iran. Journal of Industrial Microbiology and Biotechnology 35(2): 83-94. Sappington, T., B. Siegfried, and T. Guillemaud. 2006. Coordinated Diabrotica genetics research: Accelerating progress on an urgent insect pest problem. American Entomologist 52(2): 90-97. Sanchis, V. 2011. From microbial sprays to insect-resistant transgenic plants: history of the biopesticide Bacillus thuringiensis. A review. Agronomy for Sustainable Development 31(1): 217-231. Scharf, M., S. Parimi, L. Meinke, L. Chandler, and B. Siegfried. 2001. Expression and induction of three family 4 cytochrome P450 (CYP4) genes identified from insecticide- resistant and susceptible western corn rootworms, Diabrotica virgifera virgifera. Insect Molecular Biology 10(2): 139-146. Scherm, H., H. Ngugi, A. Savelle, and J. Edwards. 2004. Biological control of infection of blueberry flowers caused by Monilinia vaccinii-corymbosi. Biological Control 29(2): 199- 206. Schnepf, E., N. Crickmore, J. van Rie, D. Lereclus, J. Baum, J. Feitelson, D. Zeigler, and D. Dean. 1998. Bacillus thuringiensis and its pesticidal crystal proteins. Microbiology and Molecular Biology Reviews 62(3): 775-806. Seiedy M, Saboori A, Zahedi-Golpayegani A (2013) Olfactory response of Phytoseiulus persimilis (Acari: Phytoseiidae) to untreated and Beauveria bassiana-treated Tetranychus urticae (Acari: Tetranychidae). Experimental and Applied Acarology 60(2): 219-227. Sekar, V. 1988. The insecticidal crystal protein gene is expressed in vegetative cells of Bacillus thuringiensis var. tenebrionis. Current Microbiology 17(6): 347-349.

Semchenko, M., S. Saar, and A. Lepik. 2014. Plant root exudates mediate neighbor recognition and trigger complex behavioural changes. New Phytologist 204(3): 631-637. Shakeri, J., and H. Foster. 2007. Proteolytic activity and antibiotic production by Trichoderma harzianum in relation to pathogenicity to insects. Enzyme and microbial technology 40(4): 961-968. Siegfried, B., T. Vaughn, and T. Spencer. 2005. Baseline susceptibility of western corn rootworm (Coleoptera: Chrysomelidae) to Cry3Bb1 Bacillus thuringiensis toxin. Journal of Economic Entomology 98(4): 1320-1324. Singh, N. A. Sing, and D. Singh. 2010. Autotoxicity of maize and its mitigation by plant growth promoting rhizobacterium Paenibacillus polymyxa. Allelopathy Journal 25(1): 195- 204. Sivasupramaniam, S., G. Head, L. English, Y. Li, and T. Vaughn. 2007. A global approach to resistance monitoring. Journal of Invertebrate Pathology 95(3): 224-226. Smith, R., and G. Couche. 1991. The phylloplane as a source of Bacillus thuringiensis variants. Applied and Environmental Microbiology 57(1): 311-315. Stahly, D., D. Dingman, L. Bulla, and A. Aronson. Possible origin and function of the parasporal crystals in Bacillus thuringiensis. 1978. Biochemical and Biophysical Research Communications 84(3): 581-588. Steinhaus, E. 1956. Microbial control- the emergence of an idea. A brief history of insect pathology through the nineteenth century. Hilgardia 26(2): 107-157. Storer, N., J. Babcock, and J. Edwards. 2006. Field measures of western corn rootworm (Coleoptera: Chrysomelidae) mortality caused by Cry34/35Ab1 proteins expressed in maize event 59122 and implications for trait durability. Journal of Economic Entomology 99(4): 1381-1387. Strauss, S., J. Rudgers, J. Lau, and R. Irwin. 2002. Direct and ecological costs of resistance to herbivory. Trends in Ecology and Evolution 17(6): 278-285. Strnad, S., M. Bergman, and W. Fulton. 1986. First-instar western corn rootworm (Coleoptera: Chrysomelidae) response to carbon dioxide. Environmental Entomology 15(4): 839-842.

Strnad, S., and M. Bergman. 1987. Distribution and orientation of western corn rootworm (Coleoptera: Chrysomelidae) larvae in corn roots. Environmental Entomology 16(5): 1193- 1198. Subramanian, A., and M. Qaim. 2009. Rural poverty and employment effects of Bt cotton in India. International Association of Agricultural Economists 2009 Conference, August 16-22, Beijing, China. Szalai, M., J. Kiss, S. Kövér, and S. Toepfer. 2014. Simulating crop rotation strategies with a spatiotemporal lattice model to improve legislation for the management of the maize pest Diabrotica virgifera virgifera. Agricultural Systems 124: 39-50. Tabashnik, B. 2008. Delaying insect resistance to transgenic crops. Proceedings of the National Academy of Sciences of the United States of America. 105(49): 19029-19030. Tabashnik, B., A. Gassmann, D. Crowder, and Y. Carriére. 2008. Insect resistance to Bt crops: evidence versus theory. Nature Biotechnology 26(2): 199-202. Tallamy, D., D. Whittington, F. Defurio, D. Fontaine, P. Gorski, and P. Gothro. Sequestered cucurbitacins and pathogenicity of Metarhizium anisopliae (Monilales: Moniliaceae) on spotted cucumber beetle eggs and larvae (Coleoptera: Chrysomelidae). Environmental Entomology 27(2): 366-372. Tamiru, A., T. Bruce, C. Midega, C. Woodcock, M. Birkett, J. Pickett, and Z. Khan. 2012. Oviposition induced volatile emissions from African smallholder farmers’ maize varieties. Journal of Chemical Ecology 38(3): 231-234. Tan, S., X. Chen, D. Li, and H. Zhang. 2001. Can other host species of cotton bollworm be non-Bt refuges to prolong the effectiveness of Bt-cotton? Chinese Science Bulletin 46(21): 1804-1808. Tang, J., H. Collins, T. Metz, E. Earle, J. Zhao, R. Roush, and A. Shelton. Greenhouse tests on resistance management of Bt transgenic plants using refuge strategies. Journal of Economic Entomology 94(1): 240-247. Taylor, C. and G. Georghiou. Suppression of insecticide resistance by alteration of gene dominance and migration. Journal of Economic Entomology 72(1): 105-109.

Taylor, M. and J. Tick. 2003. Post-market oversight of biotech foods: is the system prepared? Pew Initiative on Food and Biotechnology, Washington, DC 2003. Thompson, S., R. Brandenburg, and G. Roberson. 2007. Entomopathogenic fungi detection and avoidance by mole crickets (Orthoptera: Gryllotalpidae). Environmental Entomology 36(1): 165-172. Torsvik, V., and L. Øvreas. 2002. Microbial diversity and function in soil: from genes to ecosystems. Current Opinion in Microbiology 5(3): 240-245. Townsend, M., D. Johnson, and D. Steinkraus. 1998. Laboratory studies of the interactions of environmental conditions on the susceptibility of green June beetle (Coleoptera: Scarabaeidae) grubs to entomopathogenic nematodes. Journal of Entomological Science 33(1): 40-48. Traxler, G., S. Godoy-Avila, J. Falck-Zepeda, and J. Espinoza-Arellano. Transgenic cotton in Mexico: economic and environmental impacts. Auburn, AL: Department of Agricultural Economics, Auburn University. Available from: http://www.infoagro.net/shared/docs/a2/Traxler.pdf Turlings, T., and J. Tumlinson. 1992. Systemic release of chemical signals by herbivore- injured corn. Proceedings of the National Academy of Sciences of the United States of America 89(17): 8399-8402. Van Gerrewey, T. 2015. Use of plant-growth promoting organisms (PGPO) in hydroponic potatoes for bioregenerative life support systems in space. MS thesis. University of Gent. Van Rensburg, J. First report of field resistance by the stem borer, Busseola fusca (Fuller) to Bt-transgenic maize. South African Journal of Plant and Soil 24(3): 147-151. Varma, A., S. Verma, Sudha, N. Sahay, B. Bütehorn, and P. Franken. 1999. Piriformospora indica, a cultivable plant-growth-promoting root endophyte. Applied and Environmental Microbiology 65(6): 2741-2744. Vaughn, T., T. Cavato, G. Brar, T. Coombe, T. DeGooyer, S. Ford, M. Groth, A. Howe, S. Johnson, K. Kolacz, C. Pilcher, J. Purcell, C. Romano, L. English, and J. Pershing. 2005. A method of controlling corn rootworm feeding using a Bacillus thuringiensis protein expressed in transgenic maize. Crop Science 45(3): 931-938.

Vilas-Bȏas, G., A. Peruca, and O. Arantes. Biology and taxonomy of Bacillus cereus, Bacillus anthracis, and Bacillus thuringiensis. 2007. Canadian Journal of Microbiology 53(6): 673-687. Villani, M., S. Krueger, P. Schroeder, F. Consolie, N. Consolie, L. Preston-Wilsey, and D. Roberts. 1994. Soil application effects of Metarhizium anisopliae on Japanese beetle (Coleoptera, Scarabaeidae) behavior and survival in turfgrass microcosms. Environmental Entomology 23(2): 502-513. Walker, T., H. Bais, E. Grotewold, and J. Vivanco. 2003. Root exudation and rhizosphere biology. Plant Physiology 132(1): 44-51. Wang, H., B. Coates, H. Chen, T. Sappington, T. Guillemaud, and B. Siegfried. 2013. Role of a gama-aminobutyric acid (GABA) receptor mutation in the evolution and spread of Diabrotica virgifera virgifera resistance to cyclodienes insecticides. Insect Molecular Biology 22(5): 473-484. Wang, L., J. Huang, M. You, X. Guan, and B. Liu. 2005. Effects of toxins from two strains of Verticillium lecanii (Hyphomycetes) on bioattributes of a predatory ladybeetle, Delphastus catalinae (Col., Coccinellidae). Journal of Applied Entomology 129(1): 32-38. Weatherwax, P. 1954. Indian corn in old America. Macmillan, New York. p.53. Yamada, S., Y. Takayama, M. Yamanaka, K. Ko, and I. Yamaguchi. 1990. Biological activity of antifungal substances produced by Bacillus subtilis. Journal of Pesticide Science 15(1): 95-96. Yanagawa, A., and S. Shimizu. 2007. Resistance of the termite, Coptotermes formosanus Shiraki to Metarhizium anisopliae due to grooming. Biocontrol 52(1): 75-85. Yancey, C. 1828. Extract to the editor. American Farmer 10(1): 3. Zakir, A., M. Sadek, M. Bengtsson, B. Hansson, P. Witzgall, and P. Anderson. 2013. Herbivore-induced plant volatiles provide associational resistance against an ovipositing herbivore. Journal of Ecology 101(2): 410-417.

Chapter 2

Transgenic Bt Toxin Expression in Corn Roots and Interactions with Western Corn Rootworm Under Two Refuge Mimicking Scenarios.

Introduction

Corn rootworms, Diabrotica spp. (Coleoptera: Chrysomelidae), are the most economically important pests of corn in North America, and many aspects of the corn/rootworm complex make management of these pests difficult. Yield losses and management costs resulting from these pests cost over $2 billion per year in the United States alone (Paul Mitchell, personal communication). More insecticides in the US are used to manage rootworms than any other single agricultural pest (EPA 2005). Moreover, WCR have repeatedly shown an ability to develop resistance to traditional single-pronged management methods, including not only pesticides (Ball and Weekman 1962, Meinke et al. 1998) but also crop rotation (Levine et al. 2002). Crucial in managing WCR and other pests, Bacillus thuringiensis Berliner (Bt) is a gram-positive, soil-dwelling bacterium that produces spores which contain insecticidal crystal proteins, commonly referred to as Cry proteins. Because both the gene sequence responsible for expressing these Cry proteins and the technology allowing us to insert these genes into corn via transformation events are available (Carozzi and Koziel 1997), transgenic corn containing Cry toxins effective against WCR are now on the market. These have been approved by the Environmental Protection Agency (EPA) since 2003 (Vaughn et al. 2005). Slowing the rate at which rootworms develop resistance to management methods such as transgenic Bt crops can allow farmers to preserve and utilize the best available practices. To this end, the EPA requires implementation of resistance management programs (EPA 2014) and resistance monitoring (EPA 2010 a) wherever transgenic crops are used in the United States. Thus, Bt crops must be coupled with refuges of non-Bt-expressing, susceptible plants of the same species. Under these planting scenarios, resistance development is delayed because small populations of resistant insects emerging from Bt crop fields can mate with the larger populations of susceptible pests emerging from the refuge sites (Gould 1998). Since the introduction of Bt-expressing crops, refuges have been planted in single blocks or strips away from Bt corn (EPA 2010 b). However, seed-mix refuges, in which transgenic corn is interspersed throughout the field with randomly placed non-Bt corn, are

now being used as an alternative for rootworms (EPA 2010 c). Critics of seed mix refuges (also known as “Refuge in the bag” or RIB) have voiced concerns that pests may ingest sublethal doses if resistant and susceptible plants are close enough. Larvae appear to congregate on susceptible plants in seed mix refuges (Rudeen and Gassmann 2012), which may accelerate resistance development (Murphy et al. 2010). Behavioral non-preference of Bt corn roots, increased WCR tolerance to Bt, or reduced exposure time for larvae as they move back and forth from susceptible to Bt corn roots (Murphy et al. 2010) may result in insect resistance. Under such scenarios, mathematical models show that resistance may evolve much faster than with a traditional block refuge (Mallet and Porter 1992). On the other hand, if larvae move from refuge to Bt plants and die, this will reduce the effective size of the available refuge (Head and Greenplate 2012). Moreover, Clark et al. (2006) suggested that if transgenic corn had some roots that expressed Bt at a lower rate than others, this might provide a possible route of resistance development. It is known that Bt toxin expression is not uniform. For instance, volunteer Cry3Bb corn is known to produce lower levels of toxin than intentionally planted Cry3Bb corn (Krupke 2009), while older corn expresses significantly lower levels of Cry3Bb than younger plants (Vaughn et al. 2005). Clark et al. (2006) attributed divergent behavior in neonate WCR offered Bt corn roots to the WCR being able to detect Bt toxin expression, which triggered non-preference behavior by the insects. Although WCR resistance to Bt corn has been observed multiple times, the mechanism for resistance is not currently known. WCR have been shown to develop resistance to multiple lines of rootworm-active Bt corn in the laboratory (Miehls et al. 2008, Frank et al. 2013). This resistance is nonrecessive, and attempts to find a fitness cost to resistance have been unsuccessful (Petzold-Maxwell et al. 2012). Moreover, even with the refuge implementation requirement, rootworm-active Bt corn has been shown to offer protection that varies with location under field conditions (Gray et al. 2007). Alarmingly, eleven populations in Iowa have already been confirmed resistant (Cullen et al. 2013), with populations in Colorado, Minnesota, Nebraska, and South Dakota confirmed to have reduced susceptibility to Bt (EPA 2014 a).

This project was designed to investigate WCR behavioral responses to non-Bt corn and Bt corn when presented alone (representative of block/strip refuge) or simultaneously (representative of seed mix refuge). We hypothesized that more larvae would be found on Bt than on non-Bt plants in both block/strip and seed mix refuge scenarios; that Bt expression would differ along the root length; and that this difference would be reflected in larval feeding preferences along the root length. We believe this information will help assess the role of Bt toxin behavioral escape by rootworms, which could lead to faster resistance development. Methods and Materials Experimental System and Bioassay Arenas. The Western corn rootworms used for these experiments were originally obtained from non-diapausing Bt-susceptible colonies, maintained at the USDA-ARS facilities (Brookings Co., SD) and Crop Characteristics, Inc. (Farmington, MN). Eggs, larvae, and adults were subsequently kept in a room maintained at 25°C, 75 ± 10% relative humidity, and 16:8 light: dark. Larvae were provided with sprouted corn seedlings provided by DuPont Pioneer (Johnston, IA), as described below. Adults were fed prepared adult WCR diet (Bio-Serv, Frenchtown, NJ), as needed. Corn seedlings used for experimental assays were from a transformed rootworm resistant line (Pioneer Herculex Xtra, 35F44, Bt) and a genetically similar susceptible line (Pioneer, 35F38, non-Bt). Corn seeds were soaked in water for 48 hours, Bt separately from non-Bt, and seeds that did not sprout were discarded. After 48 hours, seeds were sown in our experimental arenas as needed. Experimental arenas (Figure 1) were made from 10 or 20 cm2 (as specified) acrylic squares held 1.5 cm apart by acrylic strips and cut 0.16 cm thick gasket sheeting (Goodyear, Rancho Cucamonga, CA) along 3 sides of the perimeter of one of the plates to maintain a tight seal. The two plates were held together with two hand-made clamps across the top and bottom of the arena consisting of two boards approximately 3.75 cm wide held together with screws. Arenas were filled with a growth medium made of 10% agar and 1% Miracle-Gro ® fertilizer (ScottsMiracle-Gro ®, Marysville, OH) in water, modified after Clark et al. (2006). Corn was grown in these arenas and clear agar substrate rather than in pots containing soil to facilitate visual larval behavioral assessment under red

light. Arenas were covered with aluminum foil between observations to eliminate light penetration to corn roots, leaving only the plant tops exposed to light. Insect Behavioral Responses to Non-Bt and Bt Corn under Block/strip and Seed mix Refuge Scenarios. Block/strip Mimicking (No Choice) Scenario. The purpose of these assays was to determine if host selection behavior of larval WCR differed when presented with Bt or non- Bt plants under no-choice conditions. Therefore, we recorded settling preference along the root length on Bt and non-Bt roots independently. We hypothesized that if larvae can detect and respond to toxin expression, their behavior would be divergent with respect to Bt plants and their non-Bt counterparts. Specifically, portions of Bt root tissue that expressed the toxin at lower rates would be preferred. In order to assess larval behavior, experimental arenas (10 cm square) were sown with sprouted corn seeds of either a transformed (Bt) line or an untransformed (non-Bt) line. Two sprouted seeds of the corn line of interest were embedded into the agar in the center of the arena. No-choice assays were set up with one arena containing the Bt and another containing the non-Bt corn, so larval responses to both lines could be documented simultaneously. After corn was grown in arenas for seven days, four third instar WCR were placed on the agar substrate equidistant from Bt and non-Bt plants. Third-instar WCR were chosen because they are better able to move from plant to plant and are easier to observe. For observation purposes, each root was divided into 4 regions, each approximately 2 cm long, based on distance from the root crown and indicated by marks made on the outside of the arena. Region 1 was designated as the region starting at the root crown and region 4 was designated as the region containing the root tip. Locations of larvae (discussed below) with respect to root region were observed and recorded every 6 hours for 48 hours. Arenas were kept under light and humidity conditions described above, and position of Bt/non-Bt plants was randomized to eliminate the potential for a positional effect. Experiments were run in single replicates and repeated over time with new sets of arenas, plants, and insects to obtain a total of 20 replicates. The effect of corn genotype, observation time, root region and their two and three-way interactions on number of larvae

were analyzed using PROC MIXED (SAS Institute, Cary, NC, 2010) due to differences in population variation. Significant effects were subjected to Tukey's mean separation tests (P≤0.05). Seed Mix Mimicking (Choice) Scenario. This experiment was conducted simultaneously with the block/strip mimicking scenario assays. The purpose of these assays was to determine whether WCR larvae were able to detect and avoid Bt-expressing roots when given a choice. We hypothesized that if WCR are indeed able to detect differences between Bt and non-Bt roots, they will move from Bt to non-Bt plants over time, resulting in overall higher number of insects associated with non-Bt roots. For these purposes, two sprouted corn seeds from each of the corn lines were embedded into the agar substrate of 20 cm2 arenas, with orientation of the Bt/non-Bt seeds (left/right) randomized to prevent potential positional effect. Seeds were placed 10 cm apart to provide ample growth space and minimize root entanglement, to facilitate accurate insect behavior assessment. This distance was fairly realistic with respect to field conditions (Lauer and Rankin 2004), although our substrate was not. After corn was grown in arenas for seven days, four third instar WCR were placed on the agar substrate equidistant from Bt and non-Bt plants. The number of larvae associated with each of the root regions of either corn line were documented every 6 hours for 48 hours. Experiments were run in single replicates and repeated over time with new sets of arenas, plants, and insects to obtain a total of 20 replicates. Data for the difference between number of larvae found on non-Bt and Bt individual root regions were transformed using Log(n+4) to conform to assumptions of normality prior to statistical analyses. The effect of time, root region and their interaction on the difference between numbers of larvae found associated with each of the two corn genotypes were analyzed using ANOVA in PROC GLIMMIX (SAS Institute, Cary, NC, 2010).

Environmental Factors and Root Phenology and their Effects on Bt Toxin Expression. Effect of Plant Growth Substrate on Root Bt Expression. The purpose of this assay was to ensure that the agar planting substrate used in our experimental arenas did not interfere with toxin expression. We hypothesized that the agar substrate used in our experimental arenas would not impact toxin expression in roots of Bt corn. To test for substrate effects on root toxin expression, sprouted corn seeds were grown in plexiglass experimental arenas described above filled with either the agar substrate used in our behavioral assays or commercially acquired topsoil (Scotts Miracle-Gro, Marysville, OH). Four 10 cm2 arenas with individual Bt corn plants and four 20 cm2 arenas with one Bt and one non-Bt plant were sown for this experiment. Plants were grown for 7 days under the environmental conditions described previously. After behavioral assays were completed, an additional 48 hours, roots from each Bt plant were removed from the soil, washed and pat- dried, then divided into the same 4 regions described in our experimental assay, and stored in labeled micro-centrifuge tubes in -80 ºC until needed for toxin expression analyses. Root regions from two plants within each arena type were pooled randomly so that they would yield enough tissue for toxin analyses, yielding a total of four replicates for each root region/substrate type (agar vs soil). Toxin expression analyses were performed as described in the next section. Effects of substrate, root region, and their interaction on Cry34, Cry35 and combined (Cry34 + Cry35) toxin expression levels were tested using ANOVA in PROC MIXED (SAS Institute, Cary, NC, 2010). Significant effects were followed by Tukey’s mean separation tests (P≤0.05). Root Bt Toxin Expression Along the Root Length under Two Planting Scenarios and Insect Infestation Levels. Using our agar experimental arenas, we determined root toxin expression along the four root regions in corn grown under the two refuge planting scenarios of interest (seed mix vs block/strip) that were either left uninfested (uninfested) or were infested with four third instars for 48 hours (infested). Because stress can downregulate production of Bt toxin (Dong and Li 2006), we hypothesized that insect infestation and presence of a non-Bt plant would decrease Bt concentration.

For protein expression, roots from Bt and non-Bt seedlings used in the previous assays were washed, dried, divided into the same four root regions as in the larval assays, and analyzed for Bt expression. Root sections were stored in microcentrifuge tubes and frozen until needed for enzyme-linked immunosorbant assays (ELISAs) (Meissle 2009), which were performed at the Pioneer laboratories in Johnston, IA, in accordance with Pioneer's proprietary ELISA protocol. Root sections from multiple plants were combined as needed to obtain approximately 20 mg dry weight per section for each of the treatments so that there was enough tissue to perform ELISA. Tissue pooling resulted in variable number of replicates per planting scenario/insect infestation level combination, yielding 11 replicates for uninfested seed mix, 20 replicates for infested seed mix, 17 replicates for uninfested block/strip and 25 for infested block/strip. ELISAs were used to comparatively evaluate Bt expression for the 4 different regions of the Bt roots. Six non-Bt root samples were included as negative controls to ensure assay accuracy. Both Cry34 and Cry35 Bt protein were quantified, and both assays were validated against samples with known quantities of toxin for use with these samples. Effects of substrate, planting scenario and root region on toxin (Cry34, Cry35 and their combination) expression were compared using ANOVA in PROC GLM (SAS Institute, Cary, NC 2010). Significant effects were further analyzed using Tukey’s mean separation tests (P≤0.05). Results Insect Behavioral Responses to Non-Bt and Bt Corn under Block/strip and Seed Mix Planting Scenarios. Block/strip Mimicking (No Choice) Scenario. Number of insects associated with roots was not significantly affected by observation time or its interactions with corn line, root region, or the combination of corn line and root region. There was also no significant effect of the interaction between corn line and root region on number of larvae observed associated with corn roots in this experiment. On the other hand, number of larvae observed in contact with Bt vs non-Bt roots were found to be statistically different (F=10.09; df=1, 36; p=0.0031), with higher number of larvae observed contacting roots within non-Bt corn arenas

(t=-2.21; P≤0.0273) (Figure 2A). The number of larvae associated with each root region was also significantly different (F=29.80; df=3, 108; p<0.0001), regardless of corn genotype. Differences in number of larvae associated with regions 3 and 4 were the only ones that were not statistically different (Figure 2B). Seed Mix Mimicking (Choice) Scenario. Under this scenario, number of insects associated with roots was not significantly affected by observation time or its interaction with root region. Similarly, number of insects associated with roots was not affected by observation time or its interaction with root region. On the other hand, number of larvae in contact with non-Bt roots was significantly higher than number of larvae in contact with Bt-expressing roots (F=2.86; df=3, 150; p=0.0390) (Figure 3 A). Similarly, number of insects associated with individual root regions was significantly different (F=38.97; df=3, 114; p<0.0001), regardless of corn line. Overall, higher numbers of insects were observed contacting root region 1 (crown), and significantly lower number of insects were observed contacting root region 4 (root tip) (Figure 3B). Environmental Factors and Root Phenology and their Effects on Bt Toxin Expression. Effect of Plant Growth Substrate on Root Bt Expression. The effect of substrate type (agar or soil) on Cry34 expression was significant, but the effect of root region and the interaction between substrate type and root region were not significant. The effects of substrate type, root region, and their interaction were not significant on Cry35 expression, with a mean ± SE expression of 25.840 ± 1.620 ng/g of dry tissue in agar arenas and 22.653 ± 10.725 ng/g of dry tissue in soil arenas. Contrastingly, Cry34 expression was significantly greater in our agar arenas, with a mean ± SE expression of 56.235 ±3.696 ng/g of dry tissue, than in soil arenas, with a mean expression of 29.445 ± 2.100 ng/g of dry tissue (F=36.14; df=1, 28; p<0.0001), thus satisfying concerns that our arenas might negatively impact root toxin expression. Root Bt Toxin Expression Along the Root Length under Two Planting Scenarios and Insect Infestation Levels

There was no significant difference in the expression of either Cry34 or Cry35 Bt protein for each root region from the crown to the tip. (F=0.08; df=3, 57; p=0.9718). Interactions between root region and other effects were not significant. The effect of infestation alone on Bt toxin expression was not significant (F=0.17; df=1, 69; p=0.6794), nor was the effect of refuge scenario alone (F=0.92; df=1, 69; p=0.3413). However, the interaction between these effects was significant. In the seed mix (choice) scenario, we found significantly higher combined Bt toxin expression in roots that were infested by insects. Mean Cry34 toxin expression was significantly higher in infested root tissue (F=5.16; df=1, 29; p=0.0307), as was mean Cry35 toxin (F=7.40; df=1, 30; p=0.0109) (Figure 4 A). These trends were reversed in the block/strip (no choice) scenario, where both Cry34 (F=6.44; df=1, 40; p=0.0151and Cry35 (F=7.40; df=1, 29; p=0.0109) Bt toxin expression levels were higher in uninfested plants compared to WCR infested plants (Figure 4A). One interesting observation made during the course of this experiment was a significant increase in Cry35 toxin expression in infested plants that were grown in pairs with non-Bt corn (Figure 4 B). When plants were infested with insects, Cry34 toxin expression levels in Bt plants were elevated when roots were grown in close proximity to non-Bt corn (F=6.56; df=1, 43; p=0.0141), as were Cry35 toxin expression levels (F=14.10; df=1, 43; p=0.0005) (Figure 4 B). On the other hand, when plants were not infested with WCR larvae, this trend was reversed. That is, uninfested Bt plants expressed significantly higher (F=7.75; df=1, 27; p=0.0099) levels of Cry34 toxin when grown in isolation than when grown in close proximity to non-Bt plants (Figure 4 B). Curiously, expression levels of Cry35 Bt toxin did not differ significantly, although trends were similar (Figure 4 B). Discussion Our newly developed observational chambers showed themselves to be useful tools for soil insect behavioral research. Larvae were significantly more likely to be found on Bt- expressing than on non-Bt-expressing roots. Level of toxin expressed was affected by presence of insect infestation and by type of refuge, but not by distance from root crown.

Our experimental chambers filled with agar/Miracle Gro™ medium proved adequate for observing WCR behavior in response to corn roots. Moreover, results from this study show that use of agar substrate does not negatively impact toxin expression by corn roots. On the contrary, significantly higher expression levels of one Bt toxin were expressed by roots grown in agar- than in soil-filled arenas, and effects of treatment on Bt toxin expression were consistent regardless of substrate type. Thus, we believe that our agar arenas can be considered a suitable tool for assessing WCR responses to plant roots. The reason that toxin expression was higher in our agar arenas is unknown, but nutrient availability is known to affect Bt toxin expression (Bruns and Abel 2003). Although a nutritional comparison between soil and agar substrates was beyond the objectives of our study, it is quite plausible that levels of plant available nutrients varied between the two substrates used and this could account for the differential toxin expression observed. Our observations using these experimental arenas allowed us to determine that WCR larvae were found significantly more associated with non-Bt-expressing than with Bt-expressing corn roots. The mechanisms leading to Bt resistance in WCR are not currently known. However, researchers have previously proposed both rootworm behavior (Clark et al. 2006) and the lack of high dose toxin expression in rootworm-active Bt corn (Binning et al. 2010) as potential factors contributing to insect resistance. We hypothesized that rootworm preference for root tissue would differ with distance from root crown between non-Bt and Bt corn lines, and that this would correlate with a difference of toxin expression along the root length, as documented in the Cry3Bb experiments performed by Meissle et al. (2009). If this were true, it would mean that rootworms would ingest lower doses of Bt, a possible path to resistance development. Indeed, significantly more WCR larvae were found in contact with the crown portion of the root tissue on non-Bt than on Bt corn. Yet, contrary to our expectations, differential preference for root regions could not be attributed to toxin expression levels along the root. These unexpected results could be explained by limitations in sensitivity of the Bt toxin detection methodology used (ELISAs) or by other physical or physiological differences in root tissue which may account for insect preference. Rootworms' preference for root crowns was higher and this may be because root crowns are more nutritious and

exude more defensive compounds, which WCR use to locate food sources (Robert et al. 2012 b). Additionally, there is a pH gradient along the length of corn root tissue, and the direction of the gradient has been reported to vary with corn genotype (Gollany and Schumacher 1993). Because pH is an important factor in Bt effectiveness (Skrivanek et al. 2006), root pH may influence insect root tissue choice on Bt-expressing plants. Alternatively, it is also possible that insertion of transgenes can affect how other defense-related or nutrition-related genes are expressed. For example, transgenic Bt+CpTI cotton root tissue is believed to have reduced expression of element-binding proteins because it accumulates micronutrients at different rates than non-transgenic root tissue (Yukui et al. 2009). However, this is an area that merits further investigation within the system we studied. Although our data showed no significant effects of time on overall number of insects in contact with root tissue, we witnessed insects moving between Bt and non-Bt plant roots during the course of this experiment, when the insects were given a choice within our agar arenas (SLG, Pers. Obs.). Insects moving from plant to plant during development might result in sublethal doses of toxin, leading to resistance development. Indeed, WCR appear to aggregate on non-Bt corn in the field (Rudeen and Gassmann 2012). Larval movement away from Bt plants in a seed mix refuge has also been suggested for Helicoverpa zea (Yang et al. 2014). Another problem with this larval movement is that larval movement between plants may decrease the effective refuge size by ensuring that more insects will come into contact with transgenic Bt-expressing plants (Head and Greenplate 2012). However, it must be noted that we used 3rd instar WCR in our assays. Third instar WCR are much more mobile than neonates, but less susceptible to Bt toxin. In addition, the concentration of Bt in our arenas may not accurately reflect the concentration present in the field. Although significantly more insects contacted Bt than non-Bt corn roots, insects did still contact Bt corn roots. WCR use CO2 as a trigger to begin their search pattern, and other volatiles such as 2,4-Dihydroxy-7-methoxy-1,4-benzoxazin-3-one (DIMBOA) as more specific cues to orient themselves towards corn roots (Johnson and Nielsen 2012). Western corn rootworms prefer WCR-damaged plants over undamaged plants (Robert et al. 2012a). WCR uses plant defensive compounds to orient itself towards more nutritious root tissue

(Robert et al. 2012 b); thus, it is possible that previous damage by WCR made the Bt plants more attractive. However, we did not examine nutrient content of root tissue in this experiment. Importantly, we found that Bt toxin expression in transgenic corn is significantly impacted by the interaction of refuge planting scenario with insect infestation. These are novel findings that may contribute to our understanding of rootworm-host interactions and Bt resistance evolution in this rootworm pest system. Results from our toxin expression experiments show that Bt corn roots grown in the no-choice, block/strip planting scenario expressed lower toxin levels when they were infested by rootworms than when they were left uninfested. The reasons for this difference are not known at this time, but other environmental stressors such as nitrogen deficiency, drought, waterlogging, high temperature, and salinity are known to reduce the levels of Bt toxin expression, as reviewed by Dong and Li (2006); thus, stress incurred by insect feeding could have downregulated Bt toxin expression or caused plants to reroute resources to native defense pathways. On the other hand, our results show that when plants were grown in the seed mix, choice scenario, Bt corn roots expressed higher toxin levels when they were infested by rootworms than when they were not infested. Since one of the serious concerns regarding transgenic plants targeting rootworms is that these plants do not express a high dose of toxin, any factors leading to a change in toxin expression could affect the rate at which insect resistance to Bt develops. Our results imply that plant-plant interactions are taking place, leading to differential expression of Bt toxins in response to planting scenario and insect infestation. Toxin expression level trends in response to planting scenario/WCR infestation were similar for both Cry34 and Cry35, although differences in Cry35 were not always significant. This is likely because Cry35 is expressed at a lower rate than Cry34. In addition, we have been informed by industry professionals that Cry35 is less stable and breaks down more quickly, which might explain these results; however, we were unable to find reference to this in the literature. Transgenic corn grown in seed-mix refuges is much more likely to be grown in close proximity to non-Bt corn, because the refuge design is such that non-Bt corn is interspersed

randomly throughout the field; thus, it is worth investigating how these differences in Bt toxin expression in transgenic corn play out in the field. As noted earlier, these assays were performed with 3rd instar WCR, and our Bt concentrations may not accurately reflect those seen in the field. Results obtained as part of this study provide a significant contribution to our current understanding of WCR larval behavior and clearly documents early detection and non- preference of Bt transgenic roots by these insects. We have also determined that insect infestation and corn growth conditions (which may reflect refuge planting scenario) interact, resulting in significant changes in Bt toxin expression levels. The information contained here provides further evidence for, and sheds light on, the complexity of factors modulating WCR larval behavior and Bt toxin expression, under current resistance management mimicking scenarios. However, these experiments only scratch the surface and further research is necessary to determine if the phenomena documented in this study translate to field conditions, and how this may bear on our decisions regarding deployment of Bt and future management technologies against WCR. Acknowledgements The authors thank Dr. Fred Gould for comments and edits on an earlier version of this manuscript. We are also indebted to Dr. Consuelo Arellano (NCSU, Statistics Department) for statistical advice, and Janet Griffiths and Dina Espinoza (NCSU, Entomology) and Jennifer Anderson and Courtney Davis (Pioneer Hi-Bred) for technical support. Corn seed and technical assistance in assessing Bt toxin levels using ELISAs was generously provided by Pioneer Hi-Bred.

References

Ball H, Weekman G (1962) Insecticide resistance in adult western corn rootworm in Nebraska. J. Econ. Entomol. of Economic Entomology 55(4): 439-441. Berdegue M, Trumble J (1997) Interaction between linear furanocoumarins found in celery and a commercial Bacillus thuringiensis formulation on Spodoptera exigua (Lepidoptera: Noctuidae) larval feeding behavior. J. Econ. Entomol. 90(4): 961-966. Binning R, Lefko S, Millsap A, Thompson S, Nowatzki T (2010) Estimating western corn rootworm (Coleoptera: Chrysomelidae) larval susceptibility to event DAS-59122-7 maize. Journal of Applied Entomology 134(7): 551-561. Bruns H, Abel C (2003) Nitrogen fertility effects on Bt delta-endotoxin and nitrogen concentrations of maize during early growth. Agronomy Journal 95(1):207-211. Bryson H, Wilbur D, Burkhardt C (1953) The western corn rootworm, Diabrotica virgifera Lec. in Kansas. J. Econ. Entomol. 46(6): 995-999. Cannon R (1993) Prospects and progress for Bacillus thuringiensis-based pesticides. Pestic. Sci. 37(4): 331-335. Carozzi N, Koziel M, eds. (1997) Advances in Insect Control: The Role of Transgenic Plants. London: Taylor & Francis. Clark P, Vaughn T, Meinke L, Molina-Ochoa J, Foster J (2006) Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) larval feeding behavior on transgenic maize (MON 863) and its isoline. J. Econ. Entomol. 99(3): 722-727. Cullen E, Gray M, Gassmann A, Hibbard B (2013) Resistance to Bt corn by western corn rootworm (Coleoptera: Chrysomelidae) in the U.S. Corn Belt. Journal of Integrated Pest Management 4(3):D1-D6. Dong H, Li W (2006) Variability of endotoxin expression in Bt transgenic cotton. Journal of Agronomy and Crop Science 193(1):21-29. [EPA] Environmental Protection Agency (2005) EPA annual pesticide report. Available from http://www.epa.gov/oppfead1/annual/2005/05annualrpt.pdf

[EPA] Environmental Protection Agency (2010) Terms and conditions for Bt corn registrations. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt-corn- terms-conditions.pdf [EPA] Environmental Protection Agency (2010) Simulation models evaluation of pest resistance development to refuge in the bag concepts related to Pioneer submission. Available from http://nepis.epa.gov/Adobe/PDF/P100EBX2.pdf [EPA] Environmental Protection Agency (2010) Scientific Issues Related to Insect Resistance Management for SmartStax™ Refuge-in-the-Bag, a Plant-Incorporated Protectant (PIP) Corn Seed Blend. Available from http://www.epa.gov/scipoly/sap/meetings/2010/december/120810agenda.pdf [EPA] Environmental Protection Agency (2012) BPPD IRM Team review of corn rootworm monitoring data and unexpected damage reports for Cry3Bb1 expressing Bt Corn and academic reports of Cry3Bb1 field failures as well as corn rootworm resistance. Available from http://www.regulations.gov/" \l "%21documentDetail;D=EPA-HQ-OPP-2011-0922-0037 [EPA] Environmental Protection Agency (2014) Insect resistance management fact sheet for Bacillus thuringiensis (Bt) corn products. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_corn_refuge_2006.htm [EPA] Environmental Protection Agency (2014) White paper on corn rootworm resistance monitoring for Bt plant-incorporated protectants. Available from http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPP-2013-0490-0008 Fernandez-Cornejo J, McBride W (2000) Genetically engineered crops for pest management in US agriculture: farm-level effects. USDA. Available from http://ddr.nal.usda.gov/bitstream/10113/38264/1/CAT11079167.pdf Frank D, Zukoff A, Barry J, Higdon M, Hibbard B (2013) Development of resistance to eCry3.1Ab-expressing transgenic maize in a laboratory-selected population of Western corn rootworm (Coleoptera: Chrysomelidae). J. Econ. Entomol. 106(6): 2503-2513. Gassmann A, Petzold-Maxwell J, Keweshan R, Dunbar M (2011) Field-evolved resistance to Bt maize by Western corn rootworm. PLoS one 6(7): e22629.

Gollany H, Schumacher T (1993) Combined use of colorimetric and microelectrode methods for evaluating rhizosphere pH. Plant and Soil 154(2): 151-159. Gould F. (1998) Sustainability of transgenic insecticidal cultivars: Integrating pest genetics and ecology. Annu. Rev. Entomol. 43: 701-726. Gray M, Steffey K, Estes R, Schroeder J. (2007) Responses of transgenic maize hybrids to variant western corn rootworm larval injury. Journal of Applied Entomology 131(6): 386- 390. Head, G., and J. Greenplate. 2012. The design and implementation of insect resistance management programs for Bt crops. GM Crops and Food: Biotechnology in Agriculture and the Food Chain 3(3): 144-153. Hill R (1975) Mating, oviposition patterns, fecundity and longevity of western corn rootworm. J. Econ. Entomol. 68(3): 311-315. Johnson S, Nielsen U. 2012. Foraging in the dark- chemically mediated host plant location by belowground insect herbivores. Journal of Chemical Ecology 38(6): 604-614. Kang J, Krupke C. (2009) Likelihood of multiple mating in Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae). J. Econ. Entomol. 102(6): 2096-2100. Krupke C, Marquardt P, Johnson W,Weller S, Conle S. (2009) Volunteer corn presents new challenges for insect resistance management. Agronomy Journal 101(4): 797-799. Lauer, J., and Rankin, M. 2004. Corn response to within row plant spacing variation. Agronomy Journal 96(5): 1464-1468. Levine E, Spencer JL, Isard SA, Onstad DW, Gray ME (2002) Adaptation of the western corn rootworm to crop rotation: evolution of a new strain in response to a management practice. American Entomologist 48(2): 94-107. Mallet J, Porter P (1992) Preventing insect adaptation to insect-resistant crops: are seed mixtures or refugia the best strategy? Proc. R. Soc London, 250(1328): 165-169. Meinke L, Siegfried B, Wright R, Chandler L (1998) Adult susceptibility of Nebraska western corn rootworm (Coleoptera: Chrysomelidae) populations to selected insecticides. J. Econ. Entomol. 91(3): 594-600.

Meissle M, Pilz C, Romeis J. (2009) Susceptibility of Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) to the entomopathogenic fungus Metarhizium anisopliae when feeding on Bacillus thuringiensis Cry3Bb1-expressing maize. Appl. Environ. Microbiol. Microbiology 75(12): 3937-3943. Meihls L, Higdon M, Siegfried B, Miller N, Sappington T, Ellersieck M, Spencer T, Hibbard B. (2008) Increased survival of western corn rootworm on transgenic corn within three generations of on-plant greenhouse selection. Proc. Natl. Acad. Sci. U. S. A. 105(49): 19177-19182. Mueting S, Strain K, Lydy M (2014) Validation of an extraction method for Cry1Ab protein from soil. Environmental Toxicology and Chemistry 33(1): 18-25. Murphy A, Ginzel M, Krupke C (2010) Evaluating western corn rootworm (Coleoptera: Chrysomelidae) emergence and root damage in a seed mix refuge. J. Econ. Entomol. 103(1): 147-157. Musser R, Hum-Musser S, Eichenseer H, Peiffer M, Ervin G, Murphy J, Felton G (2002). Herbivory: caterpillar saliva beats plant defences. Nature 416: 599-600. Petzold-Maxwell J, Cibilis-Stewart X, French B, Gassmann A (2012) Adaptation by Western corn rootworm (Coleoptera: Chrysomelidae) to Bt maize: inheritance, fitness costs, and feeding preference. J. Econ. Entomol. 105(4): 1407-1418. Rasmann S, Turlings T. (2008) First insights into specificity of belowground tritrophic interactions. Oikos 117(3): 362-369. Robert, C, Erb M, Duployer M, Zwahlen C, Doyen G, and T Turlings. 2012. Herbivore- induced plant volatiles mediate host selection by a root herbivore. New Phytologist 194(4): 1061-1069. Robert, C, Veyrat N, Glauser G, Marti G, Doyen G, Villard N, Gaillard M, Köllner T, Giron D, Body M, Babst B, Ferrieri R, Turlings T, and Erb M. 2012. A specialist root herbivore exploits defensive metabolites to locate nutritious tissues. Ecology Letters 15(1): 55-64. Rudeen M, Gassmann A (2012) Effects of Cry34/35Ab1 corn on the survival and development of western corn rootworm, Diabrotica virgifera virgifera. Pest Manage. Sci. 69(6): 709-716.

Sappington T, Siegfried B, Guillemaud T (2006) Coordinated Diabrotica genetics research: Accelerating progress on an urgent insect pest problem. American Entomologist 52(2): 90- 97. Skrivanek, S., B. Ripple, J. Lopez, and M. Harris. 2006. Effects of Bacillus thuringiensis kurstaki and sodium bicarbonate in Coleopteran and Lepidopteran larval diets. Southwestern Entomologist 31(1): 55-58. Storer N, Babcock J, Edwards J (2006) Field measures of western corn rootworm (Coleoptera: Chrysomelidae) mortality caused by Cry34/35Ab1 proteins expressed in maize event 59122 and implications for trait durability J. Econ. Entomol. 99(4): 1381-1387. Tabashnik B (2008) Delaying insect resistance to transgenic crops. Proc. Natl. Acad. Sci. U. S. A. 105(49): 19029-19030. Vaughn T, Cavato T, Brar G, Coombe T, DeGooyer T, Ford S, Groth M, Howe A, Johnson S, Kolacz K, Pilcher C, Purcell J, Romano C, English L, Pershing J. (2005) A method of controlling corn rootworm feeding using a Bacillus thuringiensis protein expressed in transgenic maize. Crop Science 45(3): 931-938. Williams M, Boydston R (2013) Intraspecific and interspecific competition in sweet corn. Agronomy Journal 105(2): 503-508. Yang F, Kerns D, Head G, Leonard B, Niu Y, Huang F (2014) Occurrence, distribution, and ear damage of Helicoverpa zea (Lepidoptera: Noctuidae) in mixed plantings of non-Bt and Bt corn containing Genuity® SmartStax™ traits. Crop Protection 55: 127-132. Yukui R, Wenya W, Pinghui L, Fusuo Z. (2009) Mineral element distribution in organs of dual-toxin transgenic (Bt+CpTI) cotton seedling. Plant Biosystems 143(1):137-139.

Figure Legends Figure 1: Experimental plexiglass arenas were constructed and were filled with a clear agar/MiracleGro growth medium to facilitate rootworm observations. Markings along the root denote assignment of root regions for insect behavior and toxin expression assessments. Figure 2: Block/strip (no choice) scenario. Mean number of 3rd instar WCR larvae in contact with corn root tissue per observation time, averaged over all observations in a 48- hour period for n=20. A) Number of larvae touching root tissue per segment per observation time on Bt and non-Bt corn roots. B) Number of larvae per observation time found in contact with different segments of root, from crown to tip. Numbers represent segment, with 1 representing the segment nearest the root crown and 4 representing the segment nearest the root tip. Different letters indicate significant difference at p<0.05. Figure 3: Seed mix (choice) scenario. A) Mean number of 3rd instar WCR larvae in contact with corn root tissue over all observations in a 48-hour period for n=20. B) Number of larvae found associated with different regions of the root, from crown to tip. Numbers represent region, with 1 representing the region nearest the root crown and 4 representing the region nearest the root tip. Different letters indicate significant difference at p<0.0001. Figure 4: Amount of Bt protein (ng/g dry weight) of root tissue. A) Cry34 and Cry35 protein in roots that were infested with 3rd instar WCR larvae vs. uninfested roots. B) Cry34 and Cry35 expression in corn roots grown next to non-Bt corn (choice) vs. corn roots grown alone (no choice) when agar arenas were used. * significant at p<0.05, ** significant at p<0.0001.

Figure 2: Block/strip (no choice) scenario. Mean number of 3rd instar WCR larvae in contact with corn root tissue per observation time, averaged over all observations in a 48-hour period for n=20. A) Number of larvae touching root tissue per segment per observation time on Bt and non-Bt corn roots. B) Number of larvae per observation time found in contact with different segments of root, from crown to tip. Numbers represent segment, with 1 representing the segment nearest the root crown and 4 representing the segment nearest the root tip. Different letters indicate significant difference at p<0.05.

Figure 3: Seed mix (choice) scenario. A) Mean number of 3rd instar WCR larvae in contact with corn root tissue over all observations in a 48-hour period for n=20. B) Number of larvae found associated with different regions of the root, from crown to tip. Numbers represent region, with 1 representing the region nearest the root crown and 4 representing the region nearest the root tip. Different letters indicate significant difference at p<0.0001.

Figure 4: Amount of Bt protein (ng/g dry weight) of root tissue. A) Cry34 and Cry35 protein in roots that were infested with 3rd instar WCR larvae vs. uninfested roots. B) Cry34 and Cry35 expression in corn roots grown next to non-Bt corn (choice) vs. corn roots grown alone (no choice) when agar arenas were used. * significant at p<0.05, ** significant at p<0.0001.

Chapter 3

Responses of Western corn rootworm, Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae), larvae to soil microbe seed treatments on Bt and non-Bt corn.

Introduction Corn rootworms, Diabrotica spp. (Coleoptera: Chrysomelidae) were estimated to cost around $2 billion in the US alone in management and crop loss in 2008 (Paul Mitchell, personal communication). The Western corn rootworm, Diabrotica virgifera virgifera LeConte (Coleoptera: Chrysomelidae, WCR), is particularly pernicious due to its ability to evolve counter adaptations to commonly used management approaches. For example, pesticide resistance in WCR was documented as early as 1959, both in soil-based insecticides that target larvae (Ball and Weekman 1962), and aerial insecticides that target adults (Meinke et al 1998). Western corn rootworm populations have also evolved behavioral resistance to cultural practices such as corn-soybean crop rotation, whereby females oviposit in soybean fields rather than in corn fields, so that the following season neonates will be more likely to be in corn (Onstad et al. 2001). Transgenic crops expressing various Bacillus thuringiensis (Bt) toxins are among the most modern approaches to manage WCR. However, some populations of WCR have already evolved resistance to Bt corn (Gassmann et al. 2011). The mechanism of this resistance is not currently known, though non-preference of Bt-expressing roots by rootworm larvae may contribute to resistance development by exposing larvae to sublethal doses of the toxin (Murphy et al. 2010). The US Environmental Protection Agency (EPA) requires that a refuge of non-Bt plants be planted along with any Bt crop as a part of insect resistance management (EPA 2014). Western corn rootworm larvae have been noted to congregate around non-Bt corn in fields planted with both Bt and non-Bt corn (Rudeen and Gassmann 2012), and Bt corn roots may be deterrent to rootworm larvae (Petzold-Maxwell et al. 2012). In seed mix refuges, transgenic and non-transgenic corn seeds are randomly distributed within a bag of corn seed, leading to a random distribution of non-Bt corn in the field (EPA 2010a), while in block/strip refuges, non-Bt corn is planted in a single strip away from Bt corn (EPA 2010b). Behavior by larval WCR may affect the speed of resistance development; non-preference behavior by WCR may affect the dose of toxin received, while a lack of non-preference behavior may change the effective refuge size by killing larvae as

they move through the soil (Head and Greenplate 2012). Thus, WCR behavior is worth investigating. Little is known of WCR larval behavior, because soil-dwelling insects such as WCR larvae are difficult to observe. Another aspect of WCR ecology that has received limited attention is their interactions with soil-borne microbes. Rootworms spend their entire immature life in the soil, surrounded by a plethora of microbial life, yet there is little documentation regarding microbial interactions with these insects. Adaptations to fungal pathogens are particularly important in soil-dwelling insects such as WCR because of the soil’s large naturally occurring fungal population. Soil-borne entomopathogens such as Metarhizium anisopliae (Metchnikoff) Sorokin (Hypocreales: Clavicitipitaceae) and Beauveria bassiana (Balsamo) Vuillemin (Hypocreales: Clavicitipitaceae) are known to increase mortality in Western corn rootworms independently of corn line (Petzold-Maxwell et al. 2011). However, Pilz et al. (2008) previously found very low rates of fungal infection in rootworms collected from Hungarian fields, despite finding up to 6,046 colony-forming units of fungi per gram of soil. Thus, it appears that rootworms are able to survive despite high rates of entomopathogens present in the soil. A better understanding of soil microbes and how they interact with corn and rootworms is necessary because soil microbes can perform important functions for crop plant health, such as improving nutrient availability and protecting plants against biotic and abiotic stresses (Ryan et al. 2009), perhaps even from rootworm feeding. Bacterial and fungal entomopathogens (Laird et al. 1990 a) are one category of soil-borne microbes that have attracted attention for their potential practical applications. Entomopathogens provide pest management potential because these pathogens are considered to be nonpathogenic to humans (Laird et al.1990 a), because they may have fewer effects on beneficial arthropods than traditional insecticides (Ivie et al. 2002), and because they are otherwise unlikely to have non-target environmental harm (Laird et al.1990 b). However, entomopathogens are still only a tiny fraction of the pesticide market, due in part to their inconsistent effects under field conditions (Copping and Menn 2000).

A thorough understanding of insect-microbe behavioral interactions in the soil may help us better understand soil insect mortality with respect to entomopathogens (Baverstock et al. 2009). The more attractive and/or less repellent an entomopathogen is, the more success it may have in the field, since this would increase the likelihood that insects would come in contact with it and become infected. Some insects have been documented to display non-preference responses to entomopathogens, perhaps as a result of repellent cues (Kabaluk and Ericsson 2007, Meyling and Pell 2006, Thompson et al. 2007), which may explain the frequently disappointing field results with respect to entomopathogens as a pest management approach. However, entomopathogenic fungi can in some cases attract insects (George et al. 2013, Kepler and Bruck 2006). The mechanisms for this attraction are often unknown, but some researchers have suggested that fungi manipulate their insect hosts; for example, Beauveria bassiana may be attractive to mosquitoes, Anopheles stephensi Liston (Diptera: Culicidae) to facilitate dispersal (George et al. 2013). In some instances, behavioral response is dependent on the life stage of the insect; Japanese beetle grubs, Popillia japonica Newman (Coleoptera: Scarabaeidae) are repelled by M. anisopliae, but adults are more likely to oviposit in microcosms inundated with M. anisopliae (Villani et al. 1994). It is also important to take into consideration that the effects of pathogen cues on arthropod behavior can change over time; Japanese beetle grubs are only repelled by the fungus Metarhizium anisopliae after they have been exposed for a minimum of 48 hours (Fry et al. 1997). Another category of soil-borne microbes that may have practical benefits (Lucy et al. 2004) are plant growth-promoting organisms (PGPO), which associate with plant root tissues and enhance plant growth as they actively form colonies on or around plant roots (Kloepper and Beauchamp 1992). There are multiple mechanisms by which PGPO positively affect plant health, including not only increased production of phytohormone-mimicking substances and enzymes and increased nutrient availability, but also increased protection of plants from parasites (Vacheron et al. 2013). These PGPO can also induce plant defenses against insect pests by increasing production of plants' defensive compounds; for instance, Pseudomonas spp., combined with the fungus Beauveria bassiana, can increase activity of phenylalanine ammonialyase and at least eight other defensive enzymes in peanut plants, Arachis hypogaea

L. (Fabales: Fabaceae), reducing their susceptibility to pest insects such as Aproaerema modicella Deventer (Lepidoptera: Gelechiidae) and fungus, Athelia rolfsii Curzi (Atheliales: Atheliaceae) (Senthilraja et al. 2013). Some PGPO also have entomopathogenic activity; for example, Bacillus subtilis Ehrenberg (Bacillales: Bacillaceae) kills Diacrisia obliqua Walk (Lepidoptera: Arctiidae) in laboratory assays whether applied alone or in conjunction with insecticides (Bhattacharya and Pramanik 2003) and antibiotic peptaibols produced by Trichoderma harzianum Rifai (Hypocreales: Hypocreaceae) are lethal to Tenebrio molitor L. (Coleoptera: Tenebrionidae) when applied to the cuticle or fed to larvae (Shakeri and Foster 2007). Interestingly, some entomopathogenic fungi and bacteria also have the capacity to function in PGPO-like fashion. An important entomopathogenic fungus, B. bassiana, is known to endophytically colonize root tissue and can help reduce pest damage (Gurulingappa et al. 2010). Another well-studied entomopathogenic fungus, M. anisopliae, is known to be persistent in the rhizosphere, in that it will persist for at least a year if inoculated into soil (Hu and St Leger 2002), while congeneric M. robertsii have been shown to colonize endophytically and promote root growth (Sasan and Bidochka 2012). Similarly, the entomopathogenic bacterium, Serratia marcescens Bizio (Enterobacteriales: Enterobacteriaceae), has been shown to colonize cucumber roots and have positive effects on the plants’ growth, yield, and disease resistance (Gül et al. 2013). Additionally, Prischmann et al. (2008) found Serratia spp. associated with both corn roots and WCR eggs, and found certain strains to be associated with infected (discolored but living) adults. The adaptability of WCR, leading to resistance development against common management practices, presents the greatest challenge for sustainable corn production around the world (Gould 1998). Therefore, it is important to assess sources of WCR mortality to better understand and exploit their defenses. For these reasons, a thorough examination of the relationships between WCR, Bt and non-Bt corn, and soil microbes would be valuable for future management of these pests. The first objective of this study was to assess WCR larval behavioral responses to corn seeds (both Bt and non-Bt) treated with selected species of beneficial soil-borne microbes including entomopathogens (M. anisopliae, B. bassiana, S.

marcescens) and PGPO (B. subtilis and T. harzianum). The second objective of this study was to assess whether larval responses to sprouts from microbe-treated seeds were due to cues from microbes directly or to cues emitted after the interaction between microbe and seeds. The third objective was to assess if and how such microbes might affect adult behavioral responses and oviposition preference. Finally, the fourth objective was to assess whether insects fed on plants produced from microbe-treated seeds would survive into adulthood and whether there were any differences in developmental time, weight or sex ratio, when emerging from mesocosms representing planting scenarios that model commonly used Bt refuge types. Methods and Materials Experimental system Insects and corn seeds. Colonies of WCR were established in our laboratory using non-diapausing Bt-susceptible stock originally obtained from the USDA-ARS facilities (Brookings Co., SD) and Crop Characteristics, Inc. (Farmington, MN). Rootworms used for these experiments were kept in colonies maintained at 25°C, 75 ± 10% relative humidity, and 16:8 h light:dark. No more than 200 adults were housed per cage in BugDorm mesh cages (MegaView, Taichung, Taiwan) and fed prepared adult WCR diet (Bio-Serv, Frenchtown, NJ), with water ad libitum. Eggs were collected three times per week on 9-cm diameter plastic disposable Petri dishes (Beckton, Dickinson and Company, Franklin Lakes, NJ) filled with 1% agar and overlayed with one sheet of filter paper and four layers of cheesecloth (Fisher Scientific, Pittsburgh, PA). After 10 days incubation, eggs were transferred to larval arenas consisting of 1100mL rectangular plastic take-out food containers (Versatainer, Leola, PA), provisioned with 60 mL of corn, that had been soaked and kept in moist conditions in a seed sprouter (NK Lawn and Garden, Chattanooga, TN) for 48 hours, covered with 300 mL topsoil (Scotts Miracle-Gro, Marysville, OH) and 80 mL tapwater. All seed sprouts, soil, and insect eggs were set up in half of the arena, leaving the other half of the arena empty for future soil and seed sprout provisioning. Containers were covered with satin lining fabric (Jo-Ann, Hudson, OH) to ensure that they were kept dark and humid. To ensure that insects had enough fresh food, additional soil, corn seeds, and water were provided after 7 days on

the half of the arena left empty upon initial set up. This was enough to provide sustenance for insects to complete their development to adulthood within 4-5 weeks. Adults were hand- collected and added to adult cages. A new adult cage was set up every two weeks. Corn seeds used for insect colony maintenance were organic, rootworm-susceptible Truckers Delight variety (Coor Farm Supply, Smithfield, NC). Corn seeds used for experimental assays described below were from a rootworm-active transgenic Bt-expressing line (Pioneer Herculex Xtra, 35F44, Bt) and a genetically similar non-transgenic susceptible line (Pioneer, 35F38, non-Bt). Microbial treatments. Seeds were treated with the following commercially available formulations of microbial treatments: 1) Companion ® (Bacillus subtilis, TGrowth Products, Whiteplains, NY); 2) Mycotrol ® (Beauveria bassiana, BioWorks, Victor, NY) 3) Met52 ® (Metarhizium anisopliae, Novozymes, Brookfield, WI) 4) RootShield ® (Trichoderma harzianum, BioWorks, Victor, NY). Label rates were scaled down to 10 mL water, resulting in 312 μL Companion, 2.5 mL Mycotrol, 6.46 g Met52, and 0.6 g RootShield. Since no commercial formulations of S. marcescens could be obtained, a S. marcescens strain isolated from nematodes and maintained in pure culture in our laboratory was used for the experiments. Bacteria were kept in culture on tryptic soy agar (Becton, Dickson, and Company, Franklin Lakes, NJ). For bacterial seed treatments, a liquid culture was produced by inoculating 5mL tryptic soy broth (Difco, Detroit, MI) in a 17 x 100mm test tube (Fisher Scientific, Waltham, MA) with bacteria using a 5mm wire smear loop (Fisher Scientific, Waltham, MA) and incubating the suspension on an orbital shaker (Barnstead Lab Line, Hampton, NH) set at maximum speed for 18 h. After this incubation period, cells were pelletized with a centrifuge (Eppendorf, Hamburg, Germany) for 1 minute at 6000 RPM, growth medium was removed, and pellets were re-suspended and adjusted to the desired

CFUs (as determined by dilution plating) with dd H2O. S. marcescens suspensions were applied at 1 × 107, which has been previously described as a useful concentration for determining susceptibility of insect larvae (Ruiu et al. 2007).

Larval behavioral response to beneficial soil-borne microbes applied to non-Bt and Bt seeds in test tube arenas. Larval responses to microbe-treated non-Bt and Bt corn seeds in dual-choice test tube assays. The purpose of this experiment was to assess preference of WCR larvae for corn seeds either soaked in water (control) or soaked in suspensions of commercially available microbe formulations and lab-grown S. marcescens cultures. Arenas used to evaluate larval responses to microbial seed treatments were constructed after Fry et al. (1997) and consisted of two 5mL plastic test tubes (Fisher Scientific, Waltham, MA). The ends of the test tubes were punctured with a hot dissection needle to facilitate air flow and then covered with organza mesh (Jo-Ann Stores, Hudson, OH) to prevent WCR from escaping (Fry et al. 1997). Test tube caps were modified by removing the ends and gluing two caps together, such that when test tubes were placed in each cap two tubes were confluent with each other and insects could move from one test tube to the other (Figure 1). Corn seeds, either Bt or non-Bt,were soaked for 48 hours in 20 mL aqueous suspensions of microbes following labeled rates for commercial formulations or in water without microbes (control). For S. marcescens, seeds were soaked in a bacterial suspension of 1 x 107 as described above under “Microbial treatments.” Two seeds soaked in the microbial species of interest were then placed in one side of the arena, and two seeds soaked in water were placed in the opposite side. Only one corn genotype was tested within a single arena. Each side of the arena was filled with topsoil (Scotts Miracle-Gro, Marysville, OH) to reach the 5 mL mark of each tube. The seeds and microbes were then allowed to remain at room temperature for 24 hours to allow seed/microbe cues to disperse through the arena prior to insect release. After this 24 hour period, five 3rd instar larvae were introduced into the center of each arena. Insects were allowed to respond for 24 hours, during which time arenas were kept in the dark, at 25°C, 75 ± 10% relative humidity. After this period, number of larvae in contact with treatments at either side of the arena were counted and recorded. Experiments were run in single replicates of all microbe species/corn line combinations and

repeated over time with new microbial suspensions, seeds and insects to obtain a total of 20 replicates per seed/microbe combination. Larval behavioral response to microbe-treated non-Bt versus Bt seeds in dual-choice test tube arenas. In a separate but concurrent experiment, we assessed the larval responses to non-Bt and Bt corn seeds treated with the same microbe species in the same test tube arena. Arenas used to evaluate larval responses to microbial seed treatments were constructed out of two 5mL plastic test tubes as described previously. Corn seeds, Bt or non-Bt, were soaked for 48 hours in 20 mL aqueous suspensions of microbes following labeled rates for commercial formulations. For Serratia marcescens, seeds were soaked in a bacterial suspension of 1 x 107 as described under “Microbial treatments.” Two Bt seeds soaked in the microbial species of interest were then placed in one side of the arena, and two non-Bt seeds soaked in the microbial species of interest were placed in the opposite side of the same arena. Seeds were covered with topsoil and allowed to remain for 24 hours as described above to allow seed/microbe cues to disperse through the arena prior to insect release. After this acclimatization period, five 3rd instar larvae were introduced into the center of each arena and allowed 24 hours to respond to cues. After this period, number of larvae in contact with treatments at either side of the arena were counted and recorded. Experiments were run in single replicates of all microbe species and the experiment was repeated over time with new microbial suspensions, seeds and insects to obtain a total of 20 replicates for each microbe species. Larval response to interaction between seeds and microbes in dual-choice Petri dish arenas. The purpose of this assay was to determine if any insect behavioral responses to microbial seed treatment observed in the previous assay might be mediated by cues deriving from the microbes directly or by cues derived from the interaction of the seeds with the microbes. Microbial formulations treatments and environmental conditions were as described for the previous assay. For these purposes we observed behavioral responses of individual WCR 3rd instar larvae to non-Bt or Bt corn seeds either a) seeds soaked for 48 hours in an

aqueous microbial suspension or b) seeds soaked in water for 48 hours, then dipped for 10 seconds in the microbial suspension of interest immediately preceding bioassays. This would ensure that seeds from our second treatment were inoculated with microbes, but that the microbes did not have as much time to interact with the seeds and provide novel cues. Each arena consisted of one 9-cm diameter plastic disposable Petri dish (Beckton, Dickinson and Company, Franklin Lakes, NJ) filled with 1% agar to serve as a humid surface for seeds and insects for the assay duration. Two 10-mm diameter round plugs of the agar slab were cut out on opposite sides of the Petri dish to provide a niche to secure the corn seeds. One 10 s dipped corn seed was placed in one niche and one corn seed soaked in the same microbial suspension for 48 h was placed in the opposite niche. All microbe species in combination with either Bt or non-Bt corn seeds were tested concurrently but in separate Petri dish arenas (Figure 2). Insect behavior was monitored after releasing a single 3rd-instar WCR larva in the center of the arena. All behavioral assessments were conducted in a dark room, under red light to minimize insect disturbance. Larval movement was continuously observed and recorded by tracing onto clear acetate film (Canon, Lake Success, NY) with a fine-point permanent marker (Sharpie, Downers Grove, IL) for 15 minutes. The acetate film was bisected twice to give four quadrants: one centered around the soaked corn seed, one centered around the dipped corn seed, and two neutral quadrants that did not contain corn seeds. The number of times a larva crossed over its own path (crossovers) was recorded, and the location of the larva (near soaked corn, near dipped corn, or neutral area) at the time of crossover was also noted. The time the larva took to first contact a corn seed within the arena and the treatment first contacted was recorded. The location of the larva was recorded again at 30 min, and again at 6 and 24 h. A single replicate of both types of seeds treated with each of the microbial species was conducted on the same day. To make sure all seed/microbe treatment assays were completed within a 2 h period, up to four researchers assisted with data collection. Treatments were randomly assigned to each data collector on each replicate date, and a total of 20 replicates per seed/microbe species/time treatment combination were obtained.

Effect of microbial seed treatments on adult oviposition preference This experiment was designed to determine whether microbe seed treatments can be detected by female WCR and if any measurable behavioral response and adult oviposition preference can be detected. We performed choice experiments in which mated 14-28 day old gravid females were given oviposition arenas with and without our test microbes. To evaluate oviposition preference, a nested design with four microbial treatments nested within corn seed type (Bt or non-Bt) was used. The treatments evaluated were corn seed, either Bt or non-Bt, treated with 1) Metarhizium anisopliae (Ma), 2) Beauveria bassiana (Bb), 3) Bacillus subtilis (Bs), and 4) sterile distilled water (control). These microbial species were chosen because our earlier experiments showed that these species had significant effects on larval behavior, and because they are commercially available. Microbial treatments and concentrations were the same as used in our previous experiment. Experimental arenas were 30 cm square BugDorms (MegaView, Taichung, Taiwan) with four oviposition arenas (same as were described previously for egg collection): 1) control, seeds soaked in water only; 2) Seeds soaked in B. bassiana suspension; 3) Seeds soaked in M. anisopliae suspension; and 4) Seeds soaked in B. subtilis suspension. Oviposition arenas consisted of plastic petri dishes (90x15 mm) with 50 ml of 1% agar and one layer of Nº1 sterile filter paper (Whatman, Kent, England), over which 1.6 g of dried Scotts MiracleGro Topsoil® was placed after having been sieved through an 80-mesh screen (Tyler, Cleveland, OH). Soil was used to mimic rootworms’ natural oviposition substrate (Pruess et al. 1968), while agar was used to keep the soil moist. Microbial treatments were presented in the form of two seeds soaked in the treatment of interest for 48 hours. The two seeds were placed into small plastic containers (4x4 mm) made from the lids of 5mL plastic test tubes (Fisher Scientific, Waltham, MA) with wet sterile cotton wool to prevent seed desiccation. The tops of individual seed containers were covered with a small piece of organza fabric (Jo-Ann Stores, Hudson, OH) held snugly around the tube caps by two small rubber bands to prevent insects from directly contacting the seeds, while allowing potential seed/microbe volatile cues to permeate out of the plastic containers and into the oviposition arenas. Each oviposition arena was loosely covered with an aluminum foil

dome to provide a dark environment for female oviposition and to confine microbe/seed cues to individual arenas. Insects could access the oviposition arenas by crawling under the foil dome edges. Each container was placed in the center of an oviposition arena; thus, there were eight experimental units if both Bt and non-Bt corn are counted separately (Figure 1). Four male and eight female WCR were released per cage. The number of adults within each oviposition arena was counted two times daily, at 9:00 and 17:00, and averaged to obtain a daily treatment value for three consecutive days. At the end of the final day, the number of eggs in each oviposition dish was counted. Fourteen replicates were obtained. Olfactory response of gravid females to microbe-treated corn seeds The purpose of this experiment was to determine whether the behavioral responses and oviposition preference by mated, gravid female WCR are modulated by volatile emissions of seeds treated with the various beneficial soil microorganisms used in the previous assay. To test this, we conducted experiments using a Y-tube olfactometer (Figure 3). In each paired comparison test, five mated, gravid (14-28 day old) WCR females were exposed to volatile emissions from two odor sources: 1) corn seeds soaked for 48 h in a single-species microbial aqueous suspension and 2) corn seeds soaked in water for 48 hours (control). Microbe treatments were tested with both Bt and non-Bt corn seeds, and each seed type was tested on the same day but in separate assays. Microbial concentration was the same as described above. Assays were performed at 17:00-19:00, because preliminary experiments showed the greatest number of insect responses during this time period. Odor sources were provided by sets of five seeds placed on a small piece of aluminum foil lined with moistened sterile paper towel to prevent seeds from drying out. The duration of each assay was 20 minutes (five minutes of acclimatization to the environmental conditions within the olfactometer and 15 minutes to respond to treatments). One replicate of each seed type (Bt or non-Bt) and microbe (M. anisopliae or B. bassiana) combination were obtained each day using new insects and seeds for each assay. Bacillus subtilis was not included in this assay because of the nonsignificant results obtained with this species in the oviposition preference assays. The

experiment was repeated over time to obtain eleven replicates per seed/microbe combination. Location of treatments was rotated between the two arms of the olfactometer and treatment sequence was randomized for each seed/microbe replicate obtained to minimize potential environmental/mechanical biases on insect response. Variables measured included number of insects responding to either of the odor sources and the time it took insects to make a first choice. If the insects did not respond within the allotted 15 minutes, results were not taken into account for statistical analysis. This yielded between six and eleven replicates per treatment to be included in the statistical analyses. Insect survival, development, weight, and sex ratio under greenhouse mesocosms. The purpose of this assay was to determine whether selected microbe species used as seed treatment in our laboratory assays had an effect on insect survival, sex, weight, and time to emergence. This experiment included 5 microbial treatments (no microbe control, B. bassiana, M. anisopliae, T. harzianum, and S. marcescens) and 2 planting scenarios containing Bt/non-Bt seed mixtures. These planting scenarios were intended to mimic block/strip and seed mix refuge scenarios (80/20% for block/strip or 90/10% for seed mix), with non-Bt seeds planted in a row next to Bt seeds for the block/strip scenario and interspersed throughout the flat using a random number generator for the seed mix scenario. These ratios were used because they were being considered by the EPA at that time (EPA 2009). Seeds were planted in seedling flats, 38 x 51 x 10 cm, (Kadon, Dayton, OH) filled with potting soil (Fafard 2 Mix, Agawar, MA) and planted with Bt/non-Bt corn seed combinations as specified for each treatment. All flats received 30 plantings of three seeds each, in 3 evenly spaced rows of 10 plantings each. Seedlings were thinned to one per planting 4 d after emergence. In addition, a control flat, planted entirely with non-Bt corn without any microbial treatment, was included to serve as a reference for insect performance in the absence of Bt corn or added microbes. Microbial products used were the same as described previously for larval experiments. Corn seeds were soaked for 48 hours in microbial suspensions at label rates, or as specifically described above for S. marcescens. Corn seeds for non-Bt control flats were soaked for 48 hours in water. Individual flats were placed within 80 x 80 x 40 cm cages

constructed of PVC pipe frames (Lowe’s Companies, Inc., Mooresville, NC) covered with custom-made organza mesh (Jo-Ann Stores, Hudson, OH) sleeves to confine insects to individual treatments and to contain adults emerging from individual arenas. Experimental flats were watered as needed to maintain plant turgor. Three days after thinning corn flats, 250 10-day-old (near hatching) WCR eggs in 0.1% agar slur were added to the soil with a disposable 1mL plastic pipette (Fisher Scientific, Pittsburgh, PA) in five evenly spaces sites of 50 eggs each per site, which were then lightly covered with soil. Number, sex, and weight of adult insects emerging from each flat were recorded, in addition to time to emergence. The experiment was set up in singlets or duplicates and repeated over time to obtain six replicates for each refuge planting/microbe species combination. Each replicate was terminated when no adult insects emerged for 7 consecutive days. Statistical Analyses. For test tube behavioral response assays, analysis of variance (ANOVA) (PROC GLM, SAS Institute, Cary, NC, 2011) was used to determine effects of corn line, microbial species, and their interaction on number of WCR larvae found in the treated or untreated side of the test tube, for each replication Ho = πmicrobe = πwater where πx is the proportion difference between the microbe treatment and water (proportion of individuals out of five that selected either the microbe treatment or water). Factors found significant were followed by Tukey’s mean separation tests (P≤0.05). Data from Petri dish behavioral response assays were also analyzed via ANOVA (PROC GLM, SAS Institute, Cary, NC, 2011) to assess effects of time, corn line, microbial species, and their interactions on number of larvae found on 48-hour soaked or 10 second dipped corn seeds, for each replication Ho = πBt = πnon-Bt where πseed is the proportion difference between the Bt treatment and the non-Bt treatment (proportion of individuals out of five that selected either the Bt or non-Bt treatment). Factors that were found significant were followed by Tukey's mean separation tests (P≤0.05). For adult oviposition assays, comparison between microbial species was done using PROC GLIMMIX (SAS Institute, Cary, NC, 2011) on the proportion difference between M.

anisopliae and water and between Beauveria and water, calculated for each replication Ho =

πMetarhizium = πBeauveria where πmicrobe is the proportion difference between the microbe treatment and water (proportion of individuals out of three that selected either the microbe treatment or water), because the large amount of variation resulted in non-normality. This test was run separately for each seed type (Bt and non-Bt). For adult olfactometer assays, comparison was done using PROC TTEST because data were normal but variances were unequal between different treatments and many insects did not make a choice. Data from replicates where insects did not make a choice were not included in the analysis, and the choice was modeled as a function of time, replicate, seed, and treatment. Data from greenhouse mesocosm assays were subjected to ANOVA to determine the effects of microbial species and refuge scenario on the number, sex ratio, time until emergence, and weight of adult insects emerged from each mesocosm (Proc GLM, SAS Institute, Cary, NC, 2011). Factors that were found significant were followed by Tukey's mean separation tests (P≤0.05). Results Larval behavioral responses to corn seeds in our test tube arenas were not significantly different among microbe species tested. The interaction of microbe species with corn line was also not significant. However, seed type had a significant effect on insect choice of microbe treated vs untreated seeds, with insects significantly more likely to contact sprouted seeds that were treated with microbes, regardless of microbe species, when Bt corn seeds were used (F=9.83; df=1, 23; p<0.0021) compared to when non-Bt corn seeds were used (Figure 4A). Interestingly, for insect behavioral response to Bt versus non-Bt corn seeds in our test tube arenas, there was no non-preference of Bt seeds when both sides of the test tube were treated with microbes (F=1.92; df=1, 32; p=0.168) (Figure 4B). For the interaction between seeds and microbes in our petri dish arenas, only M. anisopliae yielded any significant difference in insect responses based on the length of time the microbe was allowed to contact the seed (10 s vs 48 h). Here, the number of insects responding to 48-hour M. anisopliae-treated corn increased over time compared to the 10 s dipped seed when Bt corn was used, but decreased over time when non-Bt corn was used.

Contrastingly, the number of insects responding to the 48 h M. anisopliae-soaked corn decreased over time compared to the 10 s dipped non-Bt corn treatments. There were no significant differences in number of crossovers, time to first contact, or position at any point of time for any other microbe species. Corn line did not have a significant impact on number of female WCR observed within oviposition arenas, regardless of microbe treatment used. Significant differences in number of females within oviposition arenas were only found for the 17:00 observation time but not for the 9:00 observation time (F=1.86, p=0.1402) (data not shown). Number of insects observed within oviposition arenas at 17:00 significantly varied in response to microbe species used to treat corn seeds (F=7.73, p<0.0001) (Figure 5A). The highest number of females was observed within arenas containing M. anisopliae-treated seeds. This was significantly different compared to all other treatments (Figure 5A). Number of females in oviposition arenas containing B. bassiana treated seeds was intermediate and only significantly higher than number of females in arenas containing seeds treated with B. subtilis, which had the lowest overall number of females visiting arenas (Fig 5 A). Effects of microbe (χ2=33.24, df=3, p<0.0001) and the interaction between microbe and corn line (χ2=22.00, df=3, p<0.0001), but not corn line alone (χ2=0.86, df=1, p=0.3524) were significant for the number of eggs laid (Figures 5B and 5C). Significantly fewer eggs were laid on B. subtilis than M. anisopliae or B. bassiana regardless of corn line. Significantly more eggs were laid on B. bassiana-treated corn when Bt corn was offered (Figure 5B); whereas, when non-Bt corn was offered, the difference between eggs on B. bassiana-treated and M. anisopliae-treated corn was not significant, although there were slightly more eggs on M. anisopliae-treated corn (Figure 5C). Olfactory responses by adults were tested and we found that 65 of 132 gravid females tested during this experiment chose any of the odor sources presented, yielding a response of 47%. Of the 65 females that responded, 41 (63%) chose the odor plume from corn seeds treated with microbes (Figure 6A). Insects presented with Bt corn seed were more likely to make a choice (χ2=8.1143, df=1, p=0.0044). The effects of replicate and microbial species treatment were not found to be significant (data not shown). Time lapsed before a choice was made was

recorded; however, effects of microbial treatment on time lapse were not significant. For choice, the only significant effect was that of replicate (χ2=23.5, df=10, p=0.0090) (data not shown). For insect survival in our greenhouse mesocosms, the effects of microbe (F=3.97, df=5, p=0.0031) and planting scenario (F=27.32, df=2, p<0.0001) on number of emerging adults were significant (Figure 7). All microbial treatments (Figures 7A, 7B) and planting scenarios (Figures 7A, 7C) resulted in fewer adults than the untreated control mesocosms. The fewest adult insects emerged from T. harzianum and B. bassiana (Figure 7B). Planting scenario was not significantly different between different microbial treatments, only between non-Bt control plots and treated plots (Figure 7C). There were no effects of replicate, microbe, planting scenario, or any interactions on sex or weight of adults. Microbe treatment effects were not significant, regardless of insect species, under the block/strip or seed mix refuge planting scenarios, when compared to their respective planting scenario with our microbe treatment. Insects emerged from non-Bt flats a mean of 11.57 (±4.69) days earlier than from Bt flats. With the exception of non-Bt flats, there were no significant effects of replicate, microbe, planting scenario, or any interaction on time to emergence. Discussion Through the course of this study, we evaluated potential effects of corn genotype and microbe-seed interactions for their effects on WCR behavior and survival. Larval WCR show non-prefrence for corn treated with these soil microbes under many circumstances. Often, WCR response to soil microbes was not significantly different regardless of the microbe species used. Although WCR show non-preference for Bt corn, presence of Bt corn appears to confound WCR’s non-preference for soil microbes. For all species except M. anisopliae, this non-preference for microbe-treated corn appears to be driven by the microbes themselves, rather than by an interaction between the microbes and corn plants. Adult WCR showed preference for M. anisopliae-treated corn in olfactometer assays. Treatment of corn seed with any of the species of soil microbes used in this study reduced survival to adulthood by WCR feeding on the subsequent corn plants, under greenhouse mesocosms.

Behavioral non-preference of Bt corn has been shown in WCR (Petzold-Maxwell et al. 2012, Rudeen and Gassmann 2012). Thus it is interesting that our results show significantly more WCR larvae found in contact with microbe-treated seeds when Bt seed was offered than when non-Bt seed was offered in our test tube (attraction/non-preference) assays. Compared to other microbial species tested, M. anisopliae appeared to have non- preference effects on WCR. This is not surprising given that it is an entomopathogen that is often present in soil samples; thus, M. anisopliae non-preference may have an evolutionary survival advantage for WCR. A non-detectable entomopathogen would have more potential for infecting insects. However, non-preference by WCR for M. anisopliae was only observed when M. anisopliae was used to treat non-Bt corn seeds. In previous research, WCR damage to non-Bt corn was lower when entomopathogens were introduced regardless of pest pressure, but WCR damage to Bt corn was only reduced when pest pressure was high (Petzold-Maxwell et al. 2013). This may indicate that WCR was less likely to avoid Bt when entomopathogens were present, although this effect is reduced when WCR abundance was high. When we tested WCR response to different corn lines (Bt or non-Bt) in test tubes where both sides contained seeds treated with soil microbes, our hypothesis about Bt interfering with WCR non-preference of M. anisopliae was further supported. Previous research (Petzold-Maxwell et al. 2012, Rudeen and Gassmann 2012) has shown non-preference of Bt by WCR, and our earlier assays showed non-preference of M. anisopliae by WCR. Insects did not avoid Bt seeds in arenas treated with M. anisopliae or, indeed, any of the soil microbes tested. Therefore, if soil microbes can interfere with Bt non-preference by WCR (i.e., seed treatment with M. anisopliae) this may be worth investigating to learn how M. anisopliae may affect WCR survival. Because we did not see Bt non-preference behavior when WCR larvae were offered microbe-treated seeds, WCR’s detection and non-preference of Bt-expressing corn lines may be mitigated by soil-borne microbes. Based on our greenhouse assays, seed treatment with soil microbes has an effect on insect survival to adulthood. We have also seen that some microbial species have significant effects on insect behavior in a variety of our assays; thus,

we would like to suggest that future research further investigate the effects of soil microbes, particularly M. anisopliae, on WCR survival under Bt field conditions. Only M. anisopliae was significantly different with respect to corn line in our Petri dish assays. Insects were more likely to contact dipped than soaked corn in non-Bt corn, with this trend reversed in Bt corn. This reversal in trend is yet another indication that Bt-microbe interactions may interfere with WCR’s sensory ability. The fact that WCR’s response to M. anisopliae -treated seed changes over time may indicate that the effects seen are caused by an interaction between the corn plant and M. anisopliae, which is known to colonize root tissue in a PGPO-like fashion (Liao et al. 2014). The change over time indicates that M. anisopliae’s effect may in fact be an effect of the interaction between M. anisopliae and the corn plant. If M. anisopliae is behaving as a PGPO and colonizing the root tissue in a way that is beneficial for the corn plant, this may indicate that M. anisopliae may affect WCR survival. Field-collected strains of M. anisopliae have proven more effective in controlling WCR than commercially available strains (Rudeen et al. 2013). However, it is important to note that the nature of our methodology precluded ensuring that there were equal numbers of spores on each treatment; thus, it is possible that any effects seen were due to different rates of inoculum imbibing into the seeds. The fact that there was no significant difference in larval response to corn sprouts from seeds treated with S. marcescens, B. subtilis, T. harzianum, or B. bassiana in the petri dish assays leads to the conclusion that the insects were responding to cues emanating from the microbes themselves rather than an interaction between the plants and the microbes. In our adult behavior assays, once again M. anisopliae was the only microbe that was significantly different from the control. Here, though, M. anisopliae treatments to corn seeds resulted in higher oviposition rates and greater attraction of gravid females. This means gravid females may be more likely to feed or oviposit on M. anisopliae -harboring plants. Further research would need to be done to elucidate whether this may affect the survival rates of the next generation of WCR. Once again, this effect was negated when Bt corn seed was used, indicating that Bt and soil microbes may interact in such a way as to disrupt insect discrimination processes.

When seed treatments with soil microbes were studied in greenhouse mesocosms, significantly more insects emerged from untreated flats. There appeared to be a trend towards more insects emerging from seed mix refuges than from block/strip refuges, although this trend was not significant. Seed mix refuges have been available as an alternative to block/strip refuges since 2008 (Onstad et al. 2011). With seed mix refuges, there is reduced time (Murphy et al. 2010) and reduced space (Onstad et al. 2011) between adults that emerge from refuge and non-refuge corn, increasing the ability of resistant adults (from non-refuge corn) to find susceptible mates (from refuge corn), and thereby minimizing numbers of homozygous-resistant WCR offspring. Other benefits of seed mix refugia include increased ease of use and increased farmer compliance (Onstad et al. 2011). However, there was a trend towards a larger percentage of adults emerging from seed mix refuges. Although seed mix strategies offer many benefits, if this is also true in the field, seed mixes may be less effective in killing WCR larvae, which must be taken into account when treatment strategies are considered. This trend was not significant, however, and it is worth noting that plants in our greenhouse flats were more crowded than corn would normally be planted in field conditions (Lauer and Rankin 2004). Overall, WCR’s response to beneficial microbe-treated seed was dependent on the type of corn line (Bt or non-Bt) offered. It has been shown that WCR avoid Bt corn (Petzold- Maxwell et al. 2012, Rudeen and Gassmann 2012); thus, the fact that non-Bt corn is in close proximity to Bt corn and larvae can move from plant to plant may explain why more insects emerged from our seed mix refuge flats than from our block/strip refuge flats. However, because no true isoline of the Bt seed exists, it is possible that some other differences between our Bt and non-Bt hybrid caused this effect, and again we must stress that corn plants in our greenhouse mesocosms were planted more densely than corn plants in the field (Lauer and Rankin 2004). We believe that results from our studies will be valuable because WCR interactions with soil microbiota are poorly understood. We found that WCR larvae and adults behaviorally respond to soil microbes. We also found that microbes interfere with WCR’s non-preference of Bt corn seed. We have identified M. anisopliae as a species of interest

because of its consistent effects on WCR behavior and T. harzianum as species of interest because survival of WCR was lowest on Trichoderma-treated flats. We found that treatments with soil microbes do reduce the survival of WCR in greenhouse mesocosms. Prior research has shown that soil microbes may influence rootworm damage to corn plants (Petzold-Maxwell et al. 2013) and our current study has indicated that they may also influence rootworm behavioral response to Bt corn seeds; however, dose responses and the mechanism of attraction/non-preference behaviors are not yet known.

References

Baverstock, J., H. Roy, and J. Pell. 2010. Entomopathogenic fungi and insect behaviour: from unsuspecting hosts to targeted vectors. BioControl 55(1): 89-102. Berdegué and Trumble. 1997. Interaction between linear furanocoumarins found in celery and a commercial Bacillus thuringiensis formulation on Spodoptera exigua (Lepidoptera: Noctuidae) larval feeding behavior. Journal of Economic Entomology 90(4): 961-966. Bhattacharya, S., and A. Pramanik. 2003. In vitro sensitivity and bio-efficacy of Bacillus subtilis as entomopathogen against lepidopteran larvae, Diacrisia obliqua. Crop Research (Hisar): 25 (1): 119-126. Ciosi, M., S. Toepfer, H. Li, et al. 2009. European populations of Diabrotica virgifera virgifera are resistant to aldrin, but not to methyl-parathion. Journal of Applied Entomology 133(4): 307-314. Clark P, Vaughn T, Meinke L, Molina-Ochoa J, Foster J (2006) Diabrotica virgifera virgifera (Coleoptera: Chrysomelidae) larval feeding behavior on transgenic maize (MON 863) and its isoline. J. Econ. Entomol. 99(3): 722-727. Copping, L., and J. Menn. 2000. Biopesticides: a review of their action, applications, and efficacy. Pest Management Science 56(8): 651-676. [EPA] Environmental Protection Agency (2005) EPA annual pesticide report. Available from http://www.epa.gov/oppfead1/annual/2005/05annualrpt.pdf [EPA] Environmental Protection Agency (2009) FIFRA scientific advisory panel open meeting. Available from http://www.epa.gov/scipoly/sap/meetings/2009/february/022309agenda.pdf [EPA] Environmental Protection Agency (2010) Scientific Issues Related to Insect Resistance Management for SmartStax™ Refuge-in-the-Bag, a Plant-Incorporated Protectant (PIP) Corn Seed Blend. Available from http://www.epa.gov/scipoly/sap/meetings/2010/december/120810agenda.pdf [EPA] Environmental Protection Agency (2010). Simulation models evaluation of pest resistance development to refuge in the bag concepts related to Pioneer submission.

Available from http://nepis.epa.gov/Adobe/PDF/P100EBX2.pdf [EPA] Environmental Protection Agency (2014) Insect resistance management fact sheet for Bacillus thuringiensis (Bt) corn products. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_corn_refuge_2006.htm Fry, R. 1997. Radiographic study of the response of Japanese beetle larvae (Coleoptera: Scarabaeidae) to soil-incorporated mycelia particles of Metarhizium anisopliae (Deuteromycetes). Journal of the New York Entomological Society 105(1-2): 113-120. Gassmann A, Petzold-Maxwell J, Keweshan R, Dunbar M (2011) Field-evolved resistance to Bt maize by Western corn rootworm. PLoS one 6(7): e22629. George J, Jenkins N, Blanford S, Thomas M, Baker T (2013) Malaria mosquitoes attracted by fatal fungus. PLoS One 8(5): 1-8. Gould F. (1998) Sustainability of transgenic insecticidal cultivars: Integrating pest genetics and ecology. Annu. Rev. Entomol. 43: 701-726. Gül, A., H. Özaktan, et al. (2013). "Rhizobacteria promoted yield of cucumber plants grown in perlite under Fusarium wilt stress." Scientia Horticulturae 153(0): 22-25. Gurulingappa, P., G. Sword, G. Murdoch, and P. McGee. Colonization of crop plants by fungal entomopathogens and their effects on two insect pests when in planta. Biological Control 55(1): 34-41. Head, G., and J. Greenplate. 2012. The design and implementation of insect resistance management programs for Bt crops. GM Crops and Food: Biotechnology in Agriculture and the Food Chain 3(3): 144-153. Hu, G., and J. St Leger. 2002. Field studies using a recombinant mycoinsecticide (Metarhizium anisopliae) reveal that it is rhizosphere competent. Applied and Environmental Microbiology 68(12): 6383-6387. Ivie, M., D. Pollock, D. Gustafson, J. Rasolomandimby, L. Ivie, and W. Swearingen. Field- based evaluation of biopesticide impacts on native biodiversity: Malagasy Coleoptera and anti-locust entomopathogenic fungi. Journal of Economic Entomology 95(4): 651-660.

Kabaluk, J., and J. Ericsson. 2007. Environmental and behavioral constraints on the infection of wireworms by Metarhizium anisopliae. Environmental Entomology 36(6): 1415-1420. Kepler, R., and D. Bruck. 2006. Examination of the interaction between the black vine weevil (Coleoptera: Curculionidae) and an entomopathogenic fungus reveals a new tritrophic interaction. Environmental Entomology 35(4): 1021-1029. Kloepper, J., and C. Beauchamp. 1992. A review of issues related to measuring colonization of plant roots by bacteria. Canadian Journal of Microbiology 38(12): 1219-1232. Laird, M., L. Lacey, and E. Davidson, Eds. 1990. Safety of Microbial Insecticides; CRC Press: Boca Raton, FL. pp. 101-113. Laird, M. 1990. Safety of Microbial Insecticides; CRC Press: Boca Raton, FL. pp. 243-247. Lauer, J., and M. Rankin. 2004. Corn response to within row plant spacing variation. Agronomy Journal 96(5): 1464-1468. Liao, X., T. O’Brien, W. Fang, and R. St. Leger. 2014. The plant beneficial effects of Metarhizium species correlate with their association with roots. Applied Microbiology and Biotechnology 98(16): 7089-7096. Lucy, M., E. Reed, and B. Glick. 2004. Applications of free living plant growth-promoting rhizobacteria. Antoine van Leeuwenhoek International Journal of General and Molecular Microbiology 86(1): 1-25. Meinke L, Siegfried B, Wright R, Chandler L (1998) Adult susceptibility of Nebraska western corn rootworm (Coleoptera: Chrysomelidae) populations to selected insecticides. J. Econ. Entomol. 91(3): 594-600. Meyling, N., and J. Pell. 2006. Detection and avoidance of an entomopathogenic fungus by a generalist insect predator. Ecological Entomology 31(2): 162-171. Murphy, A., M. Ginzel, and C. Krupke. 2010. Evaluating Western corn rootworm (Coleoptera: Chrysomelidae) emergence and root damage in a seed mix refuge. Journal of Economic Entomology 103(1): 147-157. Onstad, D., P. Mitchell, T. Hurley, J. Lundgren, R. Porter, C. Krupke, J. Spencer, C. Difonzo, T. Baute, R. Hellmich, L. Buschman, W. Hutchison, and J. Tooker. 2011. Seeds of change:

corn seed mixtures for resistance management and integrated pest management. Journal of Economic Entomology 104(2): 343-352. Onstad D, Spencer J, Guse C, Levine E, Isard S (2001) Modeling evolution of behavioral resistance by an insect to crop rotation. Entomologia Experimentalis et Applicata 100(2): 195-201. Petzold-Maxwell, J., Jaronski, S., Clifton, E., Dunbar, M., Jackson, M., and Gassmann, A. 2013. Interactions among Bt maize, entomopathogens, and rootworm species (Coleoptera: Chrysomelidae) in the field: Effects on survival, yield, and root injury. Journal of Economic Entomology 106(2): 622-632. Petzold-Maxwell J, Cibilis-Stewart X, French B, Gassmann A (2012) Adaptation by Western corn rootworm (Coleoptera: Chrysomelidae) to Bt maize: inheritance, fitness costs, and feeding preference. J. Econ. Entomol. 105(4): 1407-1418. Petzold-Maxwell, J., T. Jaronski, and A. Gassmann. 2011. Tritrophic interactions among Bt maize, an insect pest, and entomopathogens: effects on development and survival of western corn rootworm. Annals of Applied Biology 160(1): 43-55. Pilz, C., R. Wegensteiner, and S. Keller. 2008. Natural occurrence of insect pathogenic fungi and insect parasitic nematodes in Diabrotica virgifera virgifera populations. Biocontrol 53(2): 353-359. Prischmann, D., R. Lehman, A. Christie, and K. Dashiell. 2008. Characterization of bacteria isolated from maize roots: Emphasis on Serratia and infestation with corn rootworms (Chrysomelidae: Diabrotica). Applied Soil Ecology 40(3): 417-431. Pruess, K., G. Weekman, and B. Somerhalder. 1968. Western corn rootworm egg distribution and adult emergence under two corn tillage systems. Journal of Economic Entomology 61(5): 1424-1427. Rudeen M, Gassmann A. 2012. Effects of Cry34/35Ab1 corn on the survival and development of western corn rootworm, Diabrotica virgifera virgifera. Pest Manage. Sci. 69(6): 709-716.

Rudeen, M., Jaronski, S., Petzold-Maxwell, J., and Gassmann, A. 2013. Entomopathogenic fungi in cornfields and their potential to manage larval western corn rootworm Diabrotica virgifera virgifera. Journal of Invertebrate Pathology 114(3): 329-332. Ruiu L., A. Satta, and I. Floris. 2007. Susceptibility of the house fly pupal parasitoid Muscidifurax raptor (Hymenoptera: Pteromalidae) to the entomopathogenic bacteria Bacillus thuringiensis and Brevibacillus laterosporus. Biological Control 43(2): 188-194. Ryan, P., Y. Dessaux, L. Thomashow, and D. Weller. 2009. Rhizosphere engineering and management for sustainable agriculture. Plant and Soil 321(1-2): 363-383. Sappington T, Siegfried B, Guillemaud T (2006) Coordinated Diabrotica genetics research: Accelerating progress on an urgent insect pest problem. American Entomologist 52(2): 90- 97. Sasan, R., and M. Bidochka. 2012. The insect-pathogenic fungus Metarhizium robertsii (Clavicipitaceae) is also an endophyte that stimulates plant root development. American Journal of Botany 99(1): 101-107. Senthilraja, G., T. Anand, J. Kennedy, T. Raguchander, and R. Samiyappan. 2013. Plant growth promoting rhizobacteria (PGPR) and entomopathogenic fungus bioformulation enhance the expression of defense enzymes and pathogenesis-related proteins in groundnut plants against leafminer insect and collar rot pathogen. Physiological and Molecular Plant Pathology 82: 10-19. Shakeri, J., and H. Foster. 2007. Proteolytic activity and antibiotic production by Trichoderma harzianum in relation to pathogenicity to insects. Enzyme and microbial technology 40(4): 961-968. Thompson, S., R. Brandenburg, and G. Roberson. 2007. Entomopathogenic fungi detection and avoidance by mole crickets (Orthoptera: Gryllotalpidae). Environmental Entomology 36(1): 165-172. [USDA] United States Department of Agriculture. 2003. Assessing the risk of resistance to Bt corn by rotation-resistant rootworms. Available from: http://www.reeis.usda.gov/web/crisprojectpages/0197088-assessing-the-risk-of-resistance-to- bt-corn-by-rotation-resistant-rootworms.html

Vacheron, J., G. Desbrosses, M. Bouffard, B. Touraine, Y. Moenne-Loccoz, D. Muller, L. Legendre, F. Wisniewski-Dye, and C. Prigent-Combaret. 2013. Plant growth-promoting rhizobacteria and root system functioning. Frontiers in Plant Science 4(356). Villani, M., S. Krueger, P. Schroeder, F. Consolie, N. Consolie, L. Preston-Wilsey, and D. Roberts. 1994. Soil application effects of Metarhizium anisopliae on Japanese beetle (Coleoptera, Scarabaeidae) behavior and survival in turfgrass microcosms. Environmental Entomology 23(2): 502-513.

100

Figure Legends

Figure 1: Test tube arenas were constructed for our behavioral response assays. Five third- instar larvae were released in the center of an arena consisting of two test tubes affixed together. Alternate treatments were put in opposite ends of the arena, and insect choice was recorded by which side the insect was found in at 24 hours. Figure 2: Petri dish arenas were constructed for our plant/microbe interaction assays. Arenas were 10 cm Petri dishes with a layer of plain agar to retain moisture. Corn seeds treated for 48 hours or dipped for 10 seconds in microbial treatments were placed in the opposite sides of each arena, inside holes cut in agar. Third-instar larvae were released in the center and monitored for 15 minutes to determine choice of treatment. The location of the larvae was then checked at 6 hours and 24 hours. Figure 3: Y-tube olfactometer, used for adult choice experiments. A) Air tube where vacuum was attached for airflow B) Chamber where insects were introduced C) Acrylic cylinder body bifurcating into two arms, 60 cm long, 10cm wide each D) and E) Chambers where odor sources were introduced, 30cm long, 10cm wide F) and G) Flowmeter to measure rate of airflow. Figure 4: Responses of 3rd instar WCR larvae to microbe treated versus untreated non-Bt and Bt corn seeds. A) Microbial non-preference scenario. Number of larvae on each side. Different letters indicate significant differences (p<0.05). The bars represent standard error. B) Larval responses to microbe-treated non-Bt versus Bt corn seeds. Number of WCR found in each side when given the choice between Bt and non-Bt corn treated with microbes within the same test tube arena. Figure 5: A) Number of female WCR observed within oviposition arenas containing corn seeds treated with water (control), Beauveria bassiana (BB), Metarhizium anisopliae (MA), or Bacillus subtilis (BS). Different letters indicate significant difference (p<0.05). B) Mean number of eggs laid by gravid adults in oviposition arena with Bt seeds. C) Mean number of eggs laid by gravid adults in oviposition arena with non-Bt seeds.

101

Figure 6: Mean number of eggs laid by gravid WCR within oviposition chambers containing either non-Bt or Bt seeds treated with microbes or soaked in plain water (control) for 48 hours. Figure 7: Number of adult WCR emerging from greenhouse mesocosm plots. Different letters indicate significant differences (p<0.05). A) Number of adults emerging based on microbe NT= no treatment; BB= Beauveria bassiana; MA= Metarhizium anisopliae; BS= Bacillus subtilis; SM= Serratia marcescens; TH= Trichoderma harzianum. B) Number of adults emerging based on planting scenario (block strip, seed mix, or none); C) Number of adults emerging based on interaction between planting scenario and microbe. Table 1: For test tube assays, mean number of larvae in treated side, with standard errors for all values

102

103

Dipped corn

Soaked corn

Figure 2: Petri dish arenas were constructed for our plant/microbe interaction assays. Arenas were 10 cm Petri dishes with a layer of plain agar to retain moisture. Corn seeds treated for 48 hours or dipped for 10 seconds in microbial treatments were placed in the opposite sides of each arena, inside holes cut in agar. Third-instar larvae were released in the center and monitored for 15 minutes to determine choice of treatment. The location of the larvae was then checked at 6 hours and 24 hours.

104

F G E D

Figure 3: Y-tube olfactometer, used for adult choice experiments. A) Air tube where vacuum was attached for airflow B) Chamber where insects were introduced C) Acrylic cylinder body bifurcating into two arms, 60 cm long, 10cm wide each D) and E) Chambers where odor sources were introduced, 30cm long, 10cm wide F) and G) Flowmeter to measure rate of airflow.

105

Figure 4: Responses of 3rd instar WCR larvae to microbe treated versus untreated non-Bt and Bt corn seeds. A) Microbial non-preference scenario. Number of larvae on each side. Different letters indicate significant differences (p<0.05). The bars represent standard error. B) Larval responses to microbe-treated non-Bt versus Bt corn seeds. Number of WCR found in each side when given the choice between Bt and non-Bt corn treated with microbes within the same test tube arena.

106

Figure 5: A) Number of female WCR observed within oviposition arenas containing corn seeds treated with water (control), Beauveria bassiana (BB), Metarhizium anisopliae (MA), or Bacillus subtilis (BS). Different letters indicate significant difference (p<0.05). B) Mean number of eggs laid by gravid adults in oviposition arena with Bt seeds. C) Mean number of eggs laid by gravid adults in oviposition arena with non-Bt seeds.

107

Figure 6: Mean number of eggs laid by gravid WCR within oviposition chambers containing either non-Bt or Bt seeds treated with microbes or soaked in plain water (control) for 48 hours.

108

Figure 7: Number of adult WCR emerging from greenhouse mesocosm plots. Different letters indicate significant differences (p<0.05). A) Number of adults emerging based on microbe NT= no treatment; BB= Beauveria bassiana; MA= Metarhizium anisopliae; BS= Bacillus subtilis; SM= Serratia marcescens; TH= Trichoderma harzianum. B) Number of adults emerging based on planting scenario (block strip, seed mix, or none); C) Number of adults emerging based on interaction between planting scenario and microbe.

109

Table 1: For test tube assays, mean number of larvae in treated side, with standard errors for all values

Bt Non- Standard Standard Bt error Bt error non-Bt BB 2.22 2.29 0.29 0.21 MA 2.44 1.72 0.43 0.33 TH 2.16 1.24 0.26 0.30 BS 2.22 2.05 0.29 0.25 SM 2.44 1.88 0.26 0.27

110

Chapter 4

Plant-plant interactions on the expression of Bt in transgenic corn (Zea mays).

Plant-plant interactions have been described in many systems, and some prior research has suggested that transgenic plants may have decreased output of some volatiles, but the possibility that such interactions may affect how transgenes are expressed in genetically modified plants has been understudied. Our previous research on larval responses to Bt and non-Bt plants revealed that when Bt-expressing corn plants interact with non-Bt-expressing corn, this reduces Bt expression in uninfested, healthy plants. To explore this phenomenon further, we designed an experiment where we allowed transgenic Bt-expressing corn plants to interact via aerial tissue, root tissue, both, or neither, and used qRT-PCR to quantify the amount of Bt toxin expressed. We found that interactions between plants reduced the amount of toxin expressed, regardless of corn genotype. Furthermore, we determined that plant aerial tissue interactions appear to be the drivers for Bt toxin downregulation. This modulation of Bt toxin expression in response to plant-plant interaction has not been documented before, but has potential practical impacts and its nature should be further investigated. Introduction Plant-plant signaling has been well documented, primarily in interactions with herbivores and parasitoids. Corn plants are known to use both above- (Turlings and Tumlinson 1992, Tamiru et al. 2012) and belowground (Rasmann et al. 2005) chemical signaling to mediate interactions within and between plants and among plants and other organisms. Signaling compounds released at the site of wounding and insect elicitor application on corn leaves are also found in distal portions of the leaf, indicating within-plant signaling (Engelberth et al. 2012). In tomato (Solanum lycopersicum, Solanales: Solanaceae) plants, volatile organic compounds (VOCs) emitted by an insect-damaged (Spodoptera litura, Lepidoptera: Noctuidae) plant cause neighboring plants to release defensive compounds (Mescher and de Moraes 2014). Many plants are known to release VOCs that

111

are attractive to parasitoids upon damage by herbivores (as reviewed by de Vos and Jander 2010). Corn releases different blends of VOCs dependent on the species of herbivore (Helicoverpa zea and Heliothis virescens, Lepidoptera: Noctuidae), and beneficial parasitoids (Cardiochiles nigriceps, Hymenoptera: Braconidae) respond differently to these blends (De Moraes et al. 1998). These VOCs, such as green leafy volatiles and terpenoids, emitted by corn plants that have been damaged and treated with caterpillar (Noctuidae: Spodoptera exigua) regurgitant are known to attract the beneficial parasitoid wasps Cotesia marginiventris (Hymenoptera: Braconidae) (Turlings and Tumlinson 1992). In addition, corn is more attractive to the parasitoid wasp Cotesia sesamiae (Hymenoptera: Braconidae) when it has been oviposited upon by pest Lepidoptera (Chilo partellus, Hymenoptera: Crambidae) (Tamiru et al. 2012). Corn is also known to emit chemicals through its roots that can function as signals for other organisms. For example, (E)-β-caryophyllene is known to be emitted as a plant defense by attracting nematodes that attack Western corn rootworms (Chrysomelidae: Diabrotica virgifera virgifera) (Rasmann et al. 2005). However, (E)-beta- caryophyllene production has been lost in most commercially available North American corn lines (Kollner et al. 2008). In addition, synthetic (E)-beta-caryophyllene is not effective in the laboratory, which may indicate that subtle differences in quality or configuration will make a large difference (Anbesse and Ehlers 2013). In a different system, root weevils (Diaprepes abbreviates, Coleoptera: Curculionidae) on citrus (Citrus paradisi Macf. × Poncirus trifoliata L. Raf.), the coumarinic compound geijerin was shown to be the chemical responsible for nematode (Steinernema diaprepesi, Rhabdita: Steinernematidae) recruitment by the plant (Ali et al. 2010). Plant signals can also convey information to other plants, and sometimes these messages can be shared conditionally depending on the kinship of neighbors (Poelman 2013). Communication that allows plants to be more selective about synthesizing costly defensive compounds, only when an herbivore threat is imminent, could provide a fitness benefit to the plant (Baldwin et al. 2006). For instance, it is known that VOCs in corn, emitted when herbivores feed, activate the jasmonic acid defense pathway in other corn

112

plants, which leads to production and accumulation of insect defenses (Engelberth et al. 2004). Plant-plant interaction among neighboring plants can also trigger negative outcomes, such as competition through allelopathic interactions, the single most limiting factor for plant growth (Damgaard 2005). The sedentary nature of plants means that they experience competition more severely than animals that can move (Damgaard 2005). Though it is now uncontroversial that negative plant-plant interactions exist, ecological contingencies, differences in taxa, and other factors lead to differences in interactions that we have barely begun to unravel (Pearse and Karban 2013). Research discussed in Chapter 2 provides evidence that transgenic corn plants express Bt at different rates depending on their environment; that is, transgenic plants can express Bt toxin differently planted in proximity to non-Bt plants compated to when they were grown in isolation. If it were the case that nearby plants influenced Bt toxin production in rootworm- active Bt corn, this could affect the level of Bt toxins available to target WCR in the field. Since the refuge-planting scenario has the potential to influence the genotype (Bt or non-Bt) of a Bt corn plant’s neighbors have, this may also affect the amount of toxin that is produced. Interplant effects have not been documented as factors affecting toxin expression levels in Bt-expressing transgenic corn plants. This presents a novel phenomenon being explored here. Therefore, in this chapter, we examine potential effects of corn genotype and tissue type interaction on Bt toxin expression in transgenic corn. First, we investigated whether this expression was modulated by plant vicinity, regardless of corn genotype; that is, whether being grown next to a Bt or non-Bt plant would equally impact Bt toxin expression in roots of transgenic corn. Second, we investigated whether these interactions were mediated by cues emitted via root tissue or via aerial plant parts.

113

Methods and Materials Experimental system Corn seedlings used for experimental assays were from a transformed rootworm resistant line (Pioneer Herculex Xtra, 35F44, Bt) and a genetically similar susceptible line (Pioneer, 35F38, non-Bt). Corn seeds were soaked in water for 48 hours, Bt separately from non-Bt, and seeds that did not sprout within this time period were discarded. Three seeds were used for each planting, thinned to one seedling after 3 days. Plants were grown in terracotta pots (Deroma, Marshall, TX) and potting soil (Fafard 2 mix, Agawam, MA). Large pots were 25 cm wide and small pots were 15 cm wide. Planting scenarios The purpose of this experiment was to compare Bt toxin (Cry34 and Cry35) in plants that were allowed interaction via aerial tissue only, plants that were allowed interaction via root tissue only, plants that were allowed to interact via both aerial and root tissue, and plants that were not allowed to interact via aerial or root tissue. Control plants were grown singly in 15 cm pots surrounded by a Plexiglas sleeve, to measure Bt toxin expression in plants that were not allowed to interact with other plants. Full-interaction (aerial and root interaction) plants were grown in pairs with one large pot and a large Plexiglas sleeve that surrounded both plants, to measure Bt toxin expression in plants that were allowed to interact fully with neighboring plants. Root-interaction plants were grown in pairs in one large pot. One small Plexiglas sleeve surrounded each plant, so that there were two sleeves per pot, thus allowing root interaction but no aerial interaction. Aerial-interaction plants were grown singly in two pots, with a single large Plexiglas sleeve enclosing both pots, thus allowing aerial tissue interaction but no root interaction. To test for plant genotype effects on Bt toxin expression, a Bt plant was grown next to either another Bt plant or a non-Bt plant. Thus, corn line experimental setting ((Bt, Bt) or (Bt, non-Bt)) was combined with plant communication pattern (full, root, or aerial interaction). Additionally, individual Bt plants were grown as controls, which represent no interaction between plants. Thus, there were seven treatments per replicate. Ten replicates were obtained in pairs over time.

114

Ten days after emergence, the plants were removed from their pots. Plant roots were cut at the crown and washed under running water. Roots were covered in liquid nitrogen and ground into a fine powder with a mortar and pestle (Fisher Scientific, Pittsburgh, PA). Ground root samples were then stored in separate 50mL conical tubes (VWR, Radnor, PA) in a -80°C freezer until extraction was performed. RNA was extracted using Qiagen RNEasy Mini Plant Kit according to manufacturer’s instructions (Qiagen, Venlo, the Netherlands). A cDNA library was then built using Superscript III (Life Technologies, Grand Island, NY) according to manufacturer's instructions. Levels of mRNA were measured for Cry34 and Cry35 Bt toxin genes within root tissue using quantitative real-time PCR (qRT-PCR). Specific primer sequences for these two genes were obtained from Schnepf et al. (2005). Expression of mRNA for two housekeeping genes was used as a control. The GAPDH gene was chosen as a reference because it is universally expressed in all living tissues and is commonly used as a housekeeping gene (see for example Yang et al. 2015). EF1-α was used as a second control because it has been used as a housekeeping gene in similar experiments and because it is highly stable (Nicot et al. 2005). Toxin levels were controlled for expression of the housekeeping genes. To target the Bt toxin genes of interest, PCR was performed using MyTaq buffer and polymerase (Bioline, Taunton, MA). In each reaction, 1.25µL each of forward and reverse primer were used, with 5 µL buffer, 0.25 µL DNA polymerase, 15.25 µL sterile ddH2O, and 2 µL template DNA for a total of 25 µL solution. Two µL template DNA were used because this provided the best amplification results in preliminary tests. The thermocycler was set to denature proteins at 94°C for 15 seconds, then set to anneal proteins at 58°C for 10 seconds followed by 72°C for 10 seconds, for 40 cycles. Electrophoresis was used on randomly selected samples to ensure that amplification was successful. Then, qRT-PCR was performed using SYBR-Green (Qiagen, Venlo, the Netherlands) on an Applied Biosystems thermocycler (Life Technologies, Logan, UT) to quantify the amount of Cry34 and Cry35 obtained. Each well of the 96-well plate (Life Technologies,

Logan, UT) contained 3.4 µL sterile ddH2O, 0.3 µL each forward and reverse primer, 5 µL SYBR-Green dye, and 1 µL PCR product from the previous step. Again, these amounts were

115

used because previous trial runs provided the best result. The thermocycler was set to denature proteins at 95°C for 30 seconds, then set to 40 cycles of 95°C for 5 seconds, 58°C for 15 seconds, and 72°C for 10 seconds. Expression levels of Cry34 and Cry35 RNA were compared between different planting scenarios, as measured by threshold cycle (Ct, or the point at which the amplification curve meets the threshold line). Analysis of variance was used to determine the effects of planting scenario, corn line set, and their interactions on Cry34 and Cry35 expression (PROC GLM, SAS Institute, Cary, NC, 2011), for each replication Ho = πscenario, Bt,

= πscenario, non-Bt, and Ho = πfull, corn line set, = πroot, corn line set, = πaerial, corn line set. Because expression was not significantly different between Cry34 and Cry35, results for Cry34 and Cry35 were pooled. Factors that were found significant were followed by Tukey’s mean separation tests (P≤0.05). Results The effect of interaction (full, aerial, root, or none) on Bt gene expression was significant (F=6.25; df=1, 4; p<0.0001). Plants that interacted via root tissue yielded the lowest Ct values for both toxins, which indicate the highest Bt toxin gene expression, at 24.78 ± 1.25 cycles to reach threshold. Plants that had interaction via aerial tissue and plants that were allowed to communicate freely had the highest Ct, or lowest Bt toxin gene expression, at 32.99 ± 1.22 and 31.87 ± 1.26, respectively, while plants that were grown alone, had intermediate Bt toxin gene expression, with 29.02 ± 1.88 (Figure 1). Plants that were allowed to interact only via aerial tissue grouped with plants that were allowed to interact freely. Plants that were allowed to interact only via root tissue grouped with plants that were grown alone (Table 1). Toxin gene expression by plants was similar regardless of whether they were grown next to a Bt or non-Bt plant (F=1.29; df=1, 3; p=0.28). Discussion This chapter builds upon our research findings in Chapter 2 by investigating whether the vicinity of any corn plant was enough to change regulation of Bt toxin expression in corn roots regardless of neighboring plant genotype. Our data indicate that plants grown in proximity to any other corn plant indeed differentially regulated toxin gene expression.

116

Transgenic Bt plants may not always express compounds at the same rate as non-transgenic plants; for example, Bt expression in oilseed rape (Brassica napus, Brassicales: Brassicaceae) appeared to decrease plants' resource allocation to endogenous defenses against other stressors; Bt-expressing plants were less efficient at attracting parasitoid enemies of herbivores under environmental stress such as elevated O3, indicating reduced expression of volatiles (Himanen et al. 2008). Presence of a gene for Bt toxin expression does not appear to alter the content of volatile organic compounds (VOCs) produced in corn, as long as herbivore feeding pattern is controlled (Dean and de Moraes 2006). However, genotype is an important factor modulating plant competition; high plant-to-plant variability in corn has been shown to lead to poor performance when plants are grown in crowded conditions (Pagano and Maddonni 2007). We therefore investigated whether corn genotype would affect the expression of Bt. Genotype had no effect on plant-plant interactions. Factors that affect Bt toxin expression are important to document, in part because rootworm-active transgenic Bt corn already expresses a low-to-moderate dose of Bt toxin rather than an ideal high dose (EPA 2002). We also investigated at which tissue level (full, root, or aerial) interaction was occurring between plants, leading to these Bt gene expression changes. Plant-plant communication can occur via root signaling (Kollner et al. 2008) or above-ground signaling

(Muroi et al. 2011). In our experiments, Ct value was lowest (indicating highest expression level) in plants grown in isolation, and lowest in plants that were allowed to interact via aerial tissue. Plants that interacted via aerial tissue grouped with plants that interacted fully (via both aerial and root tissue); plants that interacted via root tissue grouped with plants that were grown in isolation. This indicates that interplant communication in our experiment was taking place via aerial cues such as green leaf volatiles. The category of volatiles known as green leaf volatiles (including (Z)-3-hexenal, (E)-2-hexenal, (Z)-3-hexen-1-ol, (E)-2-hexen- 1-ol, and (Z)-3-hexen-1-yl acetate), are rapid-acting; they are emitted within the first hour after a corn plant is damaged (Turlings et al. 1998). Volatile organic compounds have varied effects dependent on the organisms involved; VOCs emitted by damaged corn are attractive to both herbivores (Spodoptera littoralis, Lepidoptera: Noctuidae) (von Mérey et al. 2013)

117

and to parasitoids (Trichogramma pretiosum, Hymenoptera: Trichogrammatidae) (Saldivar et al. 2011) but repellant to adult female herbivores (Spodoptera frugiperda, Lepidoptera: Noctuidae) as oviposition sites (Signoretti et al. 2012). In our experiments, presence of a neighboring plant caused plants to downregulate rather than upregulate Bt toxin expression. There is no known mechanism by which Bt toxin expression could be regulated in the same manner as natural defensive compounds within the plant, because Bt toxins have not evolved to be expressed naturally in the plant. Though allelopathy in root tissue has been extensively studied (see, for example, Pedersen et al. 2013), allelopathy is also known to occur in aerial tissues in some plants such as Rapanea umbellata Martius (Ericales: Primulaceae) (see, for example, Novaes et al. 2013). Thus, it may be the case that aerially-transmitted allelopathic effects are causing plants to downregulate expression of multiple genes, and indirectly affects expression of the Bt transgene. However, the most interesting result of this experiment was simply that Bt expression changed in response to the presence of neighboring plants, regardless of genotype. It is known that intraspecies plant-plant communication can alter resource allocation differently depending on the cultivar; for instance, barley (Hordeum vulgare) (Poales: Poaceae) examined by Nincovic (2003) allocated biomass to roots differently depending upon the cultivar of neighboring barley plants. However, to the best of our knowledge, this is the first time that expression of a transgene has been shown to be regulated by interplant communication. Isolating individual plants for experimental purposes required us to use differently sized pots and sleeves, so it is important to note that this may have interfered with our results. However, size of pot and/or sleeve did not affect the growth or appearance of our experimental plants (SLG, personal observation). In addition, our results were in agreement with results seen in Chapter 2, where we used different types of arenas and growth media, yet the same effects on toxin expression levels were obtained for Bt plants grown in the vicinity of other plants.

118

The dose of toxin received by WCR is of importance with regards to effectivity of the Bt-expressing corn plants (Binning et al. 2010, Head and Greenplate 2012); hence, it would be interesting to investigate if Bt expression is modulated in response to neighboring plants under field conditions. Though a few researchers have investigated the interrelationship between Bt toxin production and VOC production (see for example Dean and de Moraes 2006, Himanen et al. 2008), how and under what circumstances Bt production can affect VOC production is a question that has only begun to be investigated.

119

References

Ali, J., Alborn, T., and Stelinski, L. 2010. Subterranean herbivore-induced volatiles released by citrus roots upon feeding by Diaprepes abbreviatus recruit entomopathogenic nematodes. Journal of Chemical Ecology 36(4): 361-368. Anbesse and Ehlers. 2013. Heterorhabditis sp. not attracted to synthetic (E)-beta- caryophyllene, a volatile emitted by roots upon feeding by corn rootworm. Journal of Applied Entomology 137(1-2): 88-96. Binning R, Lefko S, Millsap A, Thompson S, Nowatzki T (2010) Estimating western corn rootworm (Coleoptera: Chrysomelidae) larval susceptibility to event DAS-59122-7 maize. Journal of Applied Entomology 134(7): 551-561. Carozzi, N., and M. Koziel, eds. 1997. Advances in Insect Control: The Role of Transgenic Plants. London: Taylor & Francis. Damgaard, C. 2005. Evolutionary Ecology of Plant-Plant Interactions: An Empirical Modeling Approach. Aarhus University Press. p. 12. Dean, J., and de Moraes, C. 2006. Effects of genetic modification on herbivore-induced volatiles from maize. Journal of Chemical Ecology 32(4): 713-724. De Moraes, C., W. Lewis, P. Paré, H. Alborn, and J. Tumlinson. 1998. Herbivore-infested plants selectively attract parasitoids. Nature 393(6685): 570-573. de Vos, M., and G. Jander. 2010. Volatile communication in plant-aphid interactions. Current Opinion in Plant Biology 13(4): 366-371. Engelberth, J., H. Alborn, E. Schmelz, and J. Tumlinson. 2004. Airborne signals prime plants against insect herbivore attack. Proceedings of the National Academy of Sciences of the United States of America 101(6): 1781-1785. Engelberth, J., C. Fabiola Contreras, and S. Viswanathan. 2012. Transcriptional analysis of distant signaling induced by insect elicitors and mechanical wounding in Zea mays. PLoS One: 7(4): e34855.

120

[EPA] Environmental Protection Agency. 1998. Transmittal of the final report of the FIFRA Scientific Advisory Panel Subpanel on Bacillus thuringiensis (Bt) plant-pesticides and resistance management, meeting held on February 9 and 10, 1998. Available from [EPA] Environmental Protection Agency (2014) Insect resistance management fact sheet for Bacillus thuringiensis (Bt) corn products. Available from http://www.epa.gov/oppbppd1/biopesticides/pips/bt_corn_refuge_2006.htm [EPA] Environmental Protection Agency (2014) White paper on corn rootworm resistance monitoring for Bt plant-incorporated protectants. Available from http://www.regulations.gov/#!documentDetail;D=EPA-HQ-OPP-2013-0490-0008 [EPA] Environmental Protection Agency. 2002. FIFRA scientific advisory panel (SAP) open meeting, August 27-29, 2002. Available from http://www.epa.gov/scipoly/sap/meetings/2002/august/agenda.pdf Gassmann A, Petzold-Maxwell J, Keweshan R, Dunbar M (2011) Field-evolved resistance to Bt maize by Western corn rootworm. PLoS one 6(7): e22629. Gould, F. 1998. Sustainability of transgenic insecticidal cultivars: Integrating pest genetics and ecology. Annual Review of Entomology 43: 701-726. Himanen, S., A. Nerg, A. Nissinen, D. Pinto, C. Stewart, G. Poppy, and J. Holopainen. 2008. Effects of elevated carbon dioxide and ozone on volatile terpenoid emissions and multitrophic communication of transgenic insecticidal oilseed rape (Brassica napus). New Phytologist 181(1): 174-186. Kollner, T., M. Held, C. Lenk, I. Hiltpold, T. Turlings, J. Gershenzon, and J. Degenhardt. 2008. A maize (E)-beta-caryophyllene synthase implicated in indirect defense responses against herbivores is not expressed in most American maize varieties. Plant Cell 20(2): 482- 494. Mescher, M., and C. De Moraes. 2014. Plant biology: pass the ammunition. Nature 510: 221-222. Muroi, A., A. Ramadan, M. Nishihara, M. Yamamoto, R. Ozawa, J. Takabayashi, and G. Arimura. 2011. The composite effect of transgenic plant volatiles for acquired immunity to herbivory caused by inter-plant communications. PLoS One 6(10): e24594.

121

Nicot, N., J. Hausman, L. Hoffmann, and D. Evers. 2005. Housekeeping gene selection for real-time RT-PCR normalization in potato during biotic and abiotic stress. Journal of Experimental Botany 56(421): 2907-2914. Nincovic, V. 2003. Volatile communication between barley plants affects biomass allocation. Journal of Experimental Botany 54(389): 1931-1939. Novaes, P., Imatomi, M., Miranda, M., and Gualtieri, S. 2013. Phytotoxicity of leaf aqueous extract of Rapanea umbellata (Mart.) Mez (Primulaceae) on weeds. Acta Scientarum- Agronomy 35(2): 231-239. Pagano, E., and G. Maddonni. 2007. Intra-specific competition in maize: early established hierarchies differ in plant growth and biomass partitioning to the ear around silking. Field Crops Research 101(3): 306-320. Pearse, I., and R. Karban. 2013. Do plant-plant signals mediate herbivory consistently in multiple taxa and ecological contexts? Journal of Plant Interactions 8(3): 203-206. Pedersen, H., P. Kudsk, O. Fiehn, and I. Fomsgaard. 2013. The response of Arabidopsis to co-cultivation with clover investigating plant-plant interactions with metabolomics. Pest Management with Natural Products 1141: 189-201. Rasmann, S., T. Köllner, J. Degenhardt, I. Hiltpold, S. Toepfer, U. Kuhlmann, J. Gershenzon, and T. Turlings. 2005. Recruitment of entomopathogenic nematodes by insect-damaged maize roots. Nature 434(7034): 732-737. Rudeen M, Gassmann A (2012) Effects of Cry34/35Ab1 corn on the survival and development of western corn rootworm, Diabrotica virgifera virgifera. Pest Manage. Sci. 69(6): 709-716. Saldivar X., L. Modenez, R. Laumann, M. Borges, D. Magalhães, E. Vilela, and M. Blassioli-Moraes. 2011. Trichogramma pretiosum attraction due to the Elasmopalpus lignosellus damage in maize. Pesquisa Agropecuaria Brasiliera 46(6): 578-585. Schnepf, H., S. Lee, J. Dojillo, P. Burmeister, K. Fencil, L. Morera, L. Nygaard, K. Narva, and J. Wolt. 2005. Characterization of Cry34/Cry35 binary insecticidal proteins from diverse Bacillus thuringiensis strain collections. Applied and Environmental Microbiology 71(4): 1765-1774.

122

Signoretti, A., M. Penaflor, and J. Bento. Fall armyworm, Spodoptera frugiperda (JE Smith) (Lepidoptera: Noctuidae), female moths respond to herbivore-induced corn volatiles. Neotropical Entomology 41(1): 22-26. Sivasupramaniam, S., G. Head, L. English, Y. Li, and T. Vaughn. 2007. A global approach to resistance monitoring. Journal of Invertebrate Pathology 95(3): 224-226. Tabashnik, B. 2008. Delaying insect resistance to transgenic crops. Proceedings of the National Academy of Sciences of the United States of America. 105(49): 19029-19030. Vaughn T, Cavato T, Brar G, Coombe T, DeGooyer T, Ford S, Groth M, Howe A, Johnson S, Kolacz K, Pilcher C, Purcell J, Romano C, English L, Pershing J. (2005) A method of controlling corn rootworm feeding using a Bacillus thuringiensis protein expressed in transgenic maize. Crop Science 45(3): 931-938. Tamiru, A., T. Bruce, C. Midega, C. Woodcock, M. Birkett, J. Pickett, and Z. Khan. 2012. Oviposition induced volatile emissions from African smallholder farmers’ maize varieties. Journal of Chemical Ecology 38(3): 231-234. Turlings, T., U. Lengwiler, M. Bernasconi, and D. Wechsler. 1998. Timings of induced volatile emissions in maize seedlings. Planta 207(1): 146-152. Turlings, T., and J. Tumlinson. 1992. Systemic release of chemical signals by herbivore- injured corn. Proceedings of the National Academy of Sciences of the United States of America 89(17): 8399-8402. von Mérey, G., N. Veyrat, M. D'Alessandro, and T. Turlings. 2013. Herbivore-induced maize leaf volatiles affect attraction and feeding behavior of Spodoptera littoralis caterpillars. Frontiers in Plant Science 4(209). Yang, Z., Y. Chen, B. Hu, Z. Tan, and B. Huang. Identification and validation of reference genes for quantification of target gene expression with quantitative real-time PCR for tall fescue under four abiotic stresses. PLoS One 10(3): e0119569.

123

Figure Legends

Table 1: Ct values for expression of Cry35 and Cry35 in root tissue, by interaction treatment and plant genotype (Bt or non-Bt).

124

125

Table 1: Ct values for expression of Cry35 and Cry35 in root tissue, by interaction treatment and plant genotype (Bt or non-Bt). Sample Target N Obs Mean Std Error Bt, Full Cry34 36 35.997 2.442 interaction Cry35 36 32.615 2.444 Bt, Root Cry34 28 22.146 2.210 interaction Cry35 24 26.299 3.157 Bt, Headspace Cry34 36 36.493 2.394 interaction Cry35 36 29.957 1.988 Non-Bt, Full Cry34 32 30.405 2.916 interaction Cry35 32 27.842 2.066 Non-Bt, Root Cry34 28 22.619 2.212 interaction Cry35 28 28.280 2.406 Non-Bt, Cry34 32 31.692 2.755 Headspace interaction Cry35 36 33.691 2.594 No interaction Cry34 36 33.444 2.910 Cry35 36 24.587 2.190

126

Chapter 5

Conclusion

127

Soil microbiota such as Metarhizium anisopliae (Metchnikoff) Sorokin (Hypocreales: Clavicitipitaceae) and Beauveria bassiana (Balsamo) Vuillemin (Hypocreales: Clavicitipitaceae) can be effective at killing WCR in the laboratory, but has not shown much utility in the field. This lack of effectivity may be explained in part by evolved WCR behavioral non-preference to such entomopathogens, although environmental factors such as humidity are also likely factors. To explore this possibility, we studied the behavioral responses of larval and gravid adult WCR towards various soil microbe-treated corn and how this was influenced by corn line. We also investigated rootworms’ time to emergence, survival, and sex ratios when feeding on plants produced from Bt and non-Bt corn seeds treated with the various microbes of interest, under greenhouse conditions. We found that treatment with several different species of soil microbes confounds WCR’s ability to avoid Bt-expressing transgenic seeds in both larvae and adults. We found that exposing WCR to plants grown from microbe-treated seeds in greenhouse conditions reduced survival to adulthood, regardless of microbe species among those studied. We also found that responses to M. anisopliae were consistently significant, although adults appeared to show preference for M. anisopliae while larvae showed non-preference, suggesting that this may be a species of interest for further study with respect to its role in WCR ecology. Plant-plant interactions have been described in many systems, and some prior research has suggested that transgenic plants may have decreased output of some volatiles, but the possibility that such interactions may affect how transgenes are expressed in genetically modified plants has been understudied. Our previous research on larval responses to Bt and non-Bt plants revealed that when Bt-expressing corn plants interact with non-Bt-expressing corn, this reduces Bt expression in uninfested, healthy plants. To explore this phenomenon further, we designed an experiment where we allowed transgenic Bt-expressing corn plants to interact via aerial tissue, root tissue, both, or neither, and used qRT-PCR to quantify the amount of Bt toxin expressed. We found that interactions between plants reduced the amount of toxin expressed, regardless of corn genotype. Furthermore, we determined that plant aerial tissue interactions appear to be the drivers for Bt toxin downregulation.