• unlimited access with print and download
    $ 37 00
  • read full document, no print or download, expires after 72 hours
    $ 4 99
More info
Unlimited access including download and printing, plus availability for reading and annotating in your in your Udini library.
  • Access to this article in your Udini library for 72 hours from purchase.
  • The article will not be available for download or print.
  • Upgrade to the full version of this document at a reduced price.
  • Your trial access payment is credited when purchasing the full version.
Buy
Continue searching

Chemical ecology of plant-microbe interactions and effects on insect herbivores

Dissertation
Author: Jennifer M. Dean
Abstract:
Plants are central to most community interaction webs and thus the ability to coordinate responses to many simultaneous interactions is an essential adaptation. Interaction with one organism may make the plant more vulnerable or more resistant to attacks by very different organisms. Microorganisms can form intimate associations with plants with direct effects ranging from beneficial to antagonistic, but indirect effects of plant-microbe relationships on plant interactions with other organisms are not well understood. Here, we explore the influences of microorganisms and microbial products on herbivory and resulting plant defenses. We use the legume-rhizobia mutualism as a model-system to explore herbivore-plant interactions by first characterizing the effects of rhizobial inoculation on herbivore feeding and subsequent accumulation of plant defense signaling hormones. We found interactive effects of the legume-rhizobia mutualism on plant-herbivore interactions which were dependent on both the stage of the mutualism and on the feeding style of the herbivore. Next, we explored the effects of association with different sources of rhizobia of soybean on specialist aphid populations in an agricultural setting. We found that particular rhizobia strains can confer greater resistance to their mutualist partners than others. In order to explore the effects of a rhizobia strain-specific trait on herbivory, we focused on the rhizobial product, rhizobitoxine, which can be found in host plant tissues. We found that the presumed presence of this compound decreased herbivore feeding and damage, which could be useful in pest management of legumes. Finally, we utilize genetically modified maize to further explore the application of a microbial product used for insect resistance and its effects on herbivore feeding and induction of plant defenses. Our results contribute to understanding of the evolution of host plant defenses and could facilitate the development of more sustainable management techniques for agriculture that are informed by an understanding of the chemical ecology of plants, microbes, and insects.

iv Table of Contents

List of Figures...............................................................................................................v List of Tables...............................................................................................................vi Acknowledgements.....................................................................................................vii CHAPTER 1 - Introduction..........................................................................................1 Plant-Herbivore Interactions.....................................................................................2 Legume-Rhizobia Mutualism...................................................................................4 Model system for studying rhizobia-legume-herbivore interactions........................6 Use of microbial products in agriculture for pest control through genetic modification............................................................................................................11 Overview of chapters..............................................................................................12 References...............................................................................................................14 CHAPTER 2 -The legume-rhizobia mutualism influences herbivore feeding and plant hormone responses......................................................................................................21 Abstract...................................................................................................................21 Introduction.............................................................................................................22 Methods...................................................................................................................25 Results.....................................................................................................................30 Discussion...............................................................................................................33 References...............................................................................................................39 CHAPTER 3 - Plant-rhizobia mutualism influences aphid abundance on soybean...50 Abstract...................................................................................................................50 Introduction.............................................................................................................51 Methods...................................................................................................................53 Results.....................................................................................................................56 Discussion...............................................................................................................57 References...............................................................................................................62 CHAPTER 4 -Preliminary assessment of the effects of rhizobitoxine production on feeding by Helicoverpa zea........................................................................................71 Abstract...................................................................................................................71 Introduction.............................................................................................................71 Methods...................................................................................................................75 Results.....................................................................................................................76 Discussion...............................................................................................................77 References...............................................................................................................80 CHAPTER 5 - Effects of genetic modification on herbivore-induced volatiles from maize...........................................................................................................................83 Abstract...................................................................................................................83 Introduction.............................................................................................................84 Methods...................................................................................................................86 Results.....................................................................................................................90 Discussion...............................................................................................................91 References...............................................................................................................94

v List of Figures

Figure 1.1. Infection and nodulation process of the rhizobia-legume interaction (Lum and Hirsch, 2002)......................................................................................................5 Figure 2.1. Percent total nitrogen (determined by combustion) of young soybean plants grown without nitrogen fertilizer.................................................................43 Figure 2.2. Effects of rhizobial inoculation on soybean podworm feeding on young plants (before active nitrogen fixation) and plant responses..................................44 Figure 2.3. Effects rhizobia inoculation on soybean aphid feeding on young plants (before active nitrogen fixation) and plant responses during course of experiment and on Day 7...........................................................................................................45 Figure 2.4. Characteristics of 5-week old plants inoculated with rhizobia, but given different levels of nitrogen fertilizer.......................................................................46 Figure 2.5. Effects of rhizobial dependence for N in 5-week old soybeans on the behavior and growth of H. zea larvae.....................................................................47 Figure 2.6. Effects of rhizobial dependence for N in 5-week old soybeans on growth rate of H. zea and plant responses...........................................................................48 Figure 2.7. Effects of rhizobial dependence for N in 5-week old soybeans on reproductive rate of soybean aphids and plant responses.......................................49 Figure 3.1. Abundances of soybean aphid throughout the field season on soybean plants either treated with a commercial rhizobial inoculant at the time of planting (circles), associating solely with indigenous rhizobia (triangles), or given a nitrogen fertilizer to suppress nodulation (squares)..............................................................66 Figure 3.2. Aphid abundance (a), leaf nitrogen content (b), plant dry weight (c), and nodule dry weight (d) of soybean plants shortly before the aphid population peak at the middle of the summer.......................................................................................67 Figure 3.3. Leaf nitrogen content (% N) throughout the field season for soybean plants either treated with a commercial rhizobial inoculant at the time of planting (circles), associating solely with indigenous rhizobia (triangles), or given a nitrogen fertilizer to suppress nodulation (squares). ...........................................................68 Figure 3.4. Soybean yield in terms of number of seeds per plant, shown as mean ± 1 standard error..........................................................................................................69 Figure 3.5. Evolutionary relationships (neighbor-joining) of rhizobia isolates from field plants treated with a commercial rhizobial inoculant at the time of planting (CR) or associating solely with indigenous rhizobia (IR) and known strains (B. (Bradyrhizobium) japonicum or B. elkanii USDA plus strain number, accessions from van Berkum and Fuhrmann, 2000) based on alignment of ITS region sequences using MEGA 3.1....................................................................................70 Figure 4.1. Target reactions of rhizobitoixine. SAM, S-adenosylmethionine; ACC, 1- aminocyclopropane-1-carboxylate. From Okazaki et al. 2004...............................73 Figure 4.2. Feeding by H. zea in no-choice bioassay over 4 days.............................77 Figure 5.1. Sample of feeding patterns created by H. zea larvae on Bt and non-Bt maize leaves............................................................................................................97 Figure 5.2. Chromatograms of volatile compounds emitted on Day 3 from non-Bt maize plants subject to feeding by H. zea larvae, mechanical damage, mechanical damage with regurgitant applied, and no damage..................................................98

vi List of Tables Table 5.1. Total volatiles (mean ± SE) released during day 3 from Bt and non-Bt maize.......................................................................................................................99 Table 5.2. Major compounds released from Bt and non-Bt maize receiving an equal amount of herbivore damage on day 3 a ................................................................100 Table 5.3. Characteristics of damage holes created by H. zea larvae while controlling duration of feeding to achieve an equal amount and pattern of damage..............101

vii Acknowledgements

I am indebted to the many people have contributed to these pages in many different ways. First, I would like to thank my advisor, Dr. Consuelo De Moraes, who has taken great strides to impart her wisdom and joy of science and to prepare me for the challenges that lie ahead. She has always encouraged me to pursue new opportunities, and her constant support has made this project possible. I would also like to thank Dr. Mark Mescher for being a generous source of ideas and insight. I am also grateful to my committee members, Drs. Gretchen Kuldau, Jonathan Lynch, and Jim Tumlinson for their time, expertise, and encouragement. I would like to express my gratitude to the wonderful members of the lab, especially Casey Delphia, Justin Runyon, Dr. John Tooker, and Dr. Chris Frost, for many years of discussions, advice, and laughs. Janet Saunders and Ed Bogus have been instrumental in this work by keeping the plants growing and the lab humming beautifully. Help from many undergraduate students, particularly Adam Conrad, Jessica Dirle, Erica Smyers, Lauren Seiler, and Clare Wagner, have made even the tedious sides of research more fun. I would also like to thank the National Science Foundation, Penn State University, College of Agricultural Sciences, and Department of Entomology, for funding that opened new opportunities for my research and growth as a scientist, both at home and abroad. I am fortunate for the many people who have become dear friends in Happy Valley and who have helped me to maintain balance in my life through graduate school. I would also like to thank my wonderful family, who has given me the support and courage to make this goal a reality. My parents, Skip and Carol, and brother, Paul, have always cheered me on through my pursuits in life. And without the love and strength of my husband and daughter, I would have not accomplished what I have today. In addition to all the computer support, ingenious devices, and aesthetic critiques, Bryan helped me to keep the important things in life in a clearer perspective. Alexa has been a source of unconditional smiles, and her curiosity about the natural world constantly reminds me why I wanted to be a scientist.

1 Chapter 1

Introduction

Being primary producers, plants play a central role in most interaction webs, and have thus evolved a wide array of mechanisms to respond to encounters with other organisms. These interactions, which often occur simultaneously and range from beneficial to antagonistic, require the plant to coordinate physiological responses in a manner which minimizes fitness costs. Many recent studies have highlighted the importance of examining plant interactions with multiple organisms simultaneously in order to gain a clearer understanding of the evolution and ecology of plant defenses. Distinct attackers such as herbivores and pathogens elicit unique responses from the host plant, and combinations of attackers often produce surprising results (Cardoza et al., 2003; Cui et al., 2005; Delphia et al., 2007; Rodriguez-Saona et al., 2005; Thaler et al., 2002b; Wurst and van der Putten, 2007). In general, attack by one organism may make the plant more vulnerable or more resistant to future attacks by very different organisms (Morris et al., 2007). Beneficial organisms can also affect plant resistance to attackers. Several classes of microorganisms engage in mutualisms with plants, in which both partners benefit from the interaction. These complex, coevolved associations between plant and microbe can alter many plant traits, and thus influence other interactions, especially those between plants and herbivores. Studies exploring the effects of microbial mutualists, particularly mycorrhizal fungi and fungal endophytes, on plant- herbivore interactions have found both positive and negative consequences for the host plant and enemy (Bennett and Bever, 2007; Borowicz, 1997; Gange and West, 1994; Goverde et al., 2000; Tintjer and Rudgers, 2006). The mechanisms by which mutualistic microbes can influence herbivory are even less clear, but could include interference of plant signaling networks, changes in plant quality for the consumer, or direct interaction between microbial products and herbivores.

2 The following collection of studies explores how plant defenses against herbivores are modulated by concurrent interactions with plant-associated microbes. The primary component focuses on the well known mutualism between nitrogen- fixing bacteria (rhizobia) and legumes, and the consequential effects of this relationship on herbivore feeding and plant responses to incurred damage. We examine host plants with different stages and intensities of rhizobial inoculation exposed to feeding by two distinct herbivores in order to gain a broad understanding of these complex, three-way relationships. We also explore the influence of genetic diversity of rhizobia partners on the variation found in effects on plant-herbivore interactions. By using genetically modified maize, we conclude by exploring the effects of a microbial toxin on herbivore-induced volatile production – a component of indirect plant defenses – without the confounding effects of infection itself. This body of work aims to increase our understanding of the chemically mediated interactions between plants, microbes, and herbivores in order to contribute to the goal of sustainably managing agricultural systems through strategies informed by a sophisticated understanding of natural ecological systems. Plant-Herbivore Interactions Due to their relatively large size and immobility, plants may appear, at first glance, to be vulnerable targets for the immense diversity of herbivorous insects. Yet plants dominate terrestrial ecosystems in terms of biomass, emphasizing the longstanding question: Why is the world green? A simplistic approach suggests that herbivory is limited by resource limitations (bottom-up effects), control by natural enemies (top-down effects), or more likely, a combination of both (Gruner, 2004; Ode, 2006; Polis, 1999; Terborgh et al., 2001). Host plant quality from the perspective of an herbivore largely depends on the nutrient composition and defensive chemistry of the plant. These traits can be altered by concurrent interactions between plants and other organisms, which, in turn, drive changes in the behavior and fitness of herbivores (Borowicz, 1997; Cui et al., 2005; Rodriguez-Saona et al., 2005; Selosse et al., 2004; Walling, 2000; Wilson and Stinner, 1984). Nitrogen (N), which is essential for protein building and nucleic acids,

3 appears to be the most limiting nutritional factor for herbivores. The concentration of total N in plants (typically 2-4%) is much lower than that of primary consumers (about 8-14% N), underlying limitations to herbivore growth (Schoonhoven et al., 2005). The form of N is also of critical importance, as many plant compounds rich in N are unusable by, or even detrimental to, herbivores (Felton, 1996; Mattson, 1980). Plant defensive chemistry is also a key regulator of herbivore behavior and fitness. Some chemical defenses, such as secondary metabolites and anti-nutritive proteins, have direct negative impacts on herbivores, while others, such as volatile organic compounds, work indirectly by recruiting natural enemies to the host (reviewed in Karban and Baldwin, 1997; Schoonhoven et al., 2005). Plant defenses can either be constitutively present in tissues, or induced upon plant recognition of attack. The induction of defenses requires coordination of signaling pathways within the plant which ultimately lead to a cascade of defense-related events. The phytohormones jasmonic acid (JA) and salicylic acid (SA) are generally associated with plant defense pathways in response to herbivores and pathogens, respectively, and inhibiting the function of either increases plant susceptibility (Walling, 2000). In addition to involvement in many key plant functions, JA plays a pivotal role in many herbivorous attacks and leads to the production of both direct and indirect defensive compounds that can help prevent further damage (Chen et al., 2005; Farmer et al., 1992; Schmelz et al., 2003; Thaler et al., 2002a). However, many factors can modulate herbivore-induced accumulation and action of JA, such as simultaneous induction of SA pathways commonly associated with pathogens (Felton et al., 1999; Thaler et al., 2002b). Moreover, piercing-sucking insects often trigger SA-dependent pathways, while some non-pathogenic, root-colonizing microbes confer a general resistance against diseases through a JA-dependent pathway (Walling, 2000). As interactions with mutualistic microorganisms such as rhizobia also involve the JA and SA plant pathways, plant-herbivore interactions are likely be influenced by the presence of rhizobia.

4 Legume-Rhizobia Mutualism Legumes form symbiotic relationships with rhizobia, bacteria which are capable of fixing atmospheric nitrogen and making it available to the plants in a biochemically usable form. In return, the plants provide carbon and a protective root nodule within which the bacteria live. This specialized association allows legumes to thrive where nitrogen is a limiting factor for other plants, contributing to their ecological importance in both natural and agricultural systems. This pairing of legumes and rhizobia is credited with contributing 70 million metric tons of fixed nitrogen (N) into the earth’s soils annually (Brockwell et al., 1995). When integrated into cropping systems with non-legumes, the nitrogen transfer results in reduced need for fertilizers (O'Hara et al., 2002). The ease and low cost of inoculating legume crops with rhizobial preparations, coupled with the yield advantages associated with doing so, has made inoculation a common agricultural practice. The association of legumes with rhizobia, believed to have originated 65 million years ago, involves numerous physical and chemical changes in the host plant in response to inoculation by a compatible strain of rhizobia (Hirsch and LaRue, 1997) (Figure 1.1). Specific flavonoids exuded by legume roots are perceived by rhizobia, resulting in positive chemotaxis and production of Nod factors, various forms of lipochitooligosaccharides, which in turn initiate nodule development on legume roots (Hirsch et al., 2001). In many legumes, an individual bacterium infects the plant via a plant-produced “infection-thread” through the root hair. Rhizobia reproduce inside the infection thread and are eventually released into the plant cell. Many will differentiate into the N-fixing bacteroids, which will become surrounded by a plant-derived peribactoid membrane within the nodule (Lee and Hirsch, 2006). Once the plant no longer requires the N derived from rhizobia, usually as seeds are forming, the nodules are senesced from the roots and into the soil, at which point viable rhizobia cells are ready to infect another host. The mutualism is facultative, meaning each partner can survive and reproduce without the other. However, each partner can receive a substantial increase in fitness benefits by participating in the mutualism, especially under N-limiting conditions (West et al., 2002).

5

Figure 1.1. Infection and nodulation process of the rhizobia-legume interaction (Lum and Hirsch, 2002).

During nodule development, the host plant may initially respond to rhizobia as a pathogen, as suggested by the induction of anti-pathogen enzyme genes of phenylpropanoid metabolism (Estabrook and Senguptagopalan, 1991), the inhibition of salicylic acid biosynthesis during nodule formation (Martinez-Abarca et al., 1998), and the presence of an oxidative burst in infected roots (Santos et al., 2001). However, further plant defense mechanisms have not been noted, suggesting that the interaction with rhizobia somehow circumvents or suppresses later stages of the response pathway (Baron and Zambryski, 1995; Mithofer, 2002). Additionally, some phytohormones commonly associated with plant defenses against pathogens and herbivores, particularly salicylic acid (SA) and jasmonic acid (JA), appear to be involved in rhizobia-legume interactions, as application of these hormones will inhibit nodulation (Sato et al., 2002; Stacey et al., 2006; Sun et al., 2006). At later stages of the rhizobia-legume interaction, following nodule establishment, physiological differences have been described between inoculated legumes, and non- inoculated plants receiving supplemental nitrate fertilizers. Inoculated legumes transport nitrogen to the shoots in the form of amides or ureides, as opposed to the inorganic forms that are transported in non-inoculated legumes (Matsumoto et al., 1977b). The addition of nitrate to inoculated soybean plants reduces the amount of

6 functional nodules and also reduces the proportion of total nitrogen available as ureides while maintaining equivalent total N levels (Matsumoto et al., 1977a; McClure and Israel, 1979). Many of the physical and chemical changes induced in the host plant by compatible rhizobia are strongly influenced by the source and genotype of the bacterial partner. Different strains of rhizobia can have different effects on features of the mutualism including nitrogen-fixation efficiency, leaf chemistry, and plant fitness (Burdon et al., 1999; Fuhrmann, 1990; Lafavre and Eaglesham, 1986; Parker, 1995; van Berkum et al., 1985). For example, many of the soybean-compatible rhizobia strains commonly encountered in US soils fix less N for, yet are preferentially nodulated by, the host plant as compared to strains introduced from intentional inoculation at the time of planting (Amarger, 2001). Also, some rhizobia strains produce compounds, in addition to fixed-N products, which are translocated to the shoots, affecting leaf chemistry. The production of rhizobitoxine, an enol-ether amino acid, by particular rhizobia strains provides one example of a bacterial product found in host plant tissues that could potentially affect herbivore feeding. To our knowledge, no studies have explored how these variations in host plant effects due to different rhizobial genotypes could influence herbivory.

Model system for studying rhizobia-legume-herbivore interactions Soybean-herbivore interactions Legumes are attacked by a wide array of herbivores, and accordingly have evolved a diverse suite of both constitutive and inducible defenses. Soybeans defenses in particular have received a good deal of attention due to their worldwide economic status and breeding programs aimed at producing herbivore resistant cultivars. However, breeding for insect resistance in soybeans in the US has met limited success, and in actuality, many key defensive mechanisms are missing from

7 cultivated legumes due to selective breeding for other agronomic traits (Edwards and Singh, 2006). Constitutive defenses of soybean leaves and seeds have been well studied, and include structural protection (trichomes) and chemical defenses (e.g., lectins and flavonoids) (Edwards and Singh, 2006). Soybeans are rich in flavonoids in all plant parts tested (Romani et al., 2003). These compounds are considered to be an important component of defense against many invaders, in addition to serving as critical belowground signals to initiate the nodulation process with compatible rhizobia strains (Kosslak et al., 1987). Isoflavonoids from soybeans have been shown to have negative effects on herbivore behavior, consumption, and growth, and are inducible via herbivore feeding (Hoffmann-Campo et al., 2001; Piubelli et al., 2003; Sharma and Norris, 1991). Many other components of soybean defenses are also inducible; resulting in reductions in growth of herbivores fed previously damaged leaves. Proteinase inhibitors, which hinder the ability of herbivores to utilize ingested leaf proteins, accumulate in leaves after damage, and are implicated in growth reductions of beetles and lepidopteran larvae (Bi et al., 1994; Kraemer et al., 1987). Components of oxidative stress, such as oxidative enzymes and lipid peroxidases are also increased in soybeans following herbivory, with specific components dependent on the type of herbivore (Bi and Felton, 1995; Felton et al., 1994). Soybean aphids induced volatiles, such as methyl salicylate, which were attractive to predaceous beetles and parasitoids (Wyckhuys and Heimpel, 2007; Zhu, 2005). There is much variation among soybean cultivars in the degree of inducible and constitutive defenses (Underwood et al., 2000). The strong inducibility of soybeans defenses are hypothesized to have prevented local legume specialists from colonizing this imported crop in the US (Kogan, 1991). Most New World pests of soybeans, such as Helicoverpa zea, are very mobile and often highly polyphagous. Until just recently, there was a striking absence of soybean-colonizing aphids, which are one of the most damaging of pests to this crop at its center of origin (China) (Wu et al., 2004). As inducible defenses against herbivores must be very dependent on signaling networks, such as the

8 octadecanoid pathway, within the host plant, other organisms interacting with the plant such as rhizobia may influence the degree or timing of induction. Rhizobia Rhizobia refers to a general class of gram-negative bacteria which have the ability to infect the roots or stems of legumes and provide fixed nitrogen to the host. Nitrogen fixation is accomplished through the action of the enzyme nitrogenase, which breaks the tight triple bond of atmospheric N gas, producing ammonium through a highly energetic process: N 2 + 8e - + 8H + + 16ATP ⎯→ 2NH 3 + 16ADP + 16Pi + H 2

Oxygen flow in the root nodule is tightly controlled by a plant produced O 2 -binding protein (leghemoglobin) since the nitrogenase enzyme is denatured by oxygen, but other essential cellular functions of rhizobia are aerobic (Fisher and Newton, 2002). The mutualism between rhizobia and legumes is facultative and initiated anew with each generation of host plants; therefore rhizobia exist and reproduce in a free- living state in the soil at some point, or if a suitable host cannot be found, through their entire life. Non-symbiotic individuals do not fix N, and are at a reproductive disadvantage as compared to their legume-associated kin (West et al., 2002). There is a high degree of specificity in the relationships between legume hosts and their compatible rhizobia, mediated in part by the particular chemical interactions between plant roots and rhizobia in the soil for proper nodulation (Hirsch et al., 2001). However, it is common for a single plant species to form effective associations with numerous strains of a rhizobia species, or even with strains from different genera. For example, soybeans can successfully interact with strains from Bradyrhizobium japonicum, B. elkanii, B. lianonigense, and Sinorhizobium fredii (Amarger, 2001). Many biotic and environmental factors influence the genetic composition of rhizobia populations in the soil over time (Denton et al., 2002; Streeter, 1994; Taylor et al., 1991). Rhizobia populations also become adapted to local environmental conditions and plant genotypes, with resulting effects on mutualisms (Parker, 1995; Thrall et al., 2007). The diversity of rhizobia may also be increased through the

9 horizontal exchange of genetic material, such as the transfer of symbiotic genes from introduced rhizobia to native, non-symbiotic bacteria in the field (Barcellos et al., 2007; Spratt and Maiden, 1999; Sullivan et al., 1995). One or all of these factors can contribute to high levels of phenotypic and genotypic variation found in rhizobial isolates as compared to parental inoculant strains after many years in the soil without a host plant (Batista et al., 2007). Human intervention is another factor influencing rhizobial diversity. During the long history of legume cultivation, humans have shuffled rhizobia strains throughout many regions of the world. As a result, diverse populations of soybean- compatible rhizobia are naturally present in many agricultural soils (Amarger, 2001; Ferreira and Hungria, 2002). These “indigenous” strains are now subject to the same influences on genetic composition that occur in undisturbed settings, such as adaptation to local conditions and horizontal gene transfer with native microbes and contribute to the diverse rhizobia populations which legumes may encounter in the soil. Herbivores The soybean aphid, Aphis glycines Matsumura (Hemiptera: Aphididae), is a phloem-feeding specialist of soybeans native to areas of Asia from which soybeans origniated. As with other members of this family, soybean aphids use the stylet of their piercing-sucking mouthparts to maneuver through intercellular spaces until reaching the phloem, piercing only the sieve element cells (Powell et al., 2006). Once a suitable food source has been located, the aphid remains sessile and continues to draw out phloem over a long period of time. Since its discovery in the US in 2000, the soybean aphid has since spread throughout much of the soybean growing regions of the US and Canada and is now considered “the most significant insect threat to soybean production in North America” (Ragsdale et al., 2007). Annual outbreaks of the soybean aphid have occurred since its introduction, causing yield losses between 20 and 25% and increasing pesticide usage (Ragsdale et al., 2007; Rutledge and O'Neil, 2006; Rutledge et al., 2004). While economic thresholds suggest insecticide use after 250 aphids per plant, it is common to find

10 aphid numbers per plant in the thousands (Landis et al., 2004; Ragsdale et al., 2007)(personal observations). In addition to direct damage inflicted by phloem- feeding, the soybean aphids can also transmit plant diseases such as bean yellow mosaic virus between soybean plants and other economically important viruses between non-host crops (Wang et al., 2006). Although efforts have been made to identify cultivars of soybeans with enhanced resistance against the soybean aphids (Li et al., 2006), very little is known about the mechanisms behind host plant protection. And to date, no studies have explored the possibility of controlling aphids, or other herbivores, through selective inoculation with rhizobia. Another common pest of soybeans is Helicoverpa zea Boddie (Lepidoptera: Noctuidae), a moth species native to North America. The larvae of this species use powerful mandibles to chew through foliage and are highly polyphagous – incorporating many economically important crops into its host range – and have earned an assortment of common names, such as corn earworm, cotton bollworm, and soybean podworm. Most of the agronomic problems associated with H. zea are concentrated in the southern and mid-Atlantic regions of the US where successful overwintering occurs, but it is highly dispersive and serves as a secondary crop pest for all but the most northern regions of North America (Capinera, 2000). Although a substantial portion of its pest status stems from larval damage to cotton and corn, H. zea can also cause substantial economic losses in soybeans (Herbert et al., 1991; Yu et al., 1993).

The severity of yield loss in soybeans is dependent on the timing and intensity of H. zea infestations. Extensive defoliation during the reproductive stages of growth, such as during full flower bloom, can cause up to 50% yield reductions (reviewed in (Timsina et al., 2007). Young larvae can be found in high densities in the developing leaves, while older larvae tend to feed on the forming seed pods (Kraemer et al., 1997). Additionally, soybeans can serve as reservoirs for H. zea populations which will infest other crops (Jackson et al., 2008). Development of herbivore resistant soybean cultivars have met limited success, with products not meeting other agronomic standards (Rector et al., 2000). Also, concerns over the development of resistance to insecticides, especially Bt and pyrethroids, warrant the need for more

11 sustainable means of controlling H. zea populations (Jackson et al., 2004; Pietrantonio et al., 2007). Use of microbial products in agriculture for pest control through genetic modification Basic research into the evolution and ecology of plant defenses can eventually lead to applied outcomes, particularly towards the ongoing task of developing sustainable strategies for herbivore pest management in agricultural systems. An intimate understanding of multi-level biotic interactions such as those between plants, microbes, and herbivores are essential if control methods are derived from manipulation of these interactions. One such technology which has reached widespread commercial success in some systems is the genetic modification of crop plants to incorporate insect-resistance conferred by foreign genes. Bacillus thuringiensis (Bt), a common soil bacterium, has been utilized in sustainable farming for many years to combat agricultural pests, namely Lepidopteran larvae. During sporulation, certain Bt strains produce various crystalline inclusions containing endotoxins which are lethal to specific groups of insects (Broderick et al., 2006). Through genetic modification, bacterial genes for the production of endotoxins have been successfully inserted into the genome of some key crops, particularly cotton and maize. The constitutive production of insecticidal endotoxins in plant tissues has the potential to control target pests and consequently reduce insecticide usage. Bt-maize became commercially available in 1996, and as of 2007, 49% of all maize planted in the US incorporates Bt genes (National Agricultural Statistics Service, 2007). Large-scale implementation of transgenic crop technology in a short time span has sparked substantial debate over the social, economic, and ecological implications of GM agriculture. Ecological concerns of insect-resistant transgenic crops have often focused on resistance development by pests and potential negative effects on natural enemies (Gould, 1998; Shelton et al., 2002). These issues are intricately linked to one another, because natural enemies can influence the rate that pest populations adapt to resistant plants (Gould et al., 1991). The effects of genetic

Full document contains 110 pages
Abstract: Plants are central to most community interaction webs and thus the ability to coordinate responses to many simultaneous interactions is an essential adaptation. Interaction with one organism may make the plant more vulnerable or more resistant to attacks by very different organisms. Microorganisms can form intimate associations with plants with direct effects ranging from beneficial to antagonistic, but indirect effects of plant-microbe relationships on plant interactions with other organisms are not well understood. Here, we explore the influences of microorganisms and microbial products on herbivory and resulting plant defenses. We use the legume-rhizobia mutualism as a model-system to explore herbivore-plant interactions by first characterizing the effects of rhizobial inoculation on herbivore feeding and subsequent accumulation of plant defense signaling hormones. We found interactive effects of the legume-rhizobia mutualism on plant-herbivore interactions which were dependent on both the stage of the mutualism and on the feeding style of the herbivore. Next, we explored the effects of association with different sources of rhizobia of soybean on specialist aphid populations in an agricultural setting. We found that particular rhizobia strains can confer greater resistance to their mutualist partners than others. In order to explore the effects of a rhizobia strain-specific trait on herbivory, we focused on the rhizobial product, rhizobitoxine, which can be found in host plant tissues. We found that the presumed presence of this compound decreased herbivore feeding and damage, which could be useful in pest management of legumes. Finally, we utilize genetically modified maize to further explore the application of a microbial product used for insect resistance and its effects on herbivore feeding and induction of plant defenses. Our results contribute to understanding of the evolution of host plant defenses and could facilitate the development of more sustainable management techniques for agriculture that are informed by an understanding of the chemical ecology of plants, microbes, and insects.