Survey
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
* Your assessment is very important for improving the workof artificial intelligence, which forms the content of this project
EVOLUTIONARY ECOLOGY OF ARABIDOPSIS THALIANA: INTERACTIONS WITH BIOTIC AND ABIOTIC ENVIRONMENTAL FACTORS by Adebobola O. Imeh-Nathaniel A Dissertation Submitted to the Graduate College of Bowling Green State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY DECEMBER 2012 Committee: Dr. Maria G. Bidart- Bouzat Dr. Juan L. Bouzat Dr. Ron Woodruff Dr. Dan Pavuk Dr. Maria Rizzo (Graduate Faculty Representative) ii ABSTRACT Dr. M. Gabriela Bidart-Bouzat, Advisor Previous studies have demonstrated that plants may be affected by their interactions with both biotic and abiotic environmental factors. An important biotic factor that usually affects the performance of plants is insect herbivory. In addition, abiotic factors such as light, or more specifically ultraviolet B (UVB) radiation, can also affect growth and physiological processes in plants. Arabidopsis thaliana is an annual herbaceous plant species with documented defense strategies including resistance and tolerance. Studies have also shown that these plant responses to insect herbivory may be modified by other abiotic factors such as UVB radiation. In a series of common garden experiments, the role of insect herbivory per se and in combination with UVB radiation was evaluated. Using a population of recombinant inbred lines of A. thaliana, the first experiment (described in Chapter 1) investigated the role of plant trichomes and size on patterns of plant use by an insect community. In this experiment, it was also evaluated whether insect herbivory imposes selection on trichome production and plant size. In a second experiment (described in Chapters 2 and 3), the effect of UVB radiation on plant resistance and tolerance to insect herbivory as well as on their potential fitness costs was assessed. This experiment also evaluated the role of UVB on patterns of plant utilization by insect herbivores and whether this important abiotic factor may influence plant phenotypic responses. Results from these experiments revealed that plant trichomes influenced levels of herbivore damage and plant size had an effect on colonization of plants by insect herbivores. In addition, results from selection analyses revealed that insect herbivores exerted directional selection on trichome density in A. thaliana. The second study showed that UVB radiation influenced the expression iii of resistance and tolerance as well as their associated fitness costs. Similarly, patterns of plant utilization by insect herbivores and phenotypic responses of plants to insect herbivory were modified by UVB radiation. Overall, these results highlight the importance of evaluating complex environments, including both abiotic and biotic factors, as it relates to the evolution and maintenance of traits related to plant defense against insect herbivory iv ACKNOWLEDGMENTS First, my gratitude goes to my advisor, Dr. Gabriela Bidart-Bouzat for her unrelenting help and guidance throughout my graduate studies at Bowling State University, Ohio, and more importantly, in helping me to succeed in the completion of my Ph.D. research work. I am also very grateful to all my committee members for finding the time to attend all meetings and their useful and constructive comments throughout the course of this work. In addition, I will like to express my gratitude to past and present members of Bidart-Bouzat’s lab including Lexie Majorczyk, Sean Ryan and Julie Kemarly-Hopkins. I do appreciate the help I received from the two undergraduate assistants Curtis McKinnon and Matthew Wilkinson to perform the field experiments. I would also like to acknowledge Dr. Jeff Miner for providing a conductive working environment throughout the period of the fieldwork at the BGSU Ecological Research Station and to every member of the Department of Biological Sciences, who has in one way or another contributed to the success of this study. The research conducted was supported by BGSU and a Larry and Linda Oman Fellowship. I appreciate the support I received from my siblings. I am sincerely grateful to my mother, who was always available to help with my children during the past four years. I cannot be thankful enough to my mother. To my three boys, life could not have been better without you. Thank you for your understanding and love, especially when “mummy” did not have the time to take you for soccer practice. To my husband, you are an inspiration to me. Thanks for your support, encouragement, love, and active involvement in this study. I am so blessed to have you. This dissertation will not be complete without acknowledging all friends and well-wishers. I appreciate all your concerns, encouragement, and love. Above all, I am grateful to the Alpha v and the Omega, the beginning and the ending of all things, who has made all grace sufficient for me throughout the period of my graduate work at Bowling Green State University. vi TABLE OF CONTENTS Page CHAPTER 1. INTRASPECIFIC GENETIC VARIATION IN PLANT SIZE AND TRICHOME DENSITY INFLUENCES PATTERNS OF PLANT USE BY AN INSECT COMMUNITY Abstract…………………………………………………………………………….. 1 Introduction .……………………………………………………………………...... 3 Methods .…………………………………………………………………………… 5 Results ……………………………………………………………………………… 10 Discussion………………………………………………………………………….. 12 Tables and Figures…...……………………………………………………………... 17 References……….………………………………………………………………….. 25 CHAPTER 2. CHANGES IN ULTRAVIOLET B RADIATION MODIFY RESISTANCE AND TOLERANCE IN ARABIDOPSIS THALIANA TO INSECT HERBIVORY Abstract……………………………………………………………………………... 33 Introduction………………………………………………………………………… 34 Methods……………………………………………………………………………... 38 Results………………………………………………………………………………. 43 Discussion…………………………………………………………………………... 45 Tables and Figures ……………………………………………………………......... 50 vii References ………………………………………………………………………….. 56 CHAPTER 3. EFFECTS OF ULTRAVIOLET-B RADIATION ON PLASTICITY OF FITNESS-RELATED RESPONSES OF ARABIDOPSIS THALIANA TO INSECT HERBIVORY Abstract…………………………………………………………………………....... 67 Introduction…………………………………………………………………………. 69 Methods…………………………………………………………………………….. 72 Results……………………………………………………………………………… 75 Discussion………………………………………………………………………….. 77 Tables and Figures …...…………………………………………………………….. 82 Reference ……….………………………………………………………………….. 85 viii LIST OF TABLES Table Page CHAPTER 1 1 ANOVA F values showing the presence of genetic variation in measured phenotypic characters…………………………………………. 2A 17 Analysis of variance for relative fitness showing the effects of insect herbivores on directional selection for trichomes in A. thaliana………….. 18 2B Analysis of variance for relative fitness showing the effects of insect herbivores on stabilizing/disruptive selection for trichomes in A. thaliana…………………………………………………………………………... 18 CHAPTER 2 1 ANOVA F-values showing the effects of herbivory, UVB radiation, RILs and their interactions on trichome density (direct resistance) and an indirect resistance estimate (1-herbivore damage)………………………. 2 50 ANCOVA F-values showing the relationship between relative fitness and rosette damage across UVB environments and in each of the UVB environments……………………………………………………………... 51 3 Pearson’s correlation coefficients showing the relationship between relative fitness of undamaged plants and tolerance under ambient (UVB+) and excluded (UVB-) ultraviolet B radiation level covariance… 52 ix CHAPTER 3 1 Mixed model analysis of variance showing the effects of UVB radiation, insect herbivory and RIL on plant phenotypic variables of A. thaliana…. 82 x LIST OF FIGURES Figure Page CHAPTER 1 1 Picture of Arabidopsis thaliana plant showing the presence of 19 trichomes 2 A split- plot experimental design with two levels of herbivory 20 (herbivory and sprayed) distributed in a checkerboard pattern among ten plots 3 Analysis of covariance showing the relationship between trichome density or plant size and plant utilization patterns by insect herbivores (i.e. days to colonization and herbivore damage).................................... 21 4 Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on trichome density in the presence and absence of insect herbivores …………………………..... 5 22 Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on plant size in the presence of insect herbivores.................................................................. 6 23 Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on number of days to first herbivore colonization and insect herbivore damage………………… 24 xi CHAPTER 2 1A Picture of the experimental plots showing plant arrangements within each plot ………………………………………………………………. 1B 53 Schematic representation of experimental plot layout showing two levels of herbivory (herbivory and sprayed) and UVB (ambient and excluded) treatments distributed among twenty plots…………………. 53 2 Absorption spectrum for Mylar and Clear acetate filters……………… 3 Effects of the presence (UVB+) and exclusion (UVB-) of ultraviolet B 54 radiation on trichome density, indirect resistance, and tolerance of an A. thaliana recombinant inbred line population …………………… 4 55 Genotypic correlations showing fitness costs of trichomes, resistance , and tolerance in A. thaliana plants grown under ambient and excluded UVB radiation…………………............................................................ 5 56 Genotypic correlations between resistance and tolerance in A. thaliana plants grown under ambient and excluded UVB radiation……………. 57 CHAPTER 3 1 Reaction norms showing A. thaliana phenotypic responses to UVB radiation and insect herbivory, in terms of total fruit production , rosette biomass, and number of days to first flowering ………………. 2 83 Reaction norms showing A. thaliana phenotypic responses, in terms of days to herbivore colonization and rosette damage …………………. 86 1 CHAPTER 1 INTRASPECIFIC GENETIC VARIATION IN PLANT SIZE AND TRICHOME DENSITY INFLUENCES PATTERNS OF PLANT USE BY AN INSECT COMMUNITY ABSTRACT Intraspecific genetic variation in plant traits can have important consequences for the insect community utilizing plant resources for feeding and reproduction. For example, variation in a plant structural defensive trait such as trichome density can influence levels of herbivore damage. In addition, variability in plant size may affect host choice by insect herbivores since this trait can serve as a visual cue for insects to locate a plant host species as well as provide variable surface areas for insect oviposition and feeding. Despite the importance of these plant traits, little is known regarding how genetic variation may influence plant damage levels and colonization of plants by a natural insect community. Similarly, few studies have evaluated whether insect herbivores impose selection on trichome production and plant size. Results from a common garden experiment revealed the presence of genetic variation in trichome density and plant size in a population of recombinant inbred lines of Arabidopsis thaliana. Genetic variation in these traits significantly influenced both colonization of plants by insect herbivores and levels of herbivore damage among plants. Selection analyses revealed directional selection favoring a reduction in trichome density in the absence of insect herbivory (sprayed plots), which may be indicative of a fitness cost related to the production of this defensive trait. In terms of plant size, selection favored larger plants in field plots exposed to insect herbivory. These results corroborate the role of plant trichomes as a resistant trait favored by selection in the presence of 2 herbivory and emphasize the understudied role of plant size as an important cue in herbivore colonization of plants in the field. 3 INTRODUCTION Plants have evolved different morphological structures such as trichomes and thorns, as well as a wide variety of secondary chemical compounds including glucosinolates and phenolics, which act as defense mechanisms against insect herbivore attacks (Levin 1973; Zangerl and Bazzaz, 1992; Strauss and Agrawal, 1999; Rautio et al., 2002). Although chemical responses of plants to insect herbivores have been widely studied (Bennett and Wallsgrove 1994; McConn et al. 1997; Clauss et al. 2006; McConn et al. 1997; Kliebenstein 2004; Bidart-Bouzat et al. 2005), little is known regarding how genetic variation in structural defensive traits may influence plant damage levels and colonization of plants by natural insects. Trichomes are hair-like outgrowths derived from the epidermal cells of plants that may have defensive function against herbivore attacks (Figure1). Despite potentially important roles of trichomes in plant defense, most of the evidence suggesting that trichomes may be involved in defense were obtained from studies conducted under laboratory conditions using insect herbivores or through simulated herbivory (Traw and Bergelson 2003; Clauss et al. 2006). Thus, these studies may not directly reflect the role of trichomes on natural levels of insect herbivory. In addition, more information is required on how genetic variation in trichome density influence patterns of plant utilization by insect herbivores in natural populations. This is important because the ability of plants to respond to herbivory in an evolutionary manner will mostly depend on the availability of genetic variation in traits related to fitness. Trichomes occur in different parts of the plants such as stems, leaves, and reproductive organs, and vary within and/or between populations (Strauss and Agrawal 1999; Handley et al. 2005). They may also vary in their structure, ranging from unicellular to multicellular, and 4 anatomy, which may be branched, unbranched, spiral or hooked (Ouwerkerk et al. 1999; Hulskamp 2000). In addition, trichomes can be characterized as glandular or non-glandular based on their ability to secrete chemical materials such as terpenes, alkaloids, or fatty acid derivatives (Schilmiller et al. 2008; Wang et al. 2008). The presence of trichomes in many plant species has been associated with many physiological and defensive functions. Leaf trichomes may reduce insect herbivory by making leaves less attractive to herbivores, trapping small insects, and reducing mobility and nutritional quality of leaves, which may in turn, influence oviposition rates, egg hatching, larval growth, and survival. Glandular trichomes also produce exudates that may be toxic or serve as repellant to insect herbivores (Perez-Estrada et al. 2000; Simmons et al., 2004; Wang et al. 2004; Handley et al. 2005). In addition to the role of trichomes in defense against herbivores, previous studies have also shown that trichomes may be beneficial to herbivores and pathogens trapping natural enemies of insect herbivores (Turlings and Benrey 1998; Lovinger et al. 2000; Kennedy 2003). Apart from the role of trichomes in plant-herbivore interactions, there is evidence that they play multiple functions in plants other than that of defense against herbivores. For example, leaf trichomes may serve the purpose of resistance to other abiotic stress such as drought by reflecting light and reducing evapo-transpiration rates or by facilitating condensation of moisture onto the plant surface (Ehleringer 1984; Larcher 2001), as well as removal of excess salt (Gravano et al. 1998) and heavy metals (Sarret et al. 2006). In addition, trichomes may protect living cells from the harmful effects of ultraviolet radiation and extreme temperature changes (Skaltsa et al. 1994; Werker 2000; Agrawal et al. 2004). Since all of these functions 5 may act in concert or separately in different plant species, experimental approaches are required to determine whether trichomes play an important role in plant resistance to insect herbivory. Although the presence of trichomes can reduce levels of herbivore damage (Clauss et al. 2006; Handley et al. 2005), less information is known about the role of other phenotypic traits such as plant size on plant utilization patterns by insect herbivores in the field. Plant size is an important phenotypic trait that can affect not only the levels of damage to plants, but also host choice by herbivores. This is because herbivores use both olfactory and visual cues to choose host plants for oviposition (Singer 2004; Costa et al. 2009; Reeves et al. 2009). Several studies have also suggested that larger plants are more likely to be first (and more often) colonized than smaller plants, since plant size may be a reflection of vigor and nutrient availability, which in turn may enhance herbivore performance (Heisswolf et al. 2005; Lastra et al. 2006). It is also possible that bigger plants would be a better choice for oviposition purposes due to their ability to provide a larger surface area for oviposition, especially in herbivores that produce large amounts of eggs. For example, herbivores may deposit a cluster of eggs on a host plant that has a large surface area for oviposition (Wahlberg et al. 2004; Saastamoinen 2007), whereas less eggs per plant may be deposited on plants with smaller surface areas (Courtney 1982). Thus, time (and costs) expended by insects on searching for suitable plants for oviposition may be reduced when insects locate larger sized plants with larger surface areas (Nieminen et al. 2003; Kuussaari et al. 2004; Singer 2004). In a common garden experiment, the presence of natural variation in trichome density and plant size was assessed in a population of recombinant inbred lines of Arabidopsis thaliana. In addition, it was determined whether this variation influenced the patterns of plant use by 6 insect herbivores in the field (i.e., order of herbivore colonization and levels of herbivore damage). It was also tested whether variation in trichome density, plant size, and herbivore damage levels was significantly associated to plant fitness. Evaluation of such link between the trichome density, levels of herbivore damage, and plant fitness is important because differences in herbivore damage may impose selection for different types of defensive traits. Finally, this experiment evaluated whether selection was acting on plant size and trichome density, and whether insect herbivores were indeed an important selective pressure on these traits. As summarized in Mauricio and Rausher (1997), three conditions should be met for showing that a plant character that influences levels of insect herbivory (e.g., trichome density or plant size), is under selection by the herbivores. First, the character has to be genetically variable (potential to evolve). Second, variation in the studied character must influence levels of herbivore damage, and has to be associated with plant fitness. Lastly, it has to be demonstrated that the presence of insect herbivores indeed influences selection estimates; that is, that upon removal of the selective pressure imposed by insect herbivory, estimates of selection are modified. In this experiment, selection on plant size was measured only in the presence of herbivores but selection on trichomes was evaluated in both the presence and absence of herbivores. Specifically, insect herbivores were removed from half of the field plots, and potential changes in the patterns of selection on trichome density were evaluated by comparing selection gradients for this character in the presence and absence of insect herbivores. 7 MATERIALS AND METHODS Study Plant Arabidopsis thaliana (L.) Heynh (Brassicaceae) commonly known as the mouse ear cress is an annual herbaceous weed, which is native to Europe and Asia but has spread its distribution to North America and Africa. This plant species is naturally found in open or disturbed habitats, sandy soils, riverbanks, roadsides, rocky slopes, and meadows (Baskins and Baskins 1983). Arabidopsis thaliana has a short life cycle lasting between 6-8 weeks from germination to senescence and it is a mostly self-fertilizing species with less than 1% outcrossing rates (Hoffmann et al. 2003). Trichomes in A. thaliana are usually unicellular, branched, and non-glandular (Hulskamp and Kirik 2000). In this study, we used a population of recombinant inbred lines (100 RILs), which was derived from a cross between the ecotypes Columbia (Col-0) and Landsberg (Ler), and advanced to the eighth generation through a single seed descent (Lister and Dean 1993). Arabidopsis thaliana RILs from this cross were used because a previous study showed variation in trichome density between the two parental lines, Col-0 and Ler ecotypes (Mauricio 2005). Therefore, the use of this recombinant inbred line population allows evaluating whether genetic variation in trichome density may influence natural patterns of insect herbivory. Experimental Design and Plant Growth Conditions The experimental design consisted of a split-plot with two levels of herbivory treatment (i.e., natural insect herbivores present and excluded) assigned to ten plots distributed in a checkerboard pattern (Figure 2). Each of these treatments was replicated in five plots accounting for a total of ten plots. Naturally occurring insect herbivores were excluded in control plots using an insecticide (Proclaim by Syngenta Inc. USA) at intervals of ten days throughout the 8 experimental period. Three replicates of the 100 RILs were randomly assigned to each of the 10 plots (total of 3000 plants). For both treatments, mammalian herbivores were excluded by using wire mesh fencing. Several species of insect herbivores are known to attack A. thaliana plants in the field. Bidart-Bouzat and Kliebestein (2008) reported the presence of diamondback moths (Plutella xylostella; Plutellidae), striped and western cabbage flea beetles (Psylloides striolata and Psylloides pusilla, respectively; Chrysomelidae), cabbage aphids (Brevicoryne brassicae; Aphididae) and onion thrips (Thrips tabaci; Thripidae) as natural herbivores observed feeding on different A. thaliana genotypes. Other studies (Mauricio 1998; Mosleh Arany et al. 2005; Arany 2008) have also reported other insect herbivores feeding on filed-grown A. thaliana plants including the cabbage looper Tricoplusia ni (Noctuidae), Spodoptera exigua (Noctuidae), European cabbage butterfly Pieris rapae (Pieridae), and specialist weevils Ceutorhynchus atomus and C. contractus (Curculinoidae). During this field study, the most common herbivore observed attacking A. thaliana plants was the specialist diamondback moth (Plutella xylostella). The generalist cabbage looper (Tricoplusia ni) and the western cabbage flea beetles Psylloides pusilla were also observed feeding on the plants, but causing minor effects. In terms of growth conditions, seeds were cold-treated for 5 days at 4°C to break dormancy, and then, transferred to growth chambers for germination. Within the chambers, plants were exposed to 10 hours of light and 14 hours of dark cycles, with day and night temperatures of 20°C and 16°C, respectively. Plants were rotated within growth chambers every other day and among chambers every week to avoid chamber and position effects. After five 9 weeks of growth in the chambers, plants were transplanted to the field (Ecological Research Station, Bowling Green State University, Bowling Green, Ohio). Measurement of Plant Traits In this experiment, the following plant traits were measured: trichome density, rosette diameter, and total fruit number. Trichome density was measured using digital images of the youngest fully extended rosette leaf per plant at bolting time. Leaf area was estimated using image J software (rsbweb.nih.gov/ij/). Trichome density was calculated as the total number of trichomes (on the upper side of the leaf) divided by leaf area. In order to evaluate if trichomes are inducible in the presence of herbivores, we compared trichome density in sprayed plants to plants exposed to insect herbivores. To estimate plant size, rosette diameter was measured as the widest length of the plant rosette. A previous study has shown rosette diameter to be strongly correlated with rosette area, a measure of plant size (Mauricio 1998). Fitness was estimated as the lifetime fruit production and relative fitness as absolute fitness divided by the mean absolute fitness per herbivory level. Total fruit production was previously shown to be strongly correlated with total seed production in Arabidopsis thaliana (Mauricio and Rausher 1997; Bidart-Bouzat 2004). Since selection operates at a population level, and this selection is based on the differences in fitness between individuals; therefore, the use of relative fitness is more appropriate than using absolute fitness (Cornner and Hartl 2004). To assess whether plant genotype affected the order of colonization of insect herbivores among plants, all plants were inspected daily for insect damage and the numbers of days to first observed herbivore damage on each plant was recorded herein, referred to as colonization. To estimate insect herbivore damage to the plant rosette leaves (i.e., percent area removed by 10 herbivores), a qualitative leaf damage index adapted from Bidart-Bouzat et al. (2005) was used. This index was calculated by assigning leaves to three size categories based on leaf length (i.e., small: 0- 3cm, medium: 3 - 6cm, and large: 6-9cm), and three damage categories based on the amount of tissue removed (i.e., < 25%, 25-75 %, and > 75). Statistical Analysis To evaluate the effects of RILs (i.e., genotype) and block on trichome density, rosette diameter, days to first observed herbivore damage, rosette damage, and total number of fruits, analyses of variance using a linear random model (SAS statistical software, version 9.1) were performed. Data was either log- or ranked-transformed to correct for normality and heteroscedasticity, when necessary. In addition, we performed a series of analyses of covariance to infer whether the associations between the response variables: herbivore colonization and damage, and the covariates: trichome density and rosette diameter (estimate of plant size) were affected by the plant genotype (RIL) using the model: Y = covariate + RIL + covariate × RIL + error. To evaluate patterns of selection on trichome density, plant size, order of herbivore colonization and rosette damage, a regression analysis was used to estimate directional (β) and stabilizing/disruptive (γ) selection gradients. In addition, it was determined whether herbivores were exerting selection on trichome density, by performing analyses of variance on the directional and stabilizing selection gradients to compare patterns of selection on trichome density in the presence and absence of natural herbivores (Lande and Arnold 1983; Mauricio and Rausher 1997). For this purpose, a mixed linear (for directional selection) and quadratic (for stabilizing selection) models were used, considering the herbivory treatment as a fixed effect and 11 RIL as a random effect. The effect of block and its interactions were removed from this model because they were not significant. RESULTS Analyses of variance revealed that all traits measured in this study were highly variable (Table 1). Effects of RIL and herbivory on trichome density were significant (P= <0.0001) and plant responses to herbivory were found to be dependent on the RIL (i.e., Herbivory × RIL interaction: P= 0.0233). In addition, the effect of RIL on plant size, days to herbivore colonization, and rosette damage were all statistically significant. Analyses of covariance showed that variation in trichome density significantly influenced herbivore damage levels but it did not affect days to colonization of plants by insect herbivores (Figures 3A and 3B). On the other hand, plant size (rosette diameter) significantly influenced both the number days to herbivore colonization and herbivore damage levels (Figures 3C and 3D). In any case, the effect of RIL (plant genotype) was significant in all ANCOVA models. In addition, the relationship between the covariates and both damage and colonization was not significantly affected by RIL (Figures 3A-D). These results suggest that trichome density and plant size may play an important role in plant susceptibility/resistance to insect herbivory as well as in determining patterns of plant colonization by insect herbivores in the field. Regarding the patterns of selection on variables related to plant size and trichome density, directional regression analyses revealed that relative fitness was inversely associated with trichome density in sprayed plots but not in the plots exposed to insect herbivory (Figure 4). In other words, selection favored genotypes with lower trichome density in sprayed plots but not in 12 plants exposed to insect herbivores. Conversely, plant fitness was higher as plant size increased (Figure 5), which implies that genotypes that produce larger plants are favored by selection. There was no evidence of stabilizing or disruptive selection acting on either plant size or trichome density. In addition, regression analyses did not indicate any form of selection acting on days to herbivore colonization (Figure 6A); there was a borderline significant selection differential however, indicating directional selection favoring a decrease in rosette damage (Figure 6B). In order to detect whether insect herbivores are exerting selection on trichome density, we compared general patterns of selection on trichome density in the presence and absence of natural herbivores using both directional and stabilizing gradients of selection (Tables 2A and 2B). Results revealed that the effect of herbivory on directional selection for trichome density (i.e., trichome × herbivory interaction) was significant but the quadratic term for trichome density was not significant, suggesting that insect herbivores are exerting directional selection but not stabilizing selection on trichome density. DISCUSSION Herbivores are known to be attracted to their host plants through visual and olfactory cues (Costa et al. 2009; Reeves et al. 2009; Singer 2004). These cues include plant volatiles (Costa et al. 2009; Szendrei and Isaacs 2005), plant size (Handley et al. 2005) , shape, orientation and plant age, which may in turn influence the type and quantity of volatiles released (Moreau et al. 2003; Vergés et al. 2007; Wentzell and Kliebenstein 2008). This study assessed the role of trichome density and plant size in determining patterns of plant utilization by insect herbivores in 13 the field. Results from this study showed that larger plants were the first to be colonized by insect herbivores. This result further supports previous studies that have shown that plant size may be important in host preference (Nieminen et al. 2003; Kuussaari et al. 2004; Heisswolf et al. 2005; Lastra et al. 2006). First, larger plants may be colonized in preference of smaller plants because they are visually more conspicuous to ovipositing females when compared with smaller plants. Second, larger plants may be advantageous to herbivores in that they may be more suitable for larval development. This is because larvae movement may be restricted during early instars; thus, herbivores may show more preference for a plant that provides larger surface area and more nutrients (Nieminen et al. 2003; Kuussaari et al. 2004). In addition, the plant vigor hypothesis also predicts that vigorous plants with larger size will be preferred by herbivores because they are more likely to have higher biomass and nutritional quality than smaller plants (Heisswolf et al. 2005; Lastra et al. 2006) Furthermore, this study experimentally demonstrated the role of trichomes in resistance against insect herbivores in the annual crucifer A. thaliana. Herbivore damage levels significantly decreased with increased trichome production. In addition, the exclusion of herbivores resulted in a change in selection on trichome production, demonstrating that insect herbivory is indeed an agent of selection on this character. These results are in agreement with other studies that have shown that the presence of trichomes may lead to a reduction in the levels of insect damage. For example, Mauricio (1988), in a field experiment, demonstrated that leaf trichomes led to a decrease in herbivore damage in 144 natural populations of A. thaliana. Similarly, leaf trichomes have been found to be associated with reduced level of herbivory in the closely related Arabidopsis lyrata (Handley et al. 2005; Loe et al. 2007), as well as other 14 members of the Brassicaceae plant family (Lamb 1980; Agren and Schemske 1993). With respect to plant size, larger plants had more damage than smaller plants. This further supports the earlier observation that insect herbivores may link larger size to higher nutritional quality for larval development. Additionally, larger plant size may indirectly imply better larval performance by making them less vulnerable to plant defense traits if, as previously suggested, more resources are allocated to vegetative growth at the expense of allocation to defense (Obeso 2002). In order to investigate patterns of selection on trichome density, plant size, and plant utilization by insect herbivores, regression analyses of these variables on relative fitness were performed. Directional selection on trichome density was observed, which favored a decrease in the production of this defensive character in sprayed plots. Conversely, no evidence of selection acting on this character was found in plots exposed to insect herbivores. This result suggests the existence of a cost of trichome production in the absence of insect herbivory (Bergelson and Purrington 1996; Mauricio 1998). It is interesting to note that in contrast to what was expected, there was no evidence of directional selection for increase in trichome production in the presence of herbivory. In other words, since results showed that trichomes reduced levels of insect herbivore damage and that there was negative selection acting against herbivore damage, then, it was expected that selection would favor an increase in defense related characters such as trichomes in the presence of herbivores (Valverde et al. 2001). The lack of positive selection for trichome density in the presence of insect herbivores may imply the presence of a cost of resistance; that is, production of trichomes is expensive and the benefit derived in terms of resistance may not compensate for its cost. This may also explain the observed directional 15 selection for decreased trichome production in the sprayed plots when the benefit of resistance to herbivory was absent. This result was in agreement with a previous study by Mauricio (1998) in which trichome density was negatively associated with relative fitness, indicating the presence of a fitness cost for trichome production (but see Valverde 2001). Another possible explanation for this result could be that the levels of herbivore damage observed were not large enough to reverse the direction of selection on trichome production in the A. thaliana populations used in this study. Alternatively, there may have not been enough time for the A. thaliana population to respond to the presence of natural herbivores (Mauricio and Rausher 1997). This is because genotypes used in this study are recombinant inbred lines, which are not originated from the study area. Therefore, it may require a longer period of time in order to observe a shift in the direction of selection in this population. Anyhow, the direction of selection did show a shift towards positive selection, but the shift was not sufficiently large to be significant. Taken these results together, we predict that the cost of resistance may be acting to slow down the shift in the direction of selection for trichome density and that given a longer period of time under the same environmental conditions this selection gradient may become positive. In spite of the positive association between plant size and both herbivore damage and colonization, selection analyses revealed that selection favored larger plant size. Selection for larger plant size suggests that mechanisms leading to tolerance of insect herbivory may be favored (Paige 1992; Juenger and Bergelson 2000; Rautio et al. 2002; Freeman et al. 2003; Gonzalez-Teuber and Gianoli 2007); for example, by increased vegetative biomass growth compensating for plant tissue lost to herbivory. This is an alternative strategy of plants to minimize the detrimental effects of herbivory on fitness. Paige and Whitham (1987) have 16 documented this phenomenon of compensation in the biennial Ipomopsis aggregata, which showed a release of apical meristem dominance and the proliferation of lateral branches when browsed by mammalian herbivores like mule deer and elk. This study also demonstrated that insect herbivores impose selection on trichome density, which was documented by a significant effect of insect herbivores in modifying patterns of directional selection for trichome density. To date, only two studies have demonstrated this criterion for trichome density in the field, despite the claim that co-evolution is the main driving force for the evolution of resistance to insect herbivory (Mauricio and Rausher 1997; Valverde et al. 2001). This is because any claim that an environmental factor is having an evolutionary impact on a character must be able to demonstrate the existence of genetic variation in the character of interest, a link between variation in the resistance character and fitness, and most importantly, that the presence of the selective agent is imposing direct selection on the character. Two previous studies (Mauricio and Rausher 1997; Valverde et al. 2001) as well as the present study have experimentally manipulated the presence of the selective agent (insect herbivores) to show how its absence influenced patterns of selection on trichome density. Although this study does not rule out the possibility of other environmental factors simultaneously imposing selection on trichomes, it does support the idea that insect herbivores are indeed important selective agents influencing the evolution of defense structures in plants. 17 Table 1. ANOVA-F values showing the presence of genetic variation (RIL factor) in measured phenotypic characters. Source of variation Block Herbivory RIL Herbivory x RIL Trichomes Plant size Days to colonization 20.90*** 21.67*** 0.08 21.16*** 2.47*** 3.36*** 1.34* 1.31* - Significance levels * P<0.05, ** P<0.01, *** P<0.001 Rosette damage 16.01*** 1.81*** - 18 Table 2A. Analysis of variance for relative fitness showing the effects of insect herbivores on directional selection for trichomes in A. thaliana. Source of variation RIL Herbivory Trichome Density Herbivory x Trichome Density Error F-value 1.61 0.48 0.03 3.94 P- value 0.0095 0.4900 0.8654 0.0499 Table 2B. Analysis of variance for relative fitness showing the effects of insect herbivores on stabilizing/disruptive selection for trichomes in A. thaliana. Source of variation RIL Herbivory Trichome Density Trichome Density sq Herbivory x Trichome Density sq Error F-value 1.52 0.56 0.20 0.10 0.18 P-value 0.0211 0.4546 0.6538 0.7579 0.6713 19 Figure 1. Picture of Arabidopsis thaliana plant showing the presence of trichomes 20 A B C D E HERBIVORY SPRAYED A B C D E Figure 2. A split- plot experimental design with two levels of herbivory (herbivory and sprayed) distributed in a checkerboard pattern among ten plots. Different letters indicate different plots per treatment. 21 Source of variation RIL Trichomes Trichomes*RIL (A) Days to Colonization 16 F-value 1.43** 0.84 0.93 Source of variation RIL Rosdiameter Rosdiameter*RIL (C) F- value 1.39* 7.34** 0.91 14 12 10 8 Rosette Damage (%) 30 Source of variation RIL Trichomes Trichomes*RIL (B) F- value 1.65*** 6.89*** 1.22* Source of variation RIL Rosdiameter Rosdiameter*RIL (D) F- value 1.66*** 7.12** 1.17 20 10 150 200 250 300 Trichome Density 350 5 6 7 8 Plant Size (cm) Figure 3. Analysis of covariance showing the relationship between trichome density and days to colonization (A) or rosette damage (B), and the relationship between rosette diameter and either days to colonization (C) or rosette damage (D). * P<0.05, ** P<0.01, *** P<0.001 22 (A) Relative Fitness 1.6 β = -0.01024 γ = -0.05079 1.4 1.2 1.0 0.8 0.6 -2 0 2 Standadized Trichome Density (B) Relative Fitness 1.6 β = -0.10302* γ = 0.01621 1.4 1.2 1.0 0.8 0.6 -2 -1 0 1 2 Standardized Trichome Density Figure 4. Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on trichome density in the presence (A) and absence (B) of insect herbivores. * P<0.05 23 Relative Fitness 1.6 β = 0.16651*** γ = 0.01562 1.4 1.2 1.0 0.8 0.6 -2 0 2 Standardized Plant Size Figure 5. Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on plant size in the presence of insect herbivores. *** P<0.001 24 (A) Relative Fitness 1.6 β = 0.00849 γ = 0.05738 1.4 1.2 1.0 0.8 0.6 -2 -1 0 1 2 Standardized Days to Colonization (B) Relative Fitness 1.6 β = -0.08635† γ = 0.03505 1.4 1.2 1.0 0.8 0.6 -2 0 2 4 Standardized Rosette Damage Figure 6. Directional (β) and stabilizing/disruptive (γ) selection gradients estimated by regression selection analyses on number of days to first herbivore colonization (A) and insect herbivore damage (B). † P<0.1 25 REFERENCES Agrawal AA (2004) Resistance and susceptibility of milkweed: Competition, root herbivory, and plant genetic variation. Ecology 85:2118-2133 Agren J, Schemske DW (1993) The cost of defense against herbivores: An experimental study of trichome production in Brassica rapa. The American Naturalist 141:338-350 Arany AM et al. (2008) Glucosinolates and other metabolites in the leaves of Arabidopsis thaliana from natural populations and their effects on a generalist and a specialist herbivore. Chemoecology 18:65-71 Bennett RN, Wallsgrove RM (1994) Secondary Metabolites in Plant Defense Mechanisms. New Phytologist 127:617-633 Berenbaum M (1983) Coumarins and Caterpillars - a Case for Coevolution. Evolution 37:163179 Bergelson J, Purrington CB (1996) Surveying patterns in the cost of resistance in plants. American Naturalist 148:536-558 Bernays EA, Graham M (1988) On the evolution of host specificity phytophagus arthropods. Ecology 69:886-892 Bidart-Bouzat MG (2004) Herbivory modifies the lifetime fitness response of Arabidopsis thaliana to elevated CO2. Ecology (Washington D C) 85:297-303 Bidart-Bouzat MG, Imeh-Nathaniel A (2008) Global Change Effects on Plant Chemical Defenses against Insect Herbivores. Journal of Integrative Plant Biology 50:1339-1354 Bidart-Bouzat MG, Mithen R, Berenbaum MR (2005) Elevated CO2 influences herbivoryinduced defense responses of Arabidopsis thaliana. Oecologia (Berlin) 145:415-424 26 Calo L, Garcia I, Gotor C, Romero LC (2006) Leaf hairs influence phytopathogenic fungus infection and confer an increased resistance when expressing a Trichoderma alpha-1,3glucanase. Journal of Experimental Botany 57:3911-3920 Clauss MJ, Dietel S, Schubert G, Mitchell-Olds T (2006) Glucosinolate and trichome defenses in a natural Arabidopsis lyrata population. Journal of Chemical Ecology 32:2351-2373 Conner JK, Hartl DL (2004) A Primer of Ecological Genetics. Sinauer Associates, Inc., Sunderland, Massachusetts Costa JF, Cosio W, Cardenas M, Yabar E, Gianoli E (2009) Preference of quinoa moth: Eurysacca melanocampta meyrick (Lepidoptera: Gelechiidae) for two varieties of quinoa (Chenopodium quinoa WILLD.) in olfactometry assays. Chilean Journal of Agricultural Research 69:71-78 Courtney SP (1982) Coevolution of Pierid Butterflies and Their Cruciferous Food plants: Crucifer Apparency and Anthocharis-Cardamines (L) Oviposition. Oecologia 52:258265 Ehleringer J (1984) Ecology and ecophysiology of leaf pubescence in North American desert plants. Plenum Press, New York. Freeman RS, Brody AK, Neefus CD (2003) Flowering phenology and compensation for herbivory in Ipomopsis aggregata. Oecologia (Berlin) 136:394-401 Gonzalez-Teuber M, Gianoli E (2007) Tolerance to simulated herbivory in two populations of Convolvulus chilensis (Convolvulaceae). Acta Oecologica 32:119-123 Gravano E, Tani C, Bennici A, Gucci R (1998) The ultrastructure of glandular trichomes of Phillyrea latifolia L. (Oleaceae) leaves. Annals of Botany 81:327-335 27 Handley R, Ekbom B, Agren J (2005) Variation in trichome density and resistance against a specialist insect herbivore in natural populations of Arabidopsis thaliana. Ecological Entomology 30:284-292 Heisswolf A, Obermaier E, Poethke HJ (2005) Selection of large host plants for oviposition by a monophagous leaf beetle: nutritional quality or enemy-free space? Ecological Entomology 30: 299-306. Herms DA, Mattson WJ (1992) The Dilemma of Plants - to Grow or Defend. Quarterly Review of Biology 67:283-335 Hoffmann MH, Bremer M, Schneider K, Burger F, Stolle E, Moritz G (2003) Flower visitors in a natural population of Arabidopsis thaliana. Plant Biology 5:491-494 Holeski LM (2007) Within and between generation phenotypic plasticity in trichome density of Mimulus guttatus. Journal of Evolutionary Biology 20:2092-2100 Hulskamp M (2000) Cell morphogenesis: How plants split hairs. Current Biology 10: 308- 310 Hulskamp M, Kirik V (2000) Trichome differentiation and morphogenesis in Arabidopsis. In: Advances in Botanical Research Incorporating Advances in Plant Pathology. pp 237-260 Juenger T, Bergelson J (2000) The evolution of compensation to herbivory in scarlet gilia, Ipomopsis aggregata: Herbivore-imposed natural selection and the quantitative genetics of tolerance. Evolution 54:764-777 Kennedy GG (2003) Tomato, Pests, parasitoids, and predators: Tritrophic interactions involving the genus Lycopersicon. Annual Review of Entomology 48:51-72 Kliebenstein DJ (2004) Secondary metabolites and plant/environment interactions: a view through Arabidopsis thaliana tinged glasses. Plant Cell and Environment 27:675-684 28 Kuussaari M, Van Nouhuys S, Hellmann JJ, Singer MC (2004) Larval biology of checkerspots. In: On the Wings of Checkerspots: A Model System for Population Biology Ehrlich P.R. & Hanski I. (eds). Oxford University Press, Oxford, pp. 138–160. Lamb RJ (1980) Hairs protect pods of mustard (Brassica hirta Gisilba) from flea beetle feeding damage. Canadian Journal of Plant Science 60:1439–1440 Lambrix V, Reichelt M, Mitchell-Olds T, Kliebenstein DJ, Gershenzon J (2001) The Arabidopsis Epithiospecifier Protein Promotes the Hydrolysis of Glucosinolates to Nitriles and Influences Trichoplusia ni Herbivory. Plant Cell 13:2793-2807 Lande R, Arnold SJ (1983) The measurement of selection on correlated characters. Evolution 37:1210-1226 Larcher W (2001) Physiological Plant Ecology. Springer-Verlag, Berlin Lastra JA, Barrios LE, Rojas JC, and Rivera HP (2006). Host selection behavior of Leptophobia aripa (Lepidoptera: Pieridae). Florida Entomologist 89:127–134. Levin DA (1973) The role of trichomes in plant defense. The Quarterly Review of Biology 48:3-15 Li Q, Eigenbrode SD, Stringam GR, Thiagarajah MR (2000) Feeding and growth of Plutella xylostella and Spodoptera eridania on Brassica juncea with varying glucosinolate concentrations and myrosinase activities. Journal of Chemical Ecology 26:2401-2419 Lister C, Dean C (1993) Recombinant Inbred Lines for Mapping Rflp and Phenotypic Markers in Arabidopsis thaliana. Plant Journal 4:745-750 Loe G, Torang P, Gaudeul M, Agren J (2007) Trichome production and spatiotemporal variation in herbivory in the perennial herb Arabidopsis lyrata. Oikos 116:134-142 29 Lovinger A, Liewehr D, Lamp WO (2000) Glandular trichomes on alfalfa impede searching behavior of the potato leafhopper parasitoid. Biological Control 18:187-192 Marquis RJ (1992) Selective impacts of herbivores. University of Chicago press, Chicago Mauricio R (1998) Costs of resistance to natural enemies in field populations of the annual plant Arabidopsis thaliana. American Naturalist 151:20-28 Mauricio R (2001) An ecological genetic approach to the study of coevolution. American Zoologist 41:916-927 Mauricio R (2005) Ontogenetics of QTL: the genetic architecture of trichome density over time in Arabidopsis thaliana. Genetica 123:75-85 Mauricio R, Rausher MD (1997) Experimental manipulation of putative selective agents provides evidence for the role of natural enemies in the evolution of plant defense. Evolution 51:1435-1444 McConn M, Creelman RA, Bell E, Mullet JE, Browse J (1997) Jasmonate is essential for insect defense in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 94:5473-5477 Mosleh Arany A, de Jong TJ, Meijden VD (2005) Herbivory and abiotic factors affect population dynamics of Arabidopsis thaliana in a sand-dune area. Plant Biology 7:549– 556 Moreau G, Quiring DT, Eveleigh ES, Bauce E (2003) Advantages of a mixed diet: feeding on several foliar age classes increases the performance of a specialist insect herbivore. Oecologia 135:391–399 30 Nieminen M, Suomi J, Van Nouhuys S, Sauri P, Riekkola M (2003) Effect of Iridoid Glycoside Content on Oviposition Host Plant Choice and Parasitism In A Specialist Herbivore. Journal of Chemical Ecology 29:823-844 Obeso JR (2002) The Costs of Reproduction in Plants. New Phytologist 155: 321-348 Ouwerkerk PBF, Hallard D, Verpoorte R, Memelink J (1999) Identification of UV-B lightresponsive regions in the promoter of the tryptophan decarboxylase gene from Catharanthus roseus. Plant Molecular Biology 41:491-503 Paige KN (1992) Overcompensation in Response to Mammalian Herbivory - from Mutualistic to Antagonistic Interactions. Ecology 73:2076-2085 Perez-Estrada LB, Cano-Santana Z, Oyama K (2000) Variation in leaf trichomes of Wigandia urens: environmental factors and physiological consequences. Tree Physiology 20:629632 Rautio P et al. (2002) Developmental plasticity in birch leaves: defoliation causes a shift from glandular to nonglandular trichomes. Oikos 98:437-446 Reed DW, Pivnick KA, Underhill EW (1989) Identification of chemical oviposition stimulants for the diamond back moth (plutella xylostella) present in three species of Brasssicaceae. Entomologia Experimentalis et Applicata 53:277-286 Reeves J, Lorch P, Kershner M (2009) Vision is Important for Plant Location by the Phytophagous Aquatic Specialist Euhrychiopsis lecontei Dietz (Coleoptera: Curculionidae). Journal of Insect Behavior 22:54-64 31 Roda A, Nyrop J, English-Loeb G (2003) Leaf pubescence mediates the abundance of non-prey food and the density of the predatory mite Typhlodromus pyri. Experimental and Applied Acarology 29:193-211 Saastamoinen M (2007) Life-history, genotypic, and environmental correlates of clutch size in the Glanville fritillary butterfly. Ecological Entomology 32:235-242 Sarret G et al. (2006) Trichomes of tobacco excrete zinc as zinc-substituted calcium carbonate and other zinc-containing compounds. Plant Physiology 141:1021-1034 Schilmiller AL, Last RL, Pichersky E (2008) Harnessing plant trichome biochemistry for the production of useful compounds. Plant Journal 54:702-711 Shonle I, Bergelson J (2000) Evolutionary ecology of the tropane alkaloids of Datura stramonium L. (Solanaceae). Evolution 54:778-788 Singer MC (2004) Measurement, correlates, and importance of oviposition preference in the life of checkerspot. Biology Oxford University Press, Oxford Skaltsa H, Verykokidou E, Harvala C, Karabourniotis G, Manetas Y (1994) Uv-B Protective Potential and Flavonoid Content of Leaf Hairs of Quercus-Ilex. Phytochemistry 37:987990 Strauss SY, Agrawal AA (1999) The ecology and evolution of plant tolerance to herbivory. Trends in Ecology and Evolution 14:179-185 Szendrei Z, Isaacs R (2005) Do plant cues influence the oviposition behavior of Japanese beetles? Entomologia Experimentalis Et Applicata 117:165-174 Thompson JN (1996) The coevolutionary process. University of Chicago Press, Chicago 32 Traw MB, Bergelson J (2003) Interactive effects of jasmonic acid, salicylic acid, and gibberellin on induction of trichomes in Arabidopsis. Plant Physiology 133:1367-1375 Turlings TCJ, Benrey B (1998) Effects of plant metabolites on the behavior and development of parasitic wasps. Ecoscience 5:321-333 Valverde PL, Fornoni J, Nunez-Farfan J (2001) Defensive role of leaf trichomes in resistance to herbivorous insects in Datura stramonium. Journal of Evolutionary Biology 14:424-432 Vergés A, Becerro MA, Alcoverro T, Romero J (2007) Variation in multiple characters of vegetative and reproductive seagrass tissues influences plant–herbivore interactions. Oecologia 151:675–686 Wahlberg N, Ehrlich PR, Boggs CL, Hanski I (2004) Bay checkerspot and Glanville fritillary compared with other species. Biology Oxford University Press, Oxford. Wang EM, Hall JT, Wagner GJ (2004) Transgenic Nicotiana tabacum L. with enhanced trichome exudate cembratrieneols has reduced aphid infestation in the field. Molecular Breeding 13:49-57 Wang GD et al. (2008) Terpene Biosynthesis in Glandular Trichomes of Hop. Plant Physiology 148:1254-1266 Wentzell AM, Kliebenstein DJ (2008) Genotype, age, tissue, and environment regulate the structural outcome of glucosinolate activation. Plant Physiology 147:415-428 Werker E (2000) Trichome diversity and development. In: Advances in Botanical Research Incorporating Advances in Plant Pathology 31: 1-35 33 Yencho, G. C. and Tingey, W. M. (1994), Glandular trichomes of Solanum berthaultii alter host preference of the Colorado potato beetle, Leptinotarsa decemlineata. Entomologia Experimentalis et Applicata 70: 217–225. 34 CHAPTER 2 CHANGES IN ULTRAVIOLET B RADIATION MODIFY RESISTANCE AND TOLERANCE IN ARABIDOPSIS THALIANA TO INSECT HERBIVORY. ABSTRACT Plants have evolved two major mechanisms in response to frequent attacks from insect herbivores. While resistance reduces the levels of damage suffered by plants, tolerance reduces the negative effects of herbivory on plant fitness. Despite their roles in defense and consequent fitness advantage, the evolution of these two characters may be constrained by the fitness costs associated with their production. In addition, these costs may be influenced by variation in environmental conditions. In a common garden experiment, the effects of ultraviolet B (UVB) radiation on plant resistance and tolerance to insect herbivory was evaluated by exposing recombinant inbred lines of Arabidopsis thaliana to natural levels of insect herbivory under two levels of UVB radiation (ambient and excluded UVB). Specifically, this study evaluated the existence of genetic variation in plant resistance and tolerance to insect herbivory under variable UVB radiation and whether changes in UVB radiation modify fitness costs associated with tolerance and resistance. Results from this study revealed the presence of genetic variation in plant resistance and tolerance and that this variation is dependent on UVB radiation levels. In addition, there were differences in the fitness cost of plant resistance and tolerance across UVB environmental gradients. No trade-off between resistance and tolerance however, was observed. These results highlight implications of changes in UVB radiation on future evolution of important plant characters involved in plant defense to insect herbivory. 35 INTRODUCTION Resistance and tolerance represent two major mechanisms by which plants may reduce the detrimental effects of insect herbivores on their fitness. While resistance directly influences the amount of damage inflicted on plants by their herbivores, tolerance reduces the fitness consequences of damage (Rausher 1992; Strauss and Agrawal 1999; Stowe et al. 2000). Resistance may be expressed in the form of production of structural barriers such as trichomes, and thorns, or chemical toxins like phenolics and glucosinolates (Levin 1973; McConn et al. 1997; Strauss and Agrawal, 1999; Rautio et al., 2002 Handley et al. 2005; Clauss et al. 2006; Kliebenstein 2004; Bidart-Bouzat et al. 2005). On the other hand, tolerance through compensatory growth enables plants to maintain fitness despite biomass loss resulting from herbivore damage. Compensatory responses may include the removal of apical dominance and proliferation of lateral branches (Paige 1992; Juenger and Bergelson 2000; Rautio et al. 2002; Freeman et al. 2003), extension of flowering period (Pilson and Decker 2002; Freeman et al. 2003), and shifts in resource allocation to different plant parts or plant architecture (GonzalezTeuber and Gianoli 2007). Although herbivores may impose directional selection on resistance and tolerance (Marquis 1992; Valverde et al. 2001), several studies have failed to find this trend in both field and laboratory studies (Mauricio and Rausher 1997; Weinig et al. 2003). One major reason for this observation is the potential presence of fitness costs, which may constrain the evolution of resistance and tolerance (Fineblum and Rausher 1995; Stowe 1998; Pilson 2000). Fitness costs may arise due to potential tradeoffs between defense and other functions, such as growth and reproduction (Herms and Mattson 1992; Belsky et al. 1993; Vandermeijden et al. 1988; Obeso 36 2002; Agrawal et al. 2004). Consequently, in the absence of herbivory, more resistant or tolerant plants would have lower fitness compared to the most susceptible ones. Yet, in the presence of herbivory, the benefit of resistance or tolerance may be able to offset their associated costs. Thus, the evolution of these characters will depend on the balance between their associated costs and benefits (Obeso 2002). Multiple environmental factors such as nutrient availability, competition, elevated CO2, and ozone, have been shown to modify the expression of tolerance and resistance to insect herbivore damage (Bolsinger et al. 1992; Bidart-Bouzat et al. 2005; Zvereva and Kozlov 2006; Caldwell et al. 2007; Veteli et al. 2007; Lau and Tiffin 2009). Even though considerable work has demonstrated how environmental factors may influence costs of resistance (Bergelson and Purrington 1996; Prittinen et al. 2003), only few studies have evaluated environmental effects on the costs of tolerance (Agrawal et al. 2004; Stevens et al. 2007). In addition, both empirical and theoretical studies have not shown a consistent pattern on how the environment may modify these costs (Bergelson and Purrington 1996). For example, while some authors have predicted a higher cost of resistance or tolerance under more stressful conditions (Mutikainen et al. 2002), others have predicted a reverse trend (Herms and Mattson 1992). An understanding of how environmental factors may influence genetic variation in tolerance and resistance, as well as their associated costs, may help in predicting evolutionary trajectories of these characters in future changing environment (Tiffin 2000a). Ultraviolet B (UVB) radiation represents a small fraction of the solar spectrum (290-315 nm). The level of UVB radiation reaching the surface of the Earth has been recently changing due to the depletion of the trophospheric ozone layer, pollution, and cloud cover 37 (Intergovernmental Panel on Climate Change, IPCC 2007). Several studies have evaluated the effect of UVB radiation on plant growth and physiological processes (Hader et al. 2000; Searles et al. 2001), ranging from inhibition of photosynthetic activities to modification of flowering patterns, biomass accumulation, and reproduction (Searles et al. 2001; Tegelberg et al. 2004). With respect to plant chemistry, UVB radiation can increase accumulation of UVB-absorbing compounds such as flavonoids and phenolics, which may in turn influence plant’s interactions with higher trophic levels (Agrell et al. 2000; Searles et al. 2001; Hofmann et al. 2003; Izaguirre et al. 2003; Stratmann 2003; Bassman 2004; Izaguirre et al. 2007). For example, Izaguirre et al. (2007) reported that the same phenolics compounds commonly induced in response to insect herbivory were found to be induced in Nicotiana attenuata and Nicotiana longifolia when exposed to enhanced UVB radiation. Similarly, Stratmann (2003) detected an increase in phenylpropanoids in plant plants exposed to ambient UVB radiation when compared with plants grown under excluded UVB (using light filters) suggesting that UVB radiation may influence plant-insect interactions. Since variation in UVB levels has been shown to affect different plant characters related to defense, growth and reproduction (Veteli et al. 2003; Bassman 2004; Rousseaux et al. 2004), it is expected that these changes could have an effect on plant resistance and/or tolerance to insect herbivory. Previous studies have focused on the effect of changes in UVB radiation levels on the production of plant secondary chemicals, such as phenolics, which have a role in resistance against insect herbivores (Agrell et al. 2000; Searles et al. 2001; Hofmann et al. 2003; Izaguirre et al. 2003; Stratmann 2003; Bassman 2004; Izaguirre et al. 2007). No information appears to exist however, on the role of UVB on structural defenses, such as trichomes or 38 characters related to plant tolerance against insect herbivory. With respect to the effects of variable environments on resistance and tolerance, previous studies have shown variable responses of characters related to resistance and tolerance (Ziska and Teramura 1992; Dube and Bornman 1992; Tevini 1993; McDonald et al. 1999; Agrell et al. 2000; Lavola et al. 2000; Hofmann et al. 2003). However, there is a lack of information regarding effects of the UVB environment on potential fitness costs related to plant resistance and tolerance. This study presents data from a common garden experiment in which a population of recombinant inbred lines (RIL) of Arabidopsis thaliana was exposed to two different levels of UVB radiation (i.e., ambient- and excluded-UVB) and insect herbivory (i.e., sprayed plants and plants exposed to natural levels of insect herbivory). The main goal of this study was to determine whether there was genetic variation in resistance and tolerance to insect herbivory in this A. thaliana RIL population under variable UVB radiation levels, and to assess potential fitness costs associated with plant resistance and tolerance. An understanding of the effects of UVB on plants as well the fitness costs associated with plant response to this environmental factor in terms of resistance and tolerance is important because they may constrain the future evolution of these characters, which in turn may alter the dynamics of plant-insect interactions. For example, if costs increase with enhanced UVB levels, plant resistance and tolerance may be more commonly expressed as inducible strategies in order to minimize their costs (Simms and Rausher 1987, Herms and Mattson 1992; Bergelson and Purrington 1996; Cipollini 1998; Purrington 2000). In addition, the expression of fitness costs across different UVB environments may provide information regarding whether either plant tolerance or resistance are favored over the other, or whether these strategies are likely to be maintained at intermediate levels. 39 MATERIALS AND METHODS Study Plant Arabidopsis thaliana (L.) Heynh (Brassicaceae), also known as the mouse ear cress, is an annual herbaceous plant native to Europe and Asia with a wide geographic distribution across North America and Africa. This plant species is naturally found in open or disturbed habitats, sandy soils, riverbanks, roadsides, rocky slopes, and meadows (Baskins and Baskins 1983). Arabidopsis thaliana has a short life cycle lasting between 6-8 weeks from germination to senescence and it is mostly a self-fertilizing species with less than 1% outcrossing rates (Hoffmann et al. 2003). Because of the high selfing rates, natural populations seem to consist majorly of a single genotype. However, a significant level of genetic variation exists among different ecotypes in both reproductive and life history characters (Bergelson et al. 1998; Mauricio 1998; Nordborg and Bergelson 1999). In A. thaliana, two types of resistance to insect herbivory have been reported. These include structural defenses, such as trichomes, and chemical defenses such as glucosinolates, which are known to vary across populations (Mauricio and Rausher 1997; Valverde et al. 2001; Bidart-Bouzat et al. 2005; Clauss et al. 2006; Loe et al. 2007). In this study, 100 recombinant inbred lines (RILs) derived from a cross between the A. thaliana ecotypes Columbia (Col-0) and Landsberg (Ler), and advanced to the eighth generation through a single seed descent were used (Lister and Dean 1993). A RIL population was used in order to assess intraspecific genetic variation in plant resistance and tolerance to UVB radiation. Experimental Design A common garden experiment was performed to assess the effects of UVB radiation and plant genotype on tolerance and resistance in A. thaliana. This experiment included two UVB 40 treatment levels (total of 20 plots), which consisted of ambient (UVB+) and excluded (UVB-) ultraviolet B radiation (Figure 1). Ultraviolet B radiation was excluded from the randomly assigned UVB- plots using Mylar polyester film. This film transmits radiations only in the wavelength greater than 315 nm, thereby blocking UVB radiation but transmitting radiations in UVA regions. For the UVB+ treatment, plots were covered with clear acetate film that allowed most UVB radiation to pass through (see Figure 2). Half of the plots assigned to each UVB level were exposed to natural levels of insect herbivory and the other half were excluded from insect herbivores by spraying the plots with a pesticide (Proclaim by Syngenta Inc. USA) at ten days interval. Plants excluded from insect herbivory were inspected on a daily basis to ensure there was no herbivore damage. Each plot was surrounded with a wire-mesh fence to exclude mammalian herbivory. One hundred recombinant inbred lines (RILs) of A. thaliana were used in this study. Seeds were initially cold-treated for 5 days at 4°C to break dormancy, and then, transferred to growth chambers for germination. Within the chambers, plants were exposed to 10 hours of daylight and 14 hours of night, with day and night temperatures of 20°C and 16°C, respectively. Plants were rotated every other day within growth chambers and every week among chambers to avoid chamber and position effects. After five weeks of growth in the chambers, plants were transplanted to the field (Ecological Research Station, Bowling Green State University, Bowling Green, Ohio). Two sets of 100 RILs were randomly assigned to each of the 20 plots (total of 4000 plants). 41 Estimation of Resistance and Tolerance To evaluate the effects of UVB radiation on plant resistance and tolerance, two resistance measures (defined as direct and indirect resistance) and one measure of tolerance were estimated for each RIL. The direct measure of resistance was estimated as trichome density. Trichome density was measured using digital images of the youngest fully extended rosette leaf per plant at bolting time. Leaf area of the selected rosette leaves was estimated using image J software (www.rsbweb.nih.gov/ij). Trichome density was calculated as the total number of trichomes (on the upper side of the leaf) divided by the leaf area. The indirect measure of resistance was estimated by subtracting a proportion of the rosette area damaged from one (i.e., 1- damage). This measurement assumes that the level of herbivore damage sustained is a reflection of the level of resistance (i.e., a plant with a high resistance level will have low levels of herbivore damage and vice versa). This indirect estimate of resistance is important because it accounts for all types of resistance mechanisms present in this plant species (both structural and chemical). Rosette leaf damage (i.e., percent area removed by herbivores on the rosette leaves) was used as a measure of insect herbivore damage, which was estimated as a qualitative leaf damage index (Bidart-Bouzat et al. 2005). This index was calculated by assigning leaves to three size categories based on leaf length (i.e., small: 0- 3cm; medium: 3 - 6cm; and large: 6-9cm), and three damage categories based on the amount of rosette leaf tissue removed (i.e., < 25%, 25-75 %, and > 75%). To estimate plant tolerance, plant fitness was measured as total fruit number, which is an estimate of lifetime fitness in A. thaliana, and further compared among damaged and undamaged plants. Relative fitness was estimated as absolute fitness divided by the mean absolute fitness of 42 the population of RILs per UVB and herbivory treatment level. Tolerance was estimated by comparing the relative fitness of the damaged plants (WD) to that of the undamaged plants (WU). This may be estimated as the ratio of the relative fitness of the undamaged to the damaged plants (WD/WU) or as the difference (WD-WU) between the relative fitness of damaged and the undamaged plants (Strauss and Agrawal 1999). The choice between these two approaches depends on how the damage data were collected. For example, when damage data are categorical (i.e., presence or absence of damage), an approach using the difference in mean relative fitness would be more appropriate; however, when damage data are continuous, the use of the ratio would be more indicated. This is because tolerance estimated as a ratio is conceptually similar to using a regression coefficient of fitness on damage. This is equivalent to defining tolerance as a reaction norm along a gradient of variable levels of damage, such that the slope describes to what extent plant fitness has been affected by damage (Mauricio and Rausher 1997; Simms and Triplett 1994; Tiffin and Rausher 1999; Strauss and Agrawal 1999). In this study, rosette damage was estimated using a damage index, which vary in a continuous fashion. Therefore, tolerance was estimated as the ratio of the relative fitness of the undamaged to the damaged plants (WD/WU). Two approaches were used to evaluate costs associated to resistance and tolerance, as well as the effect of UVB radiation on these costs. First, fitness costs were detected through Pearson’s correlation coefficients as the presence of a negative association between relative fitness of the undamaged plants and either resistance or tolerance under each level of UVB radiation. Second, the existence of a potential trade-off between plant resistance and tolerance was also assessed using correlation analysis. 43 Statistical Analysis To evaluate the effects of UVB radiation and RIL on indirect resistance, and those of UVB, herbivory and RIL on trichome density, analyses of variance were performed using the PROC GLM procedure (SAS statistical software, version 9.1). Data was either log- or rankedtransformed to correct for normality and heteroscedasticity when necessary. For this purpose, a mixed linear model was used for trichome density, considering UVB radiation and herbivory as fixed effects, and block, RILs and its interactions as random effects. For indirect resistance, herbivory was not included in the model since this measure was estimated only from plants exposed to herbivory. In addition, higher-order block interactions were first included in the model but then excluded because they were not significant. Regarding tolerance, since it was not possible for a plant to be damaged and undamaged at the same time, tolerance was estimated as means for each RIL. In order to show the presence of genetic variation for tolerance, an analysis of covariance was performed including relative fitness as the dependent variable and rosette damage as the covariate using the following model: Relative fitness = Block + RIL + Rosette damage + RIL × Rosette damage + Error. Significant rosette damage × RIL interaction indicates the presence of genetic variation in tolerance. Cost of resistance was evaluated using Pearson’s correlation coefficients as estimates of potential genetic associations between the RIL means of relative fitness of undamaged plants and the two measures of resistance; that is, trichome density and indirect resistance (1-damage). Relative fitness of the undamaged plants was used based on the premise that in the absence of herbivory, the benefit of being resistant is removed and only the cost will be manifested. Thus, a significant negative genetic correlation coefficient between relative fitness and resistance in the 44 absence of damage implies the presence of a fitness cost. In the case of tolerance, however, a direct genotypic correlation between the fitness of the undamaged plant and tolerance would be biased (Tiffin and Rausher 1999). This is because using the same data to estimate relative fitness of the undamaged plants and tolerance would introduce an artifactual covariance (Tiffin and Rausher 1999). This artifactual covariance was calculated and subtracted from the estimated covariance between tolerance and relative fitness of undamaged plants. A bootstrap technique was used to generate a 95% confidence limit in order to determine if the negative covariances have confidence limits that do not include zero using a modified version of the SAS macro as described in Stinchcombe (2005). This method compiles the bootstrap distribution of each corrected covariance, and the 2.5 and 97.5 percentile of this distribution formed the 95% bootstrap percentile confidence interval. Thus, if the confidence interval failed to include zero, then the p-value is deemed to be less than or equal to 0.05, and the effect is said to be significant. RESULTS Analyses of variance revealed significant UVB effects on both direct (trichome density) and indirect resistance measures as well as genetic variation in these characters’ responses to changes in UVB variation (see Figure 3A, 3B, 3C and Table 1). Overall, exclusion of UVB and herbivores resulted in a decrease in trichome density, although this effect was more pronounced in the sprayed treatment (herbivory excluded) as indicated by a significant herbivory by UVB interaction. In addition, plant responses to UVB radiation in terms of trichome density were found to be dependent on both RIL and herbivory (i.e., significant UVB × RIL interaction) (Table 1). Effects of herbivory on trichome density were found to be marginally dependent on 45 RIL and the three-way interaction (UVB × Herbivory × RIL) was statistically not significant (Table 1). This non-significant three-way interaction indicates that UVB radiation affected trichomes density in response to herbivory in the same way across the genotypes (Table1). With respect to tolerance to insect herbivory, a significant genetic variation in tolerance was revealed by a significant RIL × rosette damage interaction in an analysis of covariance using relative fitness as the dependent variable (Figure 3D). In order to evaluate fitness costs of resistance (both direct and indirect resistance measures) and tolerance, as well as whether variation in UVB radiation influenced these costs, genotypic correlations between relative fitness of the undamaged plants and resistance or tolerance were performed. Overall, results suggest that changes in UVB radiation influence fitness costs associated with both direct and indirect resistance as well as tolerance to insect herbivory. In terms of direct resistance (trichome density), results from genotypic correlations did not reveal any significant trade-offs between relative fitness and trichome density under ambient UVB radiation levels (r=-0.06, P=0.596); (Figure 4A). However, a marginally significant cost was observed in plants excluded from UVB radiation (r=-0.17, P=0.0875); (Figure 4B). In terms of resistance, a negative genotypic correlation between relative fitness of the undamaged plants and indirect resistance was detected under ambient UVB, which suggests the presence of fitness cost in this environment (r=-0.31, P=0.0015); (Figure 4C). Conversely, under the excluded UVB regime, although an inverse trend was observed, the correlation was not significant (r=-0.13, P=0.1889); (Figure 4D). In terms of tolerance, results revealed a significant negative genotypic correlation between relative fitness of the undamaged plants and tolerance under both ambient and excluded 46 UVB radiation levels (r=-0.68, P<.0001; and r=-0.36, P=0.0003; respectively); (Figures 4E and 4F). However, after correcting for the artifactual covariance, the corrected covariance was negative under ambient UVB but not under excluded UVB (Table 3). This result suggests that cost of tolerance appears to be present only under ambient UVB. The significant negative genetic correlation observed under excluded UVB may have been due to the presence of an artifactual covariance as result of using the same data to estimate relative fitness of the undamaged plants and tolerance. Despite the presence of fitness costs for both indirect resistance and tolerance, evaluation of the potential presence of trade-offs between tolerance and trichome density did not reveal any form of trade-offs under ambient UVB (r=0.13518, P=0.1799) or excluded UVB (r=0.014, P= 0.888); (Figures 5A and B). Similarly, evaluation of trade-offs between tolerance and indirect resistance did not reveal any trade-offs under ambient (r=0.035, P=0.7281) or excluded UVB (r=-0.051, P=0.6176); (Figures 5C and D). DISCUSSION This study demonstrates that variation in UVB radiation influences resistance and tolerance to insect herbivory in A. thaliana. Although previous studies have demonstrated herbivore-induced variation in resistance characters such as trichome density and plant damage levels (Mauricio and Rausher 1997; Weinig et al. 2003; Clauss et al. 2006), no studies to date have shown whether variation in UVB radiation levels may influence intraspecific genetic variation in resistance and tolerance to insect herbivory. This study revealed significant genetic variation in the response of both direct resistance (trichome density) and indirect resistance characters to variation in UVB levels. This significant RIL× UVB interaction is very important 47 because it indicates a potential for these characters to evolve under future changes in this environmental factor. Another interesting finding from this study includes the significant interaction between UVB and herbivory for trichome density, suggesting that future changes in UVB radiation may influence plant structural defense responses to insect herbivory, which in terms may affect coevolutionary plant-insect associations. With respect to potential fitness costs associated with resistance, this study failed to find costs associated with trichome production in both ambient and excluded UVB environments (although a marginally significant result was found in excluded UVB plots). Given the detrimental effects of UVB radiation on plant fitness and related characters, the inability to detect a cost for the production of trichomes is not unexpected. This is because previous studies have linked trichome production with plant protection from harmful effects of ultraviolet radiation and extreme temperatures (Skaltsa et al. 1994; Werker 2000; Agrawal et al. 2004). Thus, trichomes may be serving a dual role in plants by protecting them from both herbivores and harmful UVB radiation, thereby reducing the negative fitness consequence of these factors on plants. In terms of indirect resistance, however, a significant cost was detected in the presence of ambient UVB but not under excluded UVB, indicating that UVB radiation modified the cost of indirect resistance in this system. This finding is concordant with previous studies that have demonstrated that environmental factors may modulate cost of resistance (Bergelson and Purrington 1996; Koricheva 2002; Siemens 2003; Osier and Lindroth 2006). For example, Osier and Lindroth (2006) demonstrated that nutrient availability modified cost of resistance in Populus tremuloides. Similarly, Siemens (2003) found the cost of resistance to differ in Arabis perennans that grew alone and Arabis perennans that grew under competition with Bouteloua 48 gracilis. It is also important to note that in this study, the measure of indirect resistance was related to the level of herbivore damage, and thus, it includes potential costs related to both structural and chemical defenses. Since costs related to structural defenses were absent in this study, the presence of this indirect resistance cost may have arisen from the constitutive chemical resistance mechanisms present in A. thaliana (Marak et al. 2003; Cipollini et al. 2003; Sampedro et al. 2011; Stowe and Marquis 2011). For example, Stowe and Marquis (2011) provided empirical evidence for the presence of constitutive costs of foliar glucosinolates in Brassica rapa by artificially selecting for foliar glucosinolates. Their results revealed that lines selected for high foliar glucosinolates showed decrease in number of flowers and mean seeds per fruit. It is also possible that the presence of fitness costs may be associated with the process of induction of chemical resistance (see Cipollini et al. 2003) With respect to fitness costs associated with tolerance, this study revealed a significant negative genetic correlation between relative fitness of undamaged plants and tolerance under both levels of UVB radiation. After correcting for an artifactual covariance however, which likely resulted from using the same data to estimate relative fitness of the undamaged plants and tolerance, the corrected covariance was only negative in plots exposed to ambient UVB radiation. This result suggests that there is evidence that costs of tolerance may exist under ambient UVB radiation levels. Little information exists on the effects of abiotic environmental changes on the cost of tolerance to insect herbivory and different results have been obtained in terms of how cost of tolerance may vary across environmental gradients. For example, while Stinchcombe (2002) detected a cost related to tolerance to be present under low soil nutrients but not under high nutrient level in the ivy leaf morning glory exposed to deer herbivory, Steven et 49 al. (2007) failed to find costs of tolerance to differ between low and high nutrient levels in aspen plants exposed to artificial defoliation. In addition, Siemens et al. (2003) found that costs of tolerance to insect herbivory in Arabis perennans were present in a non-competitive environment but absent in a competitive environment. The inconsistent patterns in the expression of fitness costs associated with tolerance in these studies suggest that these costs may be dependent on multiple factors including the plant species and the specific selective factor/s acting in each type of environment. Since theoretical models predict that tolerance and resistance serve the same function, it can be expected that plants should be either resistant or tolerant but not both (Fineblum and Rausher 1995; Tiffin 2000b; Fornoni et al. 2004). Yet, some models have predicted that both resistance and tolerance may be maintained at intermediate levels under certain conditions. These models, as summarized by Fornoni et al. (2004) and Nunez-Farfan et al. (2007), include the cost of tolerance and resistance across environmental gradients, fluctuations in both time and space with respect to insect herbivory, and characters that are under the selective pressures of factors other than herbivory. In addition, the type of damage on plants, such as rosette damage or apical damage, and type of herbivores present may also be exerting different pressures on tolerance and resistance. In this study, trade-offs between resistance and tolerance to insect herbivory were not detected. This result was in agreement with previous studies that have assessed the presence of trade-offs between resistance and tolerance. For example, Tiffin and Rausher (1999) did not detect any trade-offs between resistance and tolerance to foliovory and apical meristem damage in the common morning glory. Likewise, Weinig et al. (2003) did not find any significant trade-offs between tolerance and resistance to apical damage in A. thaliana, 50 suggesting the possibility of these two characters evolving independently of each other. Most of these studies, however, did not include potential changes in these trade-offs across environments (but see Stevens et al. 2007). The inability to detect tradeoffs between resistance and tolerance in this study could be explained by two factors. First, fitness costs for resistance and tolerance varied across environments. For example, fitness costs of tolerance and indirect resistance were only observed in plots exposed to ambient UVB but not in plots excluded from UVB. This type of variation in fitness costs across variable environments may lead to the maintenance of genetic variation in resistance and tolerance, and thus, to selection for intermediate levels of these characters across populations (Fornoni et al. 2004). Second, it is possible that defense traits used in this study may be under selection by selective agents other than insect herbivory (Stowe et al. 2000; Fornoni et al. 2003; Loe et al. 2007; Nunez-Farfan et al. 2007). For example, enhanced UVB radiation in the absence of herbivory led to an increase in trichome density. This result therefore suggests that trichomes may be playing a protective role against UVB radiation (Skaltsa et al. 1994; Werker 2000; Agrawal et al. 2004) and may be under selection in the presence of UVB radiation. Nevertheless, a selection experiment will be required to confirm whether UVB radiation is indeed imposing direct selection on trichome density. In conclusion, results from a common garden experiment revealed that changes in UVB radiation influence plant characters related to resistance and tolerance to insect herbivory, and that this environmental effect is dependent on plant genotype (RIL). This study also showed the existence of fitness costs associated with plant resistance and tolerance, which were dependent on the UVB environment; that is, they were present only under ambient UVB radiation. 51 Although previous studies have not shown a consistent pattern on the expression of fitness cost associated with resistance and tolerance (i.e. whether costs are more pronounced in the more stressful environments or less stressful environments), results from this study support the view that cost of tolerance and resistance may be more pronounced under a more stressful environment (Mutikainen et al. 2002; Sampedro et al 2011). This is because growth and other physiological processes may be compromised if resources decline or conditions become more stressful. In addition, plant responses to UVB may also come at an additional fitness cost, which becomes more pronounced in the presence of insect herbivory. The presence of these fitness costs under enhanced UVB radiation may have important evolutionary implications because selection for higher levels of resistance and tolerance to insect herbivory under enhanced UVB may result in lower fitness, and thus, may constrain the evolution of these characters. 52 Table 1. ANOVA F-values showing the effects of herbivory, UVB radiation, RILs and their interactions on trichome density (direct resistance) and an indirect resistance estimate (1herbivore damage). Source of variation Block Herbivory UVB Herbivory x UVB RIL Herbivory x RIL UVB x RIL Herbivory x UVB x RIL Trichome Indirect Density Resistance 38.59*** 1.14 220.48*** 73.79*** 92.94*** 75.74*** 0.85 1.27 1.35† 2.32*** 1.39* 1.00 - Significance levels: † P<0.1, * P<0.05, ** P<0.01, *** P<0.001 Tolerance 0.55 0.86 0.90 - 53 Table 2. ANCOVA F-values showing the relationship between relative fitness and rosette damage across UVB environments and in each of the UVB environments. A significant association in rosette damage × RIL indicates the presence of genetic variation in tolerance. Source of variation Block RIL Rosette Damage Rosette Damage x RIL Across UVB environments 17.50*** 1.27* 3.29† 1.28* UVB+ UVB- 25.84*** 1.26† 7.24** 1.07 7.09*** 1.33* 1.50 1.26† Significance levels: † P<0.1, * P<0.05, ** P<0.01, *** P<0.001 54 Table 3. Pearson’s correlation coefficients showing the relationship between relative fitness of undamaged plants and tolerance under ambient (UVB+) and excluded (UVB-) ultraviolet B radiation level. Corrected covariance represents the covariance adjusted for the artifactual covariance. Correlation coefficient Covariance Artifactual covariance Corrected covariance 95% Confidence interval UVB+ -0.68* -0.0474 -0.0317 -0.0157 (-0.036,0.003) UVB-0.36* -0.0151 -0.0229 0.0079 (-0.004,0.019) Significance levels for Pearson coefficient of correlation: * P<0.001 55 Figure 1A. Picture of the experimental plots showing plant arrangements within each plot. UVB+ UVB- UVB+ UVB- UVB+ UVB- UVB+ UVB- UVB+ UVBHERBIVORY SPRAYED UVB+ UVB- UVB+ UVB- UVB+ UVB- UVB+ UVB- UVB+ UVB- Figure 1B. Schematic representation of experimental plot layout showing two levels of herbivory (herbivory and sprayed) and UVB (ambient and excluded) treatments distributed among twenty plots. 56 Mylar Clear acetate Absorbance (AU) 3 2 1 0 200 250 300 350 400 Wavelength (nm) Figure 2. Absorption spectra for Mylar and Clear acetate filters. 450 0.94 600 Resistance 0.92 400 * 0.90 0.88 0.86 200 0.84 (B) UVB+ UVB- (D) 1.8 600 UVB+ UVB- UVB+ UVB- Tolerance 1.6 400 ** Trichome Density (Sprayed) (C) (A) ** Trichome Density (Herbivory) 57 200 1.4 1.2 1.0 0.8 0.6 UVB+ UVB- Figure 3. Effects of the presence (UVB+) and exclusion (UVB-) of ultraviolet B radiation on trichome density in the presence of herbivory (A), trichome density in the absence of herbivory (B), indirect resistance (C), and tolerance (D) of an A. thaliana recombinant inbred line population. Means of RILs are represented with closed circles and asterisks on lines represent significant differences in phenotypic means between treatments. * P< 0.05, **P<0.001 2.0 (A) UVB+ Relative Fitness (Undamaged) Relative Fitness (Undamaged) 58 r = -0.06 1.8 1.6 1.4 1.2 1.0 0.8 0.6 200 300 400 500 (B) UVBr = -0.17 1.4 1.2 1.0 0.8 600 200 (C) UVB+ Relative Fitness (Undamaged) Relative Fitness (Undamaged) 2.0 r = -0.31** 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.88 0.90 0.92 UVBr = -0.13 1.4 1.2 1.0 0.8 0.88 0.94 0.90 Relative Fitness (Undamaged) Relative Fitness (Undamaged) UVB+ r = -0.68*** 1.8 1.6 1.4 1.2 1.0 0.8 0.6 0.8 1.0 1.2 Tolerance 0.92 Resistance (E) 0.6 600 (D) Resistance 2.0 400 Trichome Density Trichome Density 1.4 1.6 (F) UVBr = -0.36** 1.4 1.2 1.0 0.8 0.6 0.8 1.0 1.2 1.4 1.6 1.8 Tolerance Figure 4. Genotypic correlations showing fitness costs of trichomes (A and B), resistance (C and D) and tolerance (E and F) in A. thaliana plants grown under ambient and excluded UVB radiation. Each point represents genotypic means of RILs. ** P<0.01, *** P<0.001 59 UVB+ (A) 1.6 UVB+ (C) r = 0.04 P= 0.73 r = 0.13 P= 0.18 Tolerance 1.4 1.2 1.0 0.8 0.6 1.8 UVB- (B) r = 0.01 P= 0.88 1.6 Tolerance UVB- (D) r =-0.05 P= 0.62 1.4 1.2 1.0 0.8 0.6 200 400 Trichome Density 600 0.84 0.86 0.88 0.90 0.92 0.94 Resistance Figure 5. Genotypic correlations between resistance and tolerance in A. thaliana plants grown under ambient and excluded UVB radiation. Each point represents genotypic means of RILs. 60 REFERENCES Agrawal AA, Conner JK, Stinchcombe JR (2004) Evolution of plant resistance and tolerance to frost damage. Ecology Letters 7:1199-1208 Agrell J, McDonald EP, Lindroth RL (2000) Effects of CO2 and light on tree phytochemistry and insect performance. Oikos 88:259-272 Agren J, Schemske DW (1993) The Cost of Defense against Herbivores - an Experimental-Study of Trichome Production in Brassica rapa. American Naturalist 141:338-350 Baskins JM, Baskins CC (1983) Seasonal changes in the germination responses of buried seeds of Arabidopsis thaliana and ecological interpretation. Botany Gazette 144:540-54 Bassman JH (2004) Ecosystem consequences of enhanced solar ultraviolet radiation: Secondary plant metabolites as mediators of multiple trophic interactions in terrestrial plant communities. Photochemistry and Photobiology 79:382-398 Belsky AJ, Carson WP, Jensen CL, Fox GA (1993) Overcompensation by Plants - Herbivore Optimization or Red Herring. Evolutionary Ecology 7:109-121 Bergelson J, Purrington CB (1996) Surveying patterns in the cost of resistance in plants. American Naturalist 148:536-558 Bergelson J, Stahl E, Dudek S, Kreitman M (1998) Genetic variation within and among populations of Arabidopsis thaliana. Genetics 148:1311-1323 Bidart-Bouzat MG, Mithen R, Berenbaum MR (2005) Elevated CO2 influences herbivoryinduced defense responses of Arabidopsis thaliana. Oecologia (Berlin) 145:415-424 Bolsinger M, Lier ME, Hughes PR (1992) Influence of ozone air pollution on plant-herbivore interactions. Part 2: Effects of ozone on feeding preference, growth and consumption rates of monarch butterflies (Danaus plexippus). Environmental Pollution 77:31-37 61 Caldwell MM, Bornman JF, Ballare CL, Flint SD, Kulandaivelu G (2007) Terrestrial ecosystems, increased solar ultraviolet radiation, and interactions with bother climate change factors. Photochemical & Photobiological Sciences 6:252-266 Cipollini D (1998) Induced defenses and phenotypic plasticity. Trends in Ecology & Evolution 13:200-200 Cipollini D, Purrington CB, Bergelson J (2003) Costs of induced responses in plants. Basic and Applied Ecology 4:79-89 Clauss MJ, Dietel S, Schubert G, Mitchell-Olds T (2006) Glucosinolate and trichome defenses in a natural Arabidopsis lyrata population. Journal of Chemical Ecology 32:2351-2373 Dube SL, Bornman JF (1992) Response of Spruce Seedlings to Simultaneous Exposure to Ultraviolet-B Radiation and Cadmium. Plant Physiology and Biochemistry 30:761-767 Fineblum WL, Rausher MD (1995) Tradeoff between resistance and tolerance to herbivore damage in a morning glory. Nature (London) 377:517-520 Fornoni J, Valverde PL, Nunez-Farfan J (2003) Quantitative genetics of plant tolerance and resistance against natural enemies of two natural populations of Datura stramonium. Evolutionary Ecology Research 5:1049-1065 Fornoni J, Nunez-Farfan J, Valverde PL, Rausher MD (2004) Evolution of mixed strategies of plant defense allocation against natural enemies. Evolution 58:1685-1695 Freeman RS, Brody AK, Neefus CD (2003) Flowering phenology and compensation for herbivory in Ipomopsis aggregata. Oecologia (Berlin) 136:394-401 Gillespie JH, Turelli M (1989) Genotype-Environment Interactions and the Maintenance of Polygenic Variation. Genetics 121:129-138 62 Gonzalez-Teuber M, Gianoli E (2007) Tolerance to simulated herbivory in two populations of Convolvulus chilensis (Convolvulaceae). Acta Oecologica 32:119-123 Hader DP, Porst M, Lebert M (2000) On site photosynthetic performance of Atlantic green algae. Journal of Photochemistry and Photobiology 57:159-168 Handley R, Ekbom B, Agren J (2005) Variation in trichome density and resistance against a specialist insect herbivore in natural populations of Arabidopsis thaliana. Ecological Entomology 30:284-292 Herms DA, Mattson WJ (1992) The Dilemma of Plants - to Grow or Defend. Quarterly Review of Biology 67:283-335 Hoffmann MH, Bremer M, Schneider K, Burger F, Stolle E, Moritz G (2003) Flower visitors in a natural population of Arabidopsis thaliana. Plant Biology 5:491-494 Hofmann RW et al. (2003) Responses to UV-B radiation in Trifolium repens L. - physiological links to plant productivity and water availability. Plant Cell and Environment 26:603612 IPCC (2007). Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom. Izaguirre MM, Mazza CA, Svatos A, Baldwin IT, Ballare CL (2007) Solar ultraviolet-B radiation and insect herbivory trigger partially overlapping phenolic responses in Nicotiana attenuata and Nicotiana longiflora. Annals of Botany 99:103-109 Izaguirre MM, Scopel AL, Baldwin IT, Ballare CL (2003) Convergent responses to stress. Solar ultraviolet-B radiation and Manduca sexta herbivory elicit overlapping transcriptional responses in field-grown plants of Nicotiana longiflora. Plant Physiology 132:1755-1767 63 Juenger T, Bergelson J (2000) The evolution of compensation to herbivory in scarlet gilia, Ipomopsis aggregata: Herbivore-imposed natural selection and the quantitative genetics of tolerance. Evolution 54:764-777 Koricheva J (2002) Meta-analysis of sources of variation in fitness costs of plant anti-herbivore defenses. Ecology 83:176–190 Kliebenstein DJ (2004) Secondary metabolites and plant/environment interactions: a view through Arabidopsis thaliana tinged glasses. Plant Cell and Environment 27:675-684 Levin DA (1973) The role of trichomes in plant defense. The Quarterly Review of Biology 48:3-15 Lau JA, Tiffin P (2009) Elevated carbon dioxide concentrations indirectly affect plant fitness by altering plant tolerance to herbivory. Oecologia 161:401-410 Lavola A, Julkunen-Tiitto R, de la Rosa TM, Lehto T, Aphalo PJ (2000) Allocation of carbon to growth and secondary metabolites in birch seedlings under UVB radiation and CO2 exposure. Physiologia Plantarum 109:260-267 Lister C, Dean C (1993) Recombinant Inbred Lines for Mapping Rflp and Phenotypic Markers in Arabidopsis thaliana. Plant Journal 4:745-750 Loe G, Torang P, Gaudeul M, Agren J (2007) Trichome production and spatiotemporal variation in herbivory in the perennial herb Arabidopsis lyrata. Oikos 116:134-142 McDonald EP, Agrell J, Lindroth RL (1999) CO2 and light effects on deciduous trees: growth, foliar chemistry, and insect performance. Oecologia 119:389-399 Marak HB, Biere A, Van Damme JMM (2003) Fitness costs of chemical defense in Plantago lanceolata L.: Effects of nutrient and competition stress. Evolution 57:2519-2530 64 Marquis RJ (1992) The selective impact of herbivory. In: Fritz RS, Simms EL (eds) Plant Resistance to Herbivory and Pathogens. Ecology, Evolution and Genetics. The University of Chicago Press, Chicago, pp 301-325 Mauricio R (1998) Costs of resistance to natural enemies in field populations of the annual plant Arabidopsis thaliana. American Naturalist 151:20-28 Mauricio R, Rausher MD (1997) Experimental manipulation of putative selective agents provides evidence for the role of natural enemies in the evolution of plant defense. Evolution 51:1435-1444 McConn M, Creelman RA, Bell E, Mullet JE, Browse J (1997) Jasmonate is essential for insect defense in Arabidopsis. Proceedings of the National Academy of Sciences of the United States of America 94:5473-5477 Mutikainen P, Walls M, Ovaska J, Keinanen M, Julkunen-Tiitto R, Vapaavuori E (2002) Costs of herbivore resistance in clonal saplings of Betula pendula. Oecologia 133:364-371 Nordborg M, Bergelson J (1999) The effect of seed and rosette cold treatment on germination and flowering time in some Arabidopsis thaliana (Brassicaceae) ecotypes. American Journal of Botany 86:470-475 Nunez-Farfan J, Fornoni J, Valverde PL (2007) The evolution of resistance and tolerance to herbivores. In: Annual Review of Ecology Evolution and Systematics 38: 541-566 Obeso JR (2002) The costs of reproduction in plants. New Phytologist 155:321-348 Osier TL, Lindroth RL (2006) Genotype and environment determine allocation to and costs of resistance in quaking aspen. Oecologia 148:293–303 Paige KN (1992) Overcompensation in Response to Mammalian Herbivory - from Mutualistic to Antagonistic Interactions. Ecology 73:2076-2085 65 Pilson D (2000) The evolution of plant response to herbivory: simultaneously considering resistance and tolerance in Brassica rapa. Evolutionary Ecology 14:457-489 Pilson D, Decker KL (2002) Compensation for herbivory in wild sunflower: Response to simulated damage by the head-clipping weevil. Ecology (Washington D C) 83:30973107 Prittinen K, Pusenius J, Koivunoro K, Roininen H (2003) Genotypic variation in growth and resistance to insect herbivory in silver birch (Betula pendula) seedlings. Oecologia 137:572-577 Purrington CB (2000) Costs of resistance. Current Opinion in Plant Biology 3:305-308 Rautio P et al. (2002) Developmental plasticity in birch leaves: defoliation causes a shift from glandular to nonglandular trichomes. Oikos 98:437-446 Rausher MD (1992) The Measurement of Selection on Quantitative Traits - Biases Due to Environmental Covariances between Traits and Fitness. Evolution 46:616-626 Rousseaux MC, Julkunen-Tiitto R, Searles PS, Scopel AL, Aphalo PJ, Ballare CL (2004) Solar UV-B radiation affects leaf quality and insect herbivory in the southern beech tree Nothofagus antarctica. Oecologia 138:505-512 Sampedro L, Moreira X, Zas R (2011) Costs of constitutive and herbivore-induced chemical defences in pine trees emerge only under low nutrient availability. Journal of Ecology 99:818-827 Searles PS, Flint SD, Caldwell MM (2001) A meta analysis of plant field studies simulating stratospheric ozone depletion. Oecologia 127:1-10 Siemens DH, Lischke H, Maggiulli N et al (2003) Cost of resistance and tolerance under competition: the defense-stress benefit hypothesis. Evolutionary Ecology 17:247–263 66 Simms EL, Rausher MD (1987) Costs and Benefits of Plant-Resistance to Herbivory. American Naturalist 130:570-581 Simms EL, Triplett J (1994) Costs and Benefits of Plant-Responses to Disease - Resistance and Tolerance. Evolution 48:1973-1985 Skaltsa H, Verykokidou E, Harvala C, Karabourniotis G, Manetas Y (1994) Uv-B Protective Potential and Flavonoid Content of Leaf Hairs of Quercus-Ilex. Phytochemistry 37:987990 Stevens MT, Waller DM, Lindroth RL (2007) Resistance and tolerance in Populus tremuloides: genetic variation, costs, and environmental dependency. Evolutionary Ecology 21:829847 Stinchcombe JR (2002) Environmental dependency in the expression of costs of tolerance to deer herbivory. Evolution 56:1063–1067 Stinchcombe JR (2005) SAS macro for correcting for the artifactual covariance between a mean and plasticity calculated as the difference between means. Available from the author Stowe KA (1998) Experimental evolution of resistance in Brassica rapa: Correlated response of tolerance in lines selected for glucosinolate content. Evolution 52:703-712 Stowe KA, Marquis RJ, Hochwender CG, Simms EL (2000) The evolutionary ecology of tolerance to consumer damage. Annual Review of Ecology and Systematics 31:565-595 Stratmann J (2003) Ultraviolet-B radiation co-opts defense signaling pathways. Trends in Plant Science 8:526-533 Strauss SY, Agrawal AA (1999) The ecology and evolution of plant tolerance to herbivory. Trends in Ecology and Evolution 14:179-185 67 Tegelberg R, Julkunen-Tiitto RJ, Aphalo PJ (2004) Red: far-red light ratio and UV-B radiation: their effects on leaf phenolics and growth of silver birch seedlings. Plant Cell and Environment 27:1005-1013 Tiffin P (2000a) Mechanisms of tolerance to herbivore damage: what do we know? Evolutionary Ecology 14:523-536 Tiffin P (2000b) Are tolerance, avoidance, and antibiosis evolutionarily and ecologically equivalent responses of plants to herbivores? American Naturalist 155:128-138 Tiffin P, Rausher MD (1999) Genetic constraints and selection acting on tolerance to herbivory in the common morning glory Ipomoea purpurea. American Naturalist 154:700-716 Valverde PL, Fornoni J, Nunez-Farfan J (2001) Defensive role of leaf trichomes in resistance to herbivorous insects in Datura stramonium. Journal of Evolutionary Biology 14:424-432 Vandermeijden E, Wijn M, Verkaar HJ (1988) Defense and Regrowth, Alternative Plant Strategies in the Struggle against Herbivores. Oikos 51:355-363 Veteli TO et al. (2007) Do elevated temperature and CO2 generally have counteracting effects on phenolic phytochemistry of boreal trees? Journal of Chemical Ecology 33:287-296 Veteli TO et al. (2003) Interactions between willows and insect herbivores under enhanced ultraviolet-B radiation. Oecologia 137:312-320 Via S, Lande R (1985) Genotype-Environment Interaction and the Evolution of Phenotypic Plasticity. Evolution 39:505-522 Via S, Lande R (1987) Evolution of Genetic-Variability in a Spatially Heterogeneous Environment - Effects of Genotype-Environment Interaction. Genetical Research 49:147-156 68 Weinig C, Stinchcombe JR, Schmitt J (2003) Evolutionary genetics of resistance and tolerance to natural herbivory in Arabidopsis thaliana. Evolution 57:1270-1280 Werker E (2000) Trichome diversity and development. In: Advances in Botanical Research Incorporating Advances in Plant Pathology, vol. 31, pp 1-35 Zavala JA, Ravetta DA (2002) The effect of solar UV-B radiation on terpenes and biomass production in Grindelia chiloensis (Asteraceae), a woody perennial of Patagonia, Argentina. Plant Ecology 161:185-191 Ziska LH, Teramura AH (1992) CO2 Enhancement of Growth and Photosynthesis in Rice (Oryza sativa): Modification by Increased Ultraviolet-B Radiation. Plant Physiology (Rockville) 99:473-481 Zvereva EL, Kozlov MV (2006) Consequences of simultaneous elevation of carbon dioxide and temperature for plant-herbivore interactions: a metaanalysis. Global Change Biology 12:27-4 69 CHAPTER 3 EFFECTS OF ULTRAVIOLET-B RADIATION ON PLASTICITY OF FITNESSRELATED RESPONSES OF ARABIDOPSIS THALIANA TO INSECT HERBIVORY ABSTRACT The amount of ultraviolet-B (UVB) radiation reaching the Earth’s surface has been changing over the past few decades due to the destruction of the stratospheric ozone layer, increased air pollution, and cloud coverage. Current changes in the environment may lead to the evolution of adaptive plastic responses in plants as well as influence interactions of plants with other organisms, such as insect herbivores. Potential evolutionary responses of organisms to environmental change however, will likely depend on the underlying genetic variation that is present in a population. Here, data from a common garden experiment is presented in which recombinant inbred lines (RILs) of Arabidopsis thaliana were exposed to two different levels of UVB radiation (i.e., ambient and excluded UVB) and natural levels of insect herbivory. This study evaluated whether variation in UVB radiation influenced fitness-related traits of A. thaliana genotypes as well as their phenotypic responses to insect herbivory. Results from this study indicate that A. thaliana RILs grown under excluded-UVB were the first to be colonized by insect herbivores and received higher levels of insect damage. Results from this study revealed that variation in UVB and herbivory levels influenced all fitness-related traits measured in a common garden experiment. These traits were affected not only by changes in the UVB environment, but were also genetically dependent, which may have implications for the evolution of these traits in future changing environments. More interestingly, UVB levels influenced plant responses to insect herbivory, in terms of total fruit production and flowering 70 time. This result is important because it implies that changes in UVB radiation may influence the dynamics of plant-insect interactions. 71 INTRODUCTION Multiple environmental changes influencing organisms and their interactions have been witnessed during the past few decades. Such changes in the environment may lead to the evolution of plastic responses, which may be reflected in variation in phenotypic traits related to growth, development, and/or reproduction. These plastic changes represent mechanisms by which organisms can respond, at least in the short term, to challenges imposed by variable or heterogeneous environments (Bradshaw 1965; Travis 1994; Schmitt et al. 1999; Sultan 2000). Phenotypic plasticity is a widely accepted adaptive strategy, particularly useful to sessile organisms like plants, which have limited capacity to escape in space from the potential detrimental effects of currently changing environments (Bradshaw 1965). To predict to what extent such plastic responses may influence the evolutionary trajectory of a species and its interactions with other species, it is imperative to evaluate whether there is genetic variation in plasticity as well as genetic variation in the plastic response to environmental factors. This is essential because the extent, to which a population may evolve in the face of a changing environment, will likely depend on the underlying genetic variation present at the population level. Ultraviolet-B radiation represents a small fraction of the solar spectrum ranging from 290 to 315 nm. The level of UVB radiation reaching the Earth’s surface has been consistently changing during the last decades due to the depletion of the trophospheric ozone layer, pollution levels, and cloud coverage (Intergovernmental Panel on Climate Change, IPCC 2007). Several studies have evaluated the effect of UVB radiation on plant growth and physiological processes as well as plant chemical responses to this factor, which appear to vary among plant species 72 (Hader 2000; Searles et al. 2001a; Searles et al. 2001b). Whereas some studies failed to find significant effects of UVB radiation on plant growth and photosynthetic processes (Searles et al. 2001a; Tegelberg et al. 2004), other studies have shown UVB-induced decreases in leaf area and growth rate (likely as a result of decreases in photosynthetic rate and increases in leaf thickness (Bornman and Teramura 1993; Stratmann 2003). Changes in UVB radiation may also modify flowering patterns, biomass accumulation, and reproductive traits. With respect to plant chemistry, UVB radiation is known to increase accumulation of UVB-absorbing compounds, such as flavonoids and phenolics, which in turn may interfere with plant interactions with their herbivores (Agrell et al. 2000; Searles et al. 2001; Hofmann et al. 2003; Izaguirre et al. 2003; Bassman 2004; Izaguirre et al. 2007). On the other hand, levels of secondary chemicals, such as terpenoids may be unaffected or even decreased in some plant species when exposed to enhanced UVB radiation levels (Singh 1996; Ambasht and Agrawal 1997). Several studies have evaluated the interactive effect of UVB radiation with other abiotic agents on plant responses. For example, abiotic factors such as elevated CO2 (Ziska and Teramura 1991; McDonald et al. 1999; Agrell et al. 2000; Lavola et al. 2000), nutrient and water stress (Hofmann et al. 2003), temperature changes (Tevini 1993), and heavy metals (Dube and Bornman 1992) have been shown to modify plant response to UVB radiation. On the other hand, only few studies have addressed the interactive effect of UVB radiation with biotic agents such as insect herbivores (Ballare et al. 1996; McCloud and Berenbaum 1999; Caputo et al. 2006; Caldwell et al. 2007). These studies, although informative, have mostly been conducted under laboratory conditions, but see Caputo et al. (2006). More importantly, they did not evaluate intraspecific variation in plant fitness responses to insect herbivory under variable UVB radiation 73 levels. In general, previous studies have mainly focused on assessing effects of UVB radiation on insect performance and the amount of insect damage suffered by plants. For examples, studies have shown that UVB may affect the nutrient levels of plants thereby leading to poor larval performance, increase in the number of days to pupation and late eclosion (Leonard and Juliano 1995). Additionally, UVB radiation may also induce responses similar to those induced by herbivory, such as production of secondary chemicals, which may be directly toxic to herbivores, and thus, reduce the amount of damage suffered by plants (Agrell et al. 2000; Searles et al. 2001; Hofmann et al. 2003; Izaguirre et al. 2003; Bassman 2004; Izaguirre et al. 2007). Furthermore, it is not known whether effects of UVB radiation on patterns of insect distribution and damage among plants in the field are genetically dependent. This information, in addition to potential genetic variation in the plant phenotypic response to both UVB radiation and insect herbivory is important because the interactive effect of these factors may lead to natural selection favoring some genotypes over others, which may in turn alter the co-evolution of plant-insect interactions. This study presents results from a common garden experiment in which recombinant inbred lines (RILs) of Arabidopsis thaliana were exposed to two different levels of UVB radiation (i.e., ambient- and excluded-UVB) as well as to the presence and absence of natural levels of insect herbivory. The interactive role of plant genotype, UVB radiation, and insect herbivory was assessed on plant fitness-related responses of the annual plant A. thaliana. Patterns of insect distribution and herbivory in the field were also evaluated to assess whether these could explain observed changes in plant phenotypic responses to insect herbivore attacks. Information from this experiment would allow us to make predictions regarding whether changes 74 in UVB radiation may influence phenotypic responses of plants to a natural insect community, and thus, the evolution of plant-insect interactions in future changing environments. MATERIALS AND METHODS Study Plant Species Arabidopsis thaliana (L.) Heynh (Capparales: Brassicaceae), also known as the mouseear cress, is an annual herbaceous plant native to Europe and Asia with a wide geographic distribution across North America and North-western Africa (Baskins and Baskins 1983; Hoffmann et al. 2003). This species is a self-fertilizing plant, characterized by a short life cycle, which is naturally found in open or disturbed habitats (Baskins and Baskins 1983; Hoffmann et al. 2003). Because of the high selfing rates (less than 1% outcrossing rates), natural populations seem to consist majorly of single genotypes. However, significant levels of genetic variation in phenotypic traits exist among different local populations of this species (Bergelson et al. 1998; Mauricio 1998; Nordborg and Bergelson 1999). Recombinant inbred lines of A. thaliana (100 RILs) derived from a cross between the ecotypes Columbia (Col-0) and Landsberg (Ler), and advanced to the eighth generation through a single seed descent were used in this study (Lister and Dean 1993). These RILs were used to generate a genetically variable population since the two parental lines (Col-0 and Ler) are known to differ in several characteristics related to growth, reproduction, phenology, and both chemical and structural defenses (Mauricio 2005; Weinig et al. 2003). Thus, genetic and phenotypic variation among these RILs allows for the assessment of intraspecific variation in the response of this plant species to both insect herbivory and UVB radiation. 75 Experimental Design To address the question of whether UVB radiation can influence plasticity of plant phenotypic responses to insect herbivory, the model plant Arabidopsis thaliana was used in a common garden experiment in which the presence/absence of natural insect herbivory and the levels of UVB radiation were manipulated. Specifically, a split-plot design was used to assess the effect of UVB radiation, insect herbivory, and plant genotype on plant phenotypic responses. The two UVB treatment levels consisted of ambient (UVB+) and excluded (UVB-) ultraviolet-B radiation randomly assigned to 20 different plots (i.e., 10 plots assigned to each UVB level). While ultraviolet-B radiation was excluded from the randomly assigned UVB- plots using Mylar polyester film, UVB+ plots were covered with clear acetate film allowing most of the UVB radiation to pass through. Mylar film transmits radiations only in the wavelength greater than 315 nm thereby blocking all the UVB radiation but transmitting radiations in the ultraviolet-A regions. On the other hand, clear acetate film allowed most UVB to pass through. Arabidopsis thaliana genotypes consisted of 100 recombinant inbred lines (RILs), which were planted at random within each of the plots. Two sets of 100 RILs were randomly assigned to each of the 20 plots (total of 4000 plants). Insect herbivores were excluded from sprayed plots by spraying with a pesticide (Proclaim by Syngenta Inc. USA) at ten days interval. Plants excluded from insect herbivores were inspected on daily basis to ensure no herbivore damage. Each plot (both herbivory and sprayed plots) was surrounded with a wire-mesh fence to exclude mammalian herbivory. In terms of growth conditions, seeds were cold-treated for 5 days at 4°C to break dormancy. Prior to transplanting to the field, plants were initially raised in a 50 well flats in 76 laboratory growth chambers. Within the chambers, plants were exposed to 10 hours of day and 14 hours of night, with day and night temperatures of 20°C and 16°C, respectively. Plants were rotated within growth chambers every other day and among chambers every week to avoid chamber and position effects. After five weeks of growth in the chambers, plants were transplanted to the field (Ecological Research Station, Bowling Green State University, Bowling Green, Ohio). Plant phenotypic traits In order to assess responses of plant RILs to herbivory under different UVB levels, variables related to phenology (flowering time), plant growth (rosette biomass), and reproduction (total fruit production) were measured in plots with herbivores present and excluded. Similarly, the effects of UVB radiation, and RIL were assessed on days to plant colonization and percent area of rosette leaves removed by insect herbivory. Flowering time was estimated as the number of days required for the first flower per plant to be fully opened. Plant rosette biomass was measured as rosette dry weight by oven drying the rosette leaves for two days at 70°C. Plant fitness was estimated as the total fruit number produced over the plant lifetime, since this measure is highly correlated with total seed production (Bidart-Bouzat 2004). To document the order of colonization of insect herbivores among plants, all plants were inspected daily for insect damage and the numbers of days to first observe herbivore damage on each plant was recorded herein, referred to as colonization. Statistical Analysis To evaluate responses of plant genotypes to herbivory under different UVB levels, analyses of variance were performed for each of the measured variables, flowering time, rosette 77 biomass and total fruit production using PROC GLM (SAS statistical software, version 9.1). Data was either log- or ranked-transformed to correct for normality and heteroscedasticity when necessary. For this purpose, a mixed linear model was used, considering the herbivory and UVB treatments as fixed effects, and plant genotype (RIL) as a random effect. In addition, similar analyses were performed on the order of insect herbivore colonization among RILs and plant rosette damage in order to evaluate the effects of UVB radiation, and genotype on patterns of plant utilization by insect herbivores. RESULTS Analyses of variance revealed that A. thaliana differed in their fitness-related responses to UVB exclusion and insect herbivory (Table 1). With respect to the number of fruits produced, both UVB exclusion and herbivory significantly affected this trait (see Table 1 and Figure 1A). Exclusion of UVB resulted in a significant increase in total fruit production and herbivory led to a general decrease in the total number of fruits (Figure 1A and B). Additionally, plant responses to herbivory in terms of total number of fruits produced were found to be dependent on UVB radiation levels (Table 1). This significant interaction between herbivory and UVB radiation levels has particular evolutionary significance since future changes in UVB radiation levels may modify plant phenotypic response to their herbivores. The effect of genotype (RIL) on total fruit number was statistically not significant (Table 1). However, separate analyses of variance for each UVB levels (when pooling herbivore treatments) revealed a significant effect of RIL on total fruit number (P= 0.0107 for UVB+; and P= <0.0001 for UVB-), indicating the existence of genetic variation in this trait. All phenotypic responses of plants to UVB radiation were found to be genotype dependent (significant UVB × RIL interaction), but those to insect herbivory were 78 not (see Table 1). In addition, the three-way interaction UVB × herbivory × RIL was statistically not significant indicating that UVB radiation levels affected total number of fruits produced in response to herbivory across the genotypes in the same manner. With respect to rosette biomass, UVB exclusion and herbivory significantly affected this trait (Table 1). Exclusion of UVB resulted in a significant increase in rosette biomass, and herbivory led to a general decrease in rosette biomass (Figure 1C and D). Overall, the effect of genotype (RIL) on rosette biomass was statistically not significant (Table 1). However, separate analyses of variance for each UVB levels revealed a significant effect of genotypes on rosette biomass (P= <.0001 for UVB+; and P= <.0001 for UVB-), indicating the existence of genetic variation in this trait. Rosette biomass responses to UVB were found to be dependent on genotype (significant UVB × RIL interactions), but not on herbivory (Table 1). Furthermore, responses of plant to herbivory were not genotype dependent. Analysis of the three-way interaction (i.e., UVB × herbivory × RIL) was not significant indicating that UVB radiation levels affected rosette biomass produced in response to herbivory across the genotypes in the same manner. Analyses of variance also revealed that UVB, herbivory, and genotype significantly affected the number of days to first flowering (Table 1). Overall, exclusion of UVB radiation resulted in a decrease in the number of days to first flowering, and herbivory led to an increase in this trait value (Figure 1E and F). For this trait, a significant interaction between herbivory and UVB radiation was observed, indicating that plant responses to UVB radiation in terms of flowering time is dependent on herbivory (Table 1). Moreover, while phenological responses of plants to UVB variation were genotype-dependent (significant UVB × RIL interaction), 79 responses to herbivory did not significantly depend on the genotype (Table 1). Finally, the threeway interaction of UVB radiation, herbivory, and genotype was significant, which implies that UVB radiation differentially affected plant response to herbivory across the genotypes. Exclusion of UVB led to a significant decrease in the number of days to plant colonization by insect herbivores (Table 1 and Figure 2A). The effects of genotype on days to herbivore colonization were marginally significant, and the effect of UVB radiation on this variable was not dependent on plant genotype (Figure 2A). In addition, changes in UVB radiation significantly affected levels of herbivore damage (Table 1 and Figure 2B). Plants excluded from UVB radiation had more damage than plants exposed to ambient UVB. Although, the effect of RILs was not significant on the levels of insect herbivore damage, separate analyses of variance under the two UVB radiation levels revealed a significant effect of RILs on the levels of herbivore damage (P=0.0055 for UVB+; and P<.0001 for UVB-). In addition, variation in damage levels as a result of differences in UVB radiation levels was also dependent on plant genotype (significant UVB × RIL; Table 1 and Figure 2B). DISCUSSION Even though several studies have investigated the effects of UVB radiation and insect herbivory on plants, information is lacking regarding the effect of these two important environmental factors on plant fitness-related responses. This study is also of evolutionary significance because the ability of plant populations to evolve and the extent to which a population may evolve in a changing environment will depend on the available genetic variation within such population. In addition, this study revealed a significant UVB × RIL interaction for 80 all measured variables suggesting the presence of evolutionary potential of these traits to respond to changes in UVB radiation. This significant interaction is important as it may lead to the maintenance of genetic variation under variable environments, and allow populations to evolve in changing environments (Via and Lande 1985; Via and Lande 1987; Gillespie and Turelli 1989). Overall, the effect of UVB radiation was detrimental to plant fitness-related variables measured in this study. Exclusion of UVB radiation led to an increase in the total number of fruits and rosette biomass produced, as well as a delay in the number of days required for plants to first flower. In the same vein, insect herbivory also had a negative effect on measured phenotypic variables. This result is consistent with other studies that have shown that plants may suffer fitness consequences because of exposure to high levels of UVB (Teramura et al. 1990; Teramura and Sullivan 1991; Grammatikopoulos et al. 1994). This may be through growth inhibition, alteration of photosynthetic enzyme activities, modification of stomata closure, and alteration of growth patterns (e.g., decreases in stem length and leaf area, and changes in flowering phenology) (Teramura 1993; Hader 2000; Searles et al. 2001a; Searles et al. 2001b Bornman and Stratmann 2003). The negative impact of UVB radiation on fitness may also be due to energy costs of synthesizing protective secondary metabolite in response this type of stress (Logemann et al. 2000). In the same manner, insect herbivory may also have negative consequences on plant fitness such as reduced seed production , increased fruit abortion, decreased growth rate, changes in plant architecture, and delayed senescence (Marquis 1992; Bidart-Bouzat 2004; Rudgers and Whitney 2006). It is important to emphasize that this study revealed a significant interaction between herbivory and UVB, indicating an effect of variation in 81 UVB levels on plasticity of plant-fitness responses to insect herbivory. Thus, future changes in UVB radiation levels may significantly affect plant-insect interactions and the fitness outcome of plants exposed to natural insect communities. With respect to patterns of plant utilization by insect herbivores, increase in UVB led to a decrease in insect herbivore damage and an increase in the number of days to herbivore colonization. This result may be due to two major reasons. Insect herbivores may detect differences in the levels of UVB radiation and present avoidance behavior to high levels of UVB radiation (Mazza et al. 1999; Mazza et al. 2002). Alternatively, UVB may mediate changes in plant morphology and chemistry, which in turn influences the amount of insect herbivory. For example, previous studies have suggested that trichomes, which are important structural barriers for insect herbivores, may also be involved in the protection of plants against harmful effects of UVB radiation. This protective mechanism is due to the ability of trichomes to reflect radiations in the UVB region (Ehleringer 1984; Karabourniotis and Bornman 1999). There is experimental evidence that trichome production was induced by UVB radiation in this study (see Chapter 2). In addition to induction of physical barriers to herbivores, UVB radiations may also increase the concentration of UVB absorbing compounds such as phenolics, flavonoids, and other secondary chemicals that may directly influence plant utilization by herbivores (Lavola et al. 2003). For instance, Izzaguire et al. (2007) found that UVB radiation induced the same phenolics compounds that were induced by herbivory, suggesting that UVB radiation may trigger the same defense pathway as insect herbivory. Information also exists on effects of UVB radiation influencing concentrations of glucosinolates, a well-studied chemical defense system in the Brasssicaceae plant family (Barro et al. 2002; Schreiner et al. 2009). For instance, Schreiner et 82 al. (2009) found up to six-fold increase in glucotropaolein, the only type of glucosinolates present in nasturtium (Tropaeolum majus) after exposure to UVB radiation. Likewise, Baro et al. (2002) observed similar increases in glucosinolate contents in the Ethiopian mustard (Brassica carinata) when exposed to UVB radiation. This UVB-mediated increase in glucosinolates has been suggested to be a stress response similar to that observed in plant defense responses to insect herbivory and pathogen attack (Schreiner et al. 2009). Taken together, these findings suggest that UVB radiation may have induced both physical and chemical defenses in our system, which in turn may have pronounced effects on patterns of plant use by insect herbivores. Intuitively, one would expect that lower herbivory impact (less plant damage) under UVB+ should be associated with enhanced plant fitness in this environment; this was not the case however, in this study. Plants grown under UVB- had more herbivore damage, yet had higher fitness. There are two possible reasons for this observation. First, it is possible that the detrimental effects of UVB radiation are more severe than the effects of herbivory on plant fitness; that is, potential fitness gains due to a reduction in levels of herbivory in UVB+ environments could not compensate for the reduction in fitness due to UVB radiation. Second, there might be a cost associated to plastic responses of plants to UVB radiation. This cost, which may be related to the production of UVB absorbing compounds or due to genetic causes such as potential deleterious effects of genes involved in plasticity (Pigliucci 2001) may account for the lower plant fitness in UVB+ plots even though insect damage was decreased in this environment. Given that the interaction between UVB and insect herbivory was significant and there was a marked decrease in the plasticity of total fruit production (estimate of fitness in this study) in 83 response to UVB exclusion, it is reasonable to infer that the effects of UVB radiation were likely more detrimental than those of insect herbivory on plant fitness. Nonetheless, potential physiological and genetic costs on plant fitness cannot be discarded, and may also at least partly explain the observed results. In conclusion, this study addressed how variation in UVB radiation levels may influence potential fitness consequences of insect herbivory in plants. Although overall, UVB radiation had a negative effect on plant fitness, this study revealed the presence of genetic variation in both UVB environments suggesting a potential for plants to evolutionary respond to future changes in UVB radiation levels. It is also important to note that although UVB radiation tended to reduce the amount of herbivore damage inflicted on plants (likely due to increased induction of protective secondary metabolites and physical structures), this reduction was not enough to compensate for the negative effects of UVB radiation on plants. 84 Table 1. Mixed model analysis of variance showing the effects of UVB radiation, insect herbivory and RIL on plant phenotypic variables of A. thaliana. Source of Variation Block Herbivory UVB Herbivory x UVB RIL Herbivory x RIL UVB x RIL Herbivory x UVB x RIL Flowering date 1.76 181.27*** 649.93*** 6.98** 4.00*** 0.92 2.38*** 1.29* Fruits 22.60*** 817.48*** 246.48*** 98.98*** 1.01 0.92 1.45* 1.03 Significance levels †P<0.1, *P<0.05, **P<0.01, ***P<0.001 Rosette Colonization Rosette Biomass Damage 6.98*** 11.49*** 0.61 235.99*** 16.48*** 155.80*** 157.43*** 0.03 0.71 1.30† 1.28 0.98 17.65*** 1.18 2.02*** 0.49 - 85 (A) (B) 600 Total Fruit Number Total Fruit Number 300 * 200 500 * 400 400 300 200 100 UVB+ UVB- Rosette Biomass 0.06 * 0.04 0.02 UVB+ UVB+ 0.12 0.10 0.08 0.06 0.04 UVB- UVB- Days to Flowering 68 68 * Days to Flowering (F) UVB- 0.02 (E) 70 UVB+ 66 64 62 60 66 * Rosette Biomass 0.08 (D) 0.14 0.10 * (C) 64 62 60 58 58 UVB+ Herbivory UVB- UVB+ Sprayed UVB- Figure 1. Reaction norms showing A. thaliana phenotypic responses, in terms of total fruit production (A-B), rosette biomass (C-D), and number of days to first flowering (E-F) to variation in UVB radiation levels in herbivory and sprayed plots. Means of RILs are represented with closed circles and represent significant differences in phenotypic means between UVB treatment levels. * P<0.001 86 (A) 22 * Days to Colonization 24 20 18 16 14 12 10 12 10 * Rosette Damage (%) (B) 8 UVB+ UVB- Figure 2. Reaction norms showing A. thaliana phenotypic responses, in terms of days to herbivore colonization (A), and rosette damage (B). Means of RILs are represented with closed circles represent significant differences in phenotypic means between UVB treatment levels. * P<0.001 87 REFERENCES Agrell J, McDonald EP, Lindroth RL (2000) Effects of CO2 and light on tree phytochemistry and insect performance. Oikos 88:259-272 Ambasht NK, Agrawal M (1997) Influence of supplemental UVB radiation on photosynthetic characteristics of rice plants. Photosynthetica 34:401-408 Ballare CL, Scopel AL, Stapleton AE, Yanovsky MJ (1996) Solar ultraviolet-B radiation affects seedling emergence, DNA integrity, plant morphology, growth rate, and attractiveness to herbivore insects in Datura ferox. Plant Physiology 112:161-170 Barro F, Fernadez-Escobar J, De la Vega M, Martin A (2002) Modification of glucosinolate and erucic acid contents in doubled haploid lines of Brassica carinata by UV treatment of isolated microspores. Euphytica 129:1-6 Baskins JM, Baskins CC (1983) Seasonal changes in the germination responses of buried seeds of Arabidopsis thaliana and ecological interpretation. Botany Gazette 144:540-543 Bassman JH (2004) Ecosystem consequences of enhanced solar ultraviolet radiation: Secondary plant metabolites as mediators of multiple trophic interactions in terrestrial plant communities. Photochemistry and Photobiology 79:382-398 Bergelson J, Stahl E, Dudek S, Kreitman M (1998) Genetic variation within and among populations of Arabidopsis thaliana. Genetics 148:1311-1323 Bidart-Bouzat MG (2004) Herbivory modifies the lifetime fitness response of Arabidopsis thaliana to elevated CO2. Ecology (Washington D C) 85:297-303 Bidart-Bouzat MG, Mithen R, Berenbaum MR (2005) Elevated CO2 influences herbivoryinduced defense responses of Arabidopsis thaliana. Oecologia (Berlin) 145:415-424 88 Bornman JF, Teramura AH (1993) Effects of ultraviolet-B radiation on terrestrial plants. Environmental UV photobiology: 427-471 Bradshaw AD (1965) Evolutionary significance of phenotypic plasticity in plants. Advances in Genetics 13: 115-155 Caldwell MM, Bornman JF, Ballare CL, Flint SD, Kulandaivelu G (2007) Terrestrial ecosystems, increased solar ultraviolet radiation, and interactions with bother climate change factors. Photochemical & Photobiological Sciences 6:252-266 Caputo C, Rutitzky M, Ballare CL (2006) Solar ultraviolet-B radiation alters the attractiveness of Arabidopsis plants to diamondback moths (Plutella xylostella L.): impacts on oviposition and involvement of the jasmonic acid pathway. Oecologia 149:81-90 Dube SL, Bornman JF (1992) Response of Spruce Seedlings to Simultaneous Exposure to Ultraviolet-B Radiation and Cadmium. Plant Physiology and Biochemistry 30:761-767 Ehleringer J (1984) Ecology and ecophysiology of leaf pubescence in North American desert plants. Plenum Press, New York Gillespie JH, Turelli M (1989) Genotype-Environment Interactions and the Maintenance of Polygenic Variation. Genetics 121:129-138 Grammatikopoulos G, Karabourniotis G, Kyparissis A, Petropoulou Y, Manetas Y (1994) Leaf Hairs of Olive (Olea europaea) Prevent Stomatal Closure by Ultraviolet-B Radiation. Australian Journal of Plant Physiology 21:293-301 Hader DP (2000) Effects of solar UVB radiation on aquatic ecosystems. Advance Space Research 26:2029-2040 89 Hoffmann MH, Bremer M, Schneider K, Burger F, Stolle E, Moritz G (2003) Flower visitors in a natural population of Arabidopsis thaliana. Plant Biology 5:491-494 Hofmann RW et al. (2003) Responses to UVB radiation in Trifolium repens L. - physiological links to plant productivity and water availability. Plant Cell and Environment 26:603612 IPCC (2007). Contribution of Working Group I to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. Cambridge University Press, Cambridge, United Kingdom. Izaguirre MM, Mazza CA, Svatos A, Baldwin IT, Ballare CL (2007) Solar ultraviolet-B radiation and insect herbivory trigger partially overlapping phenolic responses in Nicotiana attenuata and Nicotiana longiflora. Annals of Botany 99:103-109 Izaguirre MM, Scopel AL, Baldwin IT, Ballare CL (2003) Convergent responses to stress. Solar ultraviolet-B radiation and Manduca sexta herbivory elicit overlapping transcriptional responses in field-grown plants of Nicotiana longiflora. Plant Physiology 132:1755-1767 Karabourniotis G, Bornman JF (1999) Penetration of UV-A, UVB and blue light through the leaf trichome layers of two xeromorphic plants, olive and oak, measured by optical fibre microprobes. Physiologia Plantarum 105:655-661 Lavola A, Aphalo PJ, Lahti M, Julkunen-Tiitto R (2003) Nutrient availability and the effect of increasing UVB radiation on secondary plant compounds in Scots pine. Environmental and Experimental Botany 49:49-60 90 Lavola A, Julkunen-Tiitto R, de la Rosa TM, Lehto T, Aphalo PJ (2000) Allocation of carbon to growth and secondary metabolites in birch seedlings under UVB radiation and CO2 exposure. Physiologia Plantarum 109:260-267 Léonard PM, Juliano SA (1995). "Effect of leaf litter and density on fitness and population performance of the hole mosquito Aedes triseriatus". Ecological Entomology (0307 6946), 20(2), p.125. Lister C, Dean C (1993) Recombinant Inbred Lines for Mapping Rflp and Phenotypic Markers in Arabidopsis thaliana. Plant Journal 4:745-750 Logemann E, Tavernaro A, Schulz WG, Somssich IE, Hahlbrock K (2000) UV light selectively coinduces supply pathways from primary metabolism and flavonoid secondary product formation in parsley. Proceedings of the National Academy of Sciences of the United States of America 97:1903-1907 Marquis RJ (1992) Selective impacts of herbivores. University of Chicago press, Chicago Mauricio R (1998) Costs of resistance to natural enemies in field populations of the annual plant Arabidopsis thaliana. American Naturalist 151:20-28 Mauricio R (2005) Ontogenetics of QTL: the genetic architecture of trichome density over time in Arabidopsis thaliana. Genetica 123:75-85 Mazza CA, Izaguirre MM, Zavala J, Scopel AL, Ballare CL (2002) Insect perception of ambient ultraviolet-B radiation. Ecology Letters 5:722-726 Mazza CA, Zavala J, Scopel AL, Ballare CL (1999) Perception of solar UVB radiation by phytophagous insects: Behavioral responses and ecosystem implications. Proceedings of the National Academy of Sciences of the United States of America 96:980-985 91 McCloud ES, Berenbaum M (1999) Effects of enhanced UVB radiation on a weedy forb (Plantago lanceolata) and its interactions with a generalist and specialist herbivore. Entomologia Experimentalis Et Applicata 93:233-247 McDonald EP, Agrell J, Lindroth RL (1999) CO2 and light effects on deciduous trees: growth, foliar chemistry, and insect performance. Oecologia 119:389-399 Nordborg M, Bergelson J (1999) The effect of seed and rosette cold treatment on germination and flowering time in some Arabidopsis thaliana (Brassicaceae) ecotypes. American Journal of Botany 86:470-475 Pigluicci M (2001) Phenotypic plasticity. Fox C.W et al. (Ed). Evolutionary ecology: Concepts and case studies. New York, USA. Oxford University Press. 58-69. Rousseaux MC, Julkunen-Tiitto R, Searles PS, Scopel AL, Aphalo PJ, Ballare CL (2004) Solar UVB radiation affects leaf quality and insect herbivory in the southern beech tree Nothofagus antarctica. Oecologia 138:505-512 Rudgers JA, Whitney KD (2006) Interactions between insect herbivores and a plant architectural dimorphism. Journal of Ecology 94:1249-1260 Schmitt J, Dudley SA, Pigliucci M (1999) Manipulative approaches to testing adaptive plasticity: Phytochrome-mediated shade-avoidance responses in plants. American Naturalist 154:S43-S54 Schreiner M, Krumbein A, Mewis I, Ulrichs C, Huyskens-Keil S (2009) Short-term and moderate UVB radiation effects on secondary plant metabolism in different organs of nasturtium (Tropaeolum majus L.). Innovative Food Science & Emerging Technologies 10:93-96 92 Searles PS, Flint SD, Caldwell MM (2001a) A meta analysis of plant field studies simulating stratospheric ozone depletion. Oecologia 127:1-10 Searles PS, Kropp BR, Flint SD, Caldwell MM (2001b) Influence of solar UVB radiation on peatland microbial communities of southern Argentina. New Phytologist 152:213-221 Singh A (1996) Growth, physiological, and biochemical responses of three tropical legumes to enhanced UVB radiation. Canadian Journal of Botany-Revue Canadienne De Botanique 74:135-139 Stratmann J (2003) Ultraviolet-B radiation co-opts defense signaling pathways. Trends in Plant Science 8:526-533 Sultan SE (2000) Phenotypic plasticity for plant development, function and life history. Trends in Plant Science 5:537-542 Tegelberg R, Julkunen-Tiitto RJ, Aphalo PJ (2004) Red: far-red light ratio and UVB radiation: their effects on leaf phenolics and growth of silver birch seedlings. Plant Cell and Environment 27:1005-1013 Teramura AH, Sullivan JH (1991) Field Studies of UVB Radiation Effects on Plants Case Histories of Soybean and Loblolly Pine. Abrol, Y. P., Et Al. (Ed.). Impact of Global Climatic Changes on Photosynthesis and Plant Productivity; Indo-Us Workshop, New Delhi, India, January 8-12, 1991. Xlviii+824p. Asia Publishing House Ltd.: Sittingbourne, England, UK; Oxford and Ibh Publishing Co. Pvt. Ltd.: New Delhi, India. Illus. Maps: 147-161 Teramura AH, Sullivan JH, Lydon J (1990) Effects of UVB Radiation on Soybean Yield and Seed Quality - a 6-Year Field-Study. Physiologia Plantarum 80:5-11 93 Tevini M (1993) Effects of enhanced UVB radiation on terrestrial plants. UVB radiation and ozone depletion: Effects on humans, animals, plants, microorganisms, and materials: 125-153 Travis J (1994) Evaluating the adaptive role of morphological plasticity. Ecological morphology: Integrative organismal biology: 99-122 Veteli TO et al. (2003) Interactions between willows and insect herbivores under enhanced ultraviolet-B radiation. Oecologia 137:312-320 Via S, Lande R (1985) Genotype-Environment Interaction and the Evolution of Phenotypic Plasticity. Evolution 39:505-522 Via S, Lande R (1987) Evolution of Genetic-Variability in a Spatially Heterogeneous Environment - Effects of Genotype-Environment Interaction. Genetical Research 49:147-156 Weinig C, Stinchcombe JR, Schmitt J (2003) QTL architecture of resistance and tolerance traits in Arabidopsis thaliana in natural environments. Molecular Ecology 12:1153-1163 Ziska LH, Teramura AH (1991) Interaction of Elevated Carbon Dioxide and UVB Radiation on Growth and Photosynthesis in Rice. Plant Physiology (Rockville) 96:22