Investigations Into Mechanisms of Ash Resistance to The

INVESTIGATIONS INTO MECHANISMS OF ASH RESISTANCE TO THE

EMERALD ASH BORER

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Justin Graham Alexander Whitehill

Graduate Program in Plant Pathology

The Ohio State University

2011

Dissertation Committee:

Dr. Pierluigi Bonello, Advisor

Dr. Daniel A. Herms, Co-Advisor

Dr. Jennifer L. Koch

Dr. Larry Phelan

Dr. Brian McSpadden Gardener

Dr. Guo-liang Wang

! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! Copyright by

Justin Graham Alexander Whitehill

2011

! !

ABSTRACT

The emerald ash borer (EAB; Agrilus planipennis Fairmaire) is an invasive phloem- feeding, wood-boring insect that was introduced to southeastern Michigan from Asia and first discovered in the U.S. in 2002. EAB larvae kill ash (Fraxinus spp.) trees by disrupting the flow of water and nutrients throughout the main stem of host plants. While all North American species of ash are susceptible to attack by EAB, an Asian species,

Manchurian ash, is resistant. It is hypothesized that Manchurian ash resists EAB colonization by deploying targeted defenses that developed over time by virtue of its co- evolutionary history with the insect. These defenses can be broken down into constitutive and induced, physical and chemical defenses. We investigated constitutive and induced mechanisms of resistance in ash using proteomic and metabolomics approaches to study physiologically altered trees belonging to resistant and susceptible species from diverse phylogenetic backgrounds.

We found significantly increased EAB attack rates below the girdle in susceptible ash species but not in resistant Manchurian ash, while exogenous treatment with a regulatory defense phytohormone, methyl jasmonate (MeJA) provided protection from attack by

EAB at the same rate as an insecticide formulated against phloem feeding insects

(Chapter 2). Application of MeJA to the outer bark of Manchurian and white ash also had significant effects on phloem phenolic profiles (Chapter 5). Furthemore, four

proteins were constitutively expressed at significantly higher levels in the phloem tissues of Manchurian ash than in susceptible black, green, and white ashes and may therefore represent important components of constitutive resistance of Manchurian ash (Chapter

3). Comparisons of constitutive phloem phenolic profiles of six ash species revealed differences in secondary chemistry that were reflective of their phylogenetic relationships and may be associated with constitutive resistance (Chapter 4). Finally, proteomic and metabolomic characterization of oral secretions of 4th instar EAB larvae revealed a highly active molecular environment reflective of the dynamic interaction between ash and EAB

(Chapter 6).

iii

DEDICATION

To my grandparents, my mother Barbara Whitehill, my father Frederick Whitehill, and

my Karlita. Their love and support have shaped me into the person I am today.

ACKNOWLEDGMENTS

I want to first thank Karla Medina-Ortega for her love and friendship. She has given me direction in life and made me a better person. It was an amazing experience to work on the completion of our dissertations together and without her it would not have been possible.

I thank my committee members Pierluigi “Enrico” Bonello, Dan Herms, Larry

Phelan, Jennifer Koch, Brian McSpadden Gardener, and Guo-Liang Wang for their insight, support, and help throughout the course of my degree. I especially thank my advisor Enrico for his patience, understanding, and ability to distil situations and give a brutally honest opinion when it is needed most. He has always encouraged me to explore and expand my research in directions I found interesting while helping to guide me and focus my scientific endeavors. Dan has also been an excellent mentor to me throughout the course of my graduate program and provided critical insights that allowed me to see issues from a new perspective. I thank Larry for his passion and for introducing me to the world of chemical ecology and to a group of people with whom I will always share a special bond, including my Karlita. Jennifer also has been an invaluable mentor that has always been willing to discuss my research and provide me with her valuable insights.

Brian helped with my understanding of statistics, and more importantly he has always taken the time to meet with me and provide me with new approaches, ideas, and ways to

enhance my investigations. Finally, Guo-Liang’s classes in agricultural genomics provided me with a strong background and understanding of molecular plant biology that

I thought would always be beyond my intellectual grasp.

I thank all the members of the Bonello lab that contributed to my research and experience during my time at Ohio State including Chris Wallis, Francis Ockels, Alieta

Eyles, Duan Wang, Scott Williams, Shannon Quinn, Annemarie Nagle, Gerardo Suazo,

Robbie Snyder, Nelda Vasquez, Zhifeng Zheng, Patrick Sherwood, Alifia Merchant,

Tania Burgos, Ronald Battallas, Gabrielle Thottam, Amy Hill, Sourav Chakraborty, Anna

Conrad, Caterina Villari, Katie Gambone, and Ahmed Najar. I especially thank Nathan

Kleczewski who was a mentor and a big brother to me for the first three years of my program. He was a role model that showed me what it takes to be a successful graduate student. Also, Gerardo Suazo did an amazing job working with me during 2009, and I owe him a large debt of gratitude and thanks. He contributed significantly to many of my projects, and I consider it a success if his experience of working with me was even half as productive and memorable as my experience was working with him. Ronald Batallas was only a member of the lab for a few months, but he contributed significantly to this dissertation’s Chapter 6. More importantly, Ronald’s work ethic, natural biological instincts, and overall enthusiasm about research were an inspiration to me.

I also thank the Herm’s lab including Bryant Chambers for help with establishing my research plot in Bowling Green, OH and help with various aspects of field work,

Diane Hartzler for her role in establishing the plot at Bowling Green, and Vanessa

Muilenberg for insight into LC-MS analyses. vi

I thank my collaborators including Dave Bienemann, Amy Stone, Sasha Popova-

Butler, Kari Green-Church, Therese Poland, Yigen Chen, Don Cippollini, Dave Mackey,

Karla Medina, and Stephen Opiyo. Also, Sasha Popova-Butler was an excellent collaborator in my proteomic work and I was able to learn much from her over the years.

I especially thank Dave Bienemann for his enthusiasm and friendship throughout my graduate career. He not only contributed significantly to the establishment, maintenance, and care of my ash tree plot in Bowling Green, OH, but he became a friend and colleague in the process.

I would like to thank Jeff Lehman from Otterbein College for his support, mentorship, and friendship since we first met in 2001. He provided me with a direction in life, taught me what it takes to be successful in science, and most importantly made me believe in myself. Finally, I want to thank my family. My paternal grandmother,

Marguerite Whitehill is a strong woman that is a true inspiration. I also want to thank my grandparents that are no longer with us (Helen Heeger, Jerome Heeger, and Clarence

Whitehill). My father, Frederick Whitehill has always provided me with love and support, pushed me to pursue excellence and to never put forth anything other than my best, and my mother Barbara Whitehill has always given me nothing but unconditional love and support and always pushed me to think outside the box. Thank you for everything and this dissertation and degree is a testament to you both.

vii

VITA

1982 ...... Born - Bellefontaine, OH

2001 ...... Graduated Bellefontaine High School

2006 ...... B.S. Life Science, Otterbein College,

Westerville, OH

2006 ...... B.S. General Music, Otterbein College,

Westerville, OH

2006 to present ...... Graduate Research Associate, Department

of Plant Pathology, The Ohio State

University

2009 ...... M.S. Plant Pathology, The Ohio State

University

2009 to present ...... Ph. D. Candidate Plant Pathology, The Ohio

State University

viii

PUBLICATIONS

Cipollini D., Wang Q., Whitehill J.G.A., Powell J., Bonello P., and Herms D.A. 2011. Distinguishing Defensive Characteristics in the Phloem of Ash Species Resistant and Susceptible to Emerald Ash Borer. Journal of Chemical Ecology: 37: 450-459.

Chen Y., Whitehill J.G.A., Bonello P., and Poland T. 2011. Differential Response in Foliar Chemistry of Three Ash Species to Emerald Ash Borer Adult Feeding. Journal of Chemical Ecology 37: 29-39.

Whitehill J.G.A., Lehman J.S., and Bonello P. 2007. Ips pini (Curculionidae: Scolytinae) is a Vector of the Fungal Pathogen, Sphaeropsis sapinea (Coelomycetes), to Austrian Pines, Pinus nigra (Pinaceae). Environmental Entomology 36: 114-120.

FIELDS OF STUDY

Major Field: Plant Pathology

Specialization: Plant Molecular Biology and Biotechnology

TABLE OF CONTENTS

Abstract ...... ii! Dedication ...... iv! Acknowledgments ...... v! Vita ...... viii! Publications ...... ix! Fields of Study ...... ix! Table of Contents ...... x! List of Tables ...... xiv! List of Figures ...... xvi! Chapter 1 ...... 1! LITERATURE REVIEW ...... 1! INTRODUCTION ...... 1! POTENTIAL AND REALIZED IMPACTS OF THE EMERALD ASH BORER INVASION ...... 4! ECOLOGY OF EAB AND ASH ...... 9! PHYLOGENY AND PHYLOGEOGRAPHY OF THE GENUS FRAXINUS ...... 11! PLANT DEFENSE AGAINST HERBIVORES ...... 16! Phenolics ...... 18! Lignin ...... 20! Defensive Proteins ...... 21! MECHANISMS OF RESISTANCE TO WOODBORERS ...... 22! HYPOTHESES AND OBJECTIVES OF MY WORK ...... 23! STRATEGIES AND TOOLS USED TO STUDY DEFENSE RESPONSES IN ASH ...... 24! Girdling ...... 26! The Use of Phytohormones ...... 27! Characterization of Insect Oral Secretions ...... 29! Proteomics and Metabolomics ...... 29! Chapter 2 ...... 32! Species-Specific Effects of Girdling, Insecticide Application, and Methyl Jasmonate Priming on EAB Colonization of Ash ...... 32! ABSTRACT ...... 32! x

INTRODUCTION ...... 33! MATERIALS AND METHODS ...... 37! Experimental Design ...... 37! Sampling ...... 39! Statistical Analysis ...... 40! RESULTS ...... 40! EAB Colonization of Treated Ash Species within the RCB Design ...... 40! EAB Colonization of Treated Black Ash ...... 41! DISCUSSION ...... 41! ACKNOWLEDGEMENTS ...... 46! Chapter 3 ...... 47! Interspecific Proteomic Comparisons Reveal Ash Phloem Genes Potentially Involved in Constitutive Resistance to the Emerald Ash Borer ...... 47! ABSTRACT ...... 47! INTRODUCTION ...... 49! MATERIALS AND METHODS ...... 52! Experimental Design ...... 52! Protein Extraction and Purification ...... 52! 1-D SDS-PAGE ...... 53! Sample preparation for DIGE ...... 54! IPG Strip pI Range and First Dimension Isoelectric Focusing ...... 56! Strip Equilibration and Second Dimension ...... 59! Gel Imaging and Statistical Analysis ...... 59! Preparative Gel, Protein Spot-cutting, and Trypsin Digestion ...... 62! Nano LC-MS/MS, Protein Identification, and Data Deposition ...... 63! Protein Gene Ontology Annotation and Identification of Putative Defense and Increased Susceptibility Genes ...... 65! RESULTS ...... 67! Principal Component Analysis ...... 67! Proteomic Differences between Manchurian and Black, Green, and White Ash . 70! Protein Identification, Classification, and Gene Ontology Annotation ...... 71! Identification of Putative Defense and Susceptibility-Related Genes ...... 72! DISCUSSION ...... 73! Proteomic Analysis of Non-Model Plants and Inter-species Comparisons Using DIGE ...... 73! Putative Resistance-Related Genes in Manchurian Ash ...... 76! Major Allergen ...... 76! Phenylcoumaran Benzylic Ether Reductase ...... 77! Aspartic Protease ...... 78! Thylakoid-Bound Ascorbate Peroxidase ...... 79! Conclusions ...... 80! ACKNOWLEDGEMENTS ...... 82! Chapter 4 ...... 93! xi

Interspecific Comparisons of Ash Constitutive Phloem Phenolic Chemistry ...... 93! ABSTRACT ...... 93! INTRODUCTION ...... 95! MATERIALS AND METHODS ...... 97! Experimental Design and Sampling ...... 97! Phenolic and Lignin Extraction ...... 98! Selection of Phenolic Compounds Using HPLC-UV for Qualitative, Quantitative, and Statistical Analyses ...... 99! Analysis and Identification of Soluble Phenolics with HPLC-ESI-MS-PDA .... 100! Identification and Quantification of Phenolics ...... 101! Statistical Analyses ...... 103! RESULTS ...... 104! Selection of Phenolic Compounds Using HPLC-UV for Qualitative, Quantitative, and Statistical Analyses ...... 104! Simple Phenolics and Phenolic acids ...... 105! Coumarins ...... 106! Lignans ...... 108! Monolignols ...... 109! Secoiridoids ...... 109! Phenylethanoids ...... 111! Flavonoids ...... 112! Coumarins-secoiridoids ...... 112! Unknowns ...... 113! Qualitative and Quantitative Differences in Phenolic Profiles ...... 113! Principal Component, Cluster, and Biplot Analyses ...... 115! Lignin ...... 122! DISCUSSION ...... 122! ACKNOWLEDGEMENTS ...... 127! Chapter 5 ...... 147! Effects of Methyl Jasmonate on Induced Responses of Manchurian and White Ash to Emerald Ash Borer Elicitors and a Necrotrophic Pathogen ...... 147! ABSTRACT ...... 147! INTRODUCTION ...... 149! MATERIALS AND METHODS ...... 151! Experimental Design ...... 151! Larval Homogenate Preparation ...... 156! Isolation of Botryosphaeria ...... 157! Analysis of Phenolics and Lignin ...... 157! Statistical Analyses ...... 158! RESULTS ...... 158! Identification of Phenolics ...... 158! Effect of MeJA Treatment, Species, and Subtreatment on Individual Phenolics and Lignin ...... 164! xii

DISCUSSION ...... 164! ACKNOWLEDGEMENTS ...... 168! Chapter 6 ...... 173! Characterization of 4th Instar EAB Larval Oral Secretions: A Molecular Window into EAB – Ash Interactions ...... 173! ABSTRACT ...... 173! INTRODUCTION ...... 175! MATERIALS AND METHODS ...... 176! Collection of EAB Oral Secretion ...... 176! Extraction of Proteins and Metabolites ...... 178! Protein Extraction and Preparation for Shotgun Proteomics ...... 179! Shotgun Proteomic Analysis ...... 180! Protein Gene Ontology Annotation ...... 182! Metabolomics ...... 182! RESULTS ...... 184! Shotgun Proteomics ...... 184! Metabolomics ...... 184! DISCUSSION ...... 185! ACKNOWLEDGEMENTS ...... 193! Chapter 7 ...... 201! Conclusions and Suggestions for Future Work ...... 201! BIBLIOGRAPHY ...... 213!

xiii

LIST OF TABLES

Table 3.1. Details of DIGE in-gel comparisons: M = Manchurian, G = green, W = white, B = black ash. Numbers associated with each species represent biological replicates...... 61! Table 3.2. Number of proteins differentially expressed between North American ashes and Manchurian ash. Differences between species reflect phylogenetic relatedness (Figure 3.1)...... 69! Table 3.3. Phloem proteins putatively related to defense identified in Manchurian ash. 83! Table 3.4. Phloem proteins putatively related to susceptibility identified in black, green, and white ash...... 84! Table 3.5. Proteins identified by MS and MASCOT analysis from Manchurian ash with an average ratio of 2 or greater when compared to black, green, and white ash...... 85! Table 3.6. Proteins identified by MS and MASCOT analysis from black, green, and white ash with an average ratio of 2 or greater when compared to Manchurian ash. . 90! Table 4.1. Phenolic compounds identified by HPLC-UV and LC-MS-PDA from samples collected on June 2nd, 2008 from susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash grown in a common garden in Bowling Green, OH...... 128! Table 4.2. Phenolic compounds identified by HPLC-UV and LC-MS-PDA from samples collected on August 8th, 2008 from susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash grown in a common garden in Bowling Green, OH...... 134! Table 4.3. Contents of individual phenolic compounds and lignin in susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash in samples collected on June 2nd, 2008. Compounds are separated into groups by phenolic compound class, while species are separated into the sections to which they belong within the genus Fraxinus (Wallander 2008). Contents are expressed in mg g- 1 FW ± SEM (N = 8). Different letters within a row indicate significantly different means by the protected LSD test (! = 0.05). A solid black line separates black ash data from the other species because it was not part of the original experimental design, but can be visually compared to the other species...... 139! Table 4.4. Contents of individual phenolic compounds and lignin in susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash in samples collected on June 2nd, 2008. Compounds are separated into groups by phenolic compound class, while species are separated into the sections to which they xiv

belong within the genus Fraxinus (Wallander 2008). Contents are expressed in mg g- 1 FW ± SEM (N = 8). Different letters within a row indicate significantly different means by the protected LSD test (! = 0.05). A solid black line separates black ash data from the other species because it was not part of the original experimental design, but can be visually compared to the other species...... 143! Table 5.1. Characteristics of HPLC peaks analyzed with a photo diode array detector (PDA) and putative identifications based on retention time and UV spectrum based on identities presented in Chapter 4...... 169! 2 Table 5.2. " values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of species, treatment, and subtreatment effects on HPLC peaks shared between F. americana and F. mandshurica...... 170! 2 Table 5.3. " values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of treatment and subtreatment effects on individual HPLC peaks in F. mandshurica...... 171! 2 Table 5.4. " values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of treatment and subtreatment effects on individual HPLC peaks in F. americana...... 172! Table 6.1. Plant proteins identified by MS and MASCOT analysis in EAB oral secretions with a Mascot score of 60 or greater...... 196! Table 6.2. Bacterial proteins identified by MS and MASCOT analysis in EAB oral secretions with a Mascot score of 60 or greater...... 198! Table 6.3. Metabolites identified by LC-MS-PDA from EAB oral secretions of 4th instar larvae feeding on susceptible green ash in the field. Numbers in bold text represent the dominant ion on which subsequent fragmentation patterns are based...... 200!

LIST OF FIGURES

Figure 1.1. Phylogenetic tree of the genus Fraxinus. Species used in this study are outlined. Reproduced with permission from Eva Wallander and Springer publishing...... 13! Figure 2.1. Effects of species and treatment on EAB attack rates (number of emergence holes per trees). Species (F3,127 = 13.526; P < 0.001) and treatment (F3,127 = 17.349; P < 0.001) had significant effects. The insecticide and MeJA treatments had statistically similar effects, but were both different from the other two treatments (protected LSD test, P < 0.05). The species x treatment interaction was also significant (F3,127 = 2.889; P < 0.01). Treatment effects for black ash were also significant (F3,31 = 13.864; P < 0.001) similar to those observed in the other species, including the statistical equivalence of the MeJA and insecticide treatments. Error bars are the standard error of the mean (s.e.m.) and different letters indicate significantly different means separated by the protected LSD test (! = 0.05)...... 42! Figure 3.1. A 1-D SDS-PAGE gel of protein extracts (20 µg per lane) from Manchurian, black, green, and white ash phloem tissues showing the high quality of the extracts. Protein extracts are pools from eight biological replicates...... 55! Figure 3.2. A 2-D SDS-PAGE gel (pI 3-10) of a pooled protein extract consisting of equal parts derived from 32 individual biological replicates (n = 8 each for Manchurian, black, green, and white ash). Most of the proteins are found in the 4-7 pI range, which was subsequently used in all DIGE analyses...... 57! Figure 3.3. 2-D DIGE gel overlay. The gel image represents proteins (pI 4-7) derived from Manchurian ash (Cy3– red spots) and black ash (Cy5– green). The internal standard (composed of equal parts from all four ash protein extracts) is visible in this gel as Cy2- blue spots. Yellow indicates non-differentially expressed, overlapping spots common to all species. Numbered protein spots are associated with resistant Manchurian ash and susceptible black, green, and white ash (see Tables 3.3 and 3.4)...... 58! Figure 3.4. PCA analysis of ash phloem proteomic profiles. PC1 clearly separates black (!) and Manchurian (!) ash from white (!) and green (!) ash, while PC2 separates Manchurian and black ash but not white and green ash. Each color point represents a single biological replicate (n=8/species)...... 68!

xvi

Figure 3.5. Bar graphs of mean log standardized abundance (LSA) values for proteins shown in Tables 2 and 3. Panels A, B, C, D: Proteins expressed at higher levels (> 5- fold; P < 0.01) in resistant Manchurian (M) ash than in susceptible North American black (B), white (W), and green (G) ash. Panels E and F: Proteins expressed at higher levels (> 5-fold; P < 0.01) in susceptible black, green, and white ash than resistant Manchurian ash. Dots are individual biological replicates within genotype/species, while crosses are the means. The zero value for LSA corresponds to the internal standard (IS). N = 8, except in Manchurian ash PR-10, where n = 7 due to a lack of a match for this protein in gel 1 (Table 3.1) to the master gel image...... 74! Figure 4.1. PCA plot of phenolic data in six species of taxa. Numbers represent individual ash species: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon...... 116! Figure 4.2. Approximately unbiased (AU) confidence value test from bootstrapping for clusters obtained using the first 2 PC scores from the six ash taxa. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group...... 117! Figure 4.3. AU confidence value test for clusters obtained using the first 2 PC scores from metabolites. Numbers at the end of lines prefaced with A correspond to individual phenolic metabolites as listed in Table 4.1 and 4.2. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group...... 118! Figure 4.4. Relationship between PC1 and PC2 scores obtained for six species of ash and 68 metabolites. Black dots with black numbers refer to individual ash taxa:1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon. Red arrows ending with an ‘A’ followed by a number correspond to individual phenolic metabolites as listed in Table 4.1 and 4.2. The biplot analysis visualizes the compounds that are the main drivers of the differentiation of the various taxa and clusters...... 119! Figure 4.5. Relationship between PC1 and PC2 scores for individual peak areas for 61 compounds (excluding the most characteristic compounds that were associated with individual species – see Figure 4.2 and 4.4) identified in six species of ash. Numbers correspond to individual ash taxa: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon. . 120! Figure 4.6. Approximately unbiased (AU) confidence value test from bootstrapping for clusters obtained using the first two PC scores from the six ash species based on 61 phenolic metabolites. Numbers at the end of a line correspond to individual ash species: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group...... 121! Figure 5.1. MeJA or water were applied to a 20 cm branch segment (between orange tape markers), while the sub-treatment was applied within a 1 cm2 bark window positioned on the upper side of the branch, 5 cm from the stem-branch junction. ... 153! xvii

Figure 5.2. Diagrammatic representation of artificial wound created on ash branch where EAB larval homogenate was applied and phloem tissue collected around wound for chemical analyses...... 154! Figure 5.3. Environmental data were collected daily throughout the course of the experiment in both the MeJA and H2O control experimental blocks. Data points represent the mean (± SEM) of three separate measurements taken at 10 am, 12 pm, and 2 pm. While on some days there might have been significant differences between the two blocks (e.g. on day 8 for temperature and days 3, 8, and 10 for humidity), the two blocks did not differ overall (ANOVA)...... 155! Figure 5.4. Overlay of HPLC chromatograms showing effects of MeJA in the absence of a subtreatment: black trace - Manchurian ash treated with MeJA; magenta trace - Manchurian ash treated with H2O; blue trace - white ash treated with MeJA; orange trace - white ash treated with H2O. Peaks correspond to identities presented in Table 5.1 ...... 159! Figure 5.5. Concentrations of individual compounds shared between Manchurian and white ash presented as absorbance units µl-1 extract. Bars are means of individual peak areas (± SEM) for each compound (n = 4 for Manchurian ash and n = 5 for white ash where each n is composed of information from three separate subsamples). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes...... 160! Figure 5.6. Concentrations of individual compounds specific to Manchurian ash presented as absorbance units µl-1 extract. Bars are means of individual peak areas (± SEM) for each compound (n = 4 for Manchurian ash where each n is composed of information from three separate subsamples). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes...... 161! Figure 5.7. Concentrations of individual compounds specific to white ash. Bars are means of individual peak areas (± SEM) for each compound (n = 5 for white ash where each n is composed of information from three separate subsamples). EAD 1 = elenolic acid derivative 1 and ORC 1 = oleuropein related compound 1. Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes...... 162! Figure 5.8. Lignin concentration (± SEM) in the phloem of Manchurian and white ash (n = 4 for Manchurian ash, and n = 5 for white ash). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal- Wallis test. Means are used in place of medians for presentation purposes...... 163! Figure 6.1. A 4th instar EAB larva that has just been removed from feeding on a green ash tree in the field. The liquid at the insect mouth is characteristic of actively feeding EAB larvae that are removed from trees in the field. The oral secretion droplet was collected and used in our proteomic and metabolomic analyses...... 177!

xviii

Figure 6.2. GO annotation term relationships obtained from the QuickGO resource (Binns et al. 2009) for biological functions of 15 terms associated with proteins identified as plant proteins in EAB oral secretions. Organization of proteins in Table 6.2 are based on these relationships, with the parent category serving as a point of organization for the daughter terms...... 194! Figure 6.3. GO annotation term relationships obtained from the QuickGO resource (Binns et al. 2009) for biological functions of 9 terms associated with proteins identified as bacterial proteins in EAB oral secretions. Organization of proteins in Table 6.2 are based on these relationships, with the parent category serving as a point of organization for the daughter terms...... 195! Figure 7.1. Black circles show the entrance holes of EAB neonates in an artificial inoculation experiment using eggs inserted under a flap of outer bark. Neonates immediately died upon initiation of feeding on drought stressed Manchurian ash trees...... 206! Figure 7.2. Close-up of the posterior end of a dead EAB neonate that fed on a drought stressed Manchurian ash tree. The arrow is pointing directly at the dead EAB neonate...... 207! Figure 7.3. The gall-like structure on the main stem of this Manchurian ash tree is the result of a wound periderm forming over an EAB larval gallery...... 208! Figure 7.4. Anatomy of a Manchurian ash wound periderm. The numbers 1 – 3 are used to denote the innermost layer (1) expanding out to the outermost layer of the wound periderm (3). (1) Is the end of an EAB larval gallery. Notice the thick callus tissue that formed at the end of the gallery and surrounding the gallery. The EAB gallery itself was initially created at the cambial-phloem interface, but callus tissue completely grew over and covered the initial EAB gallery (2). On top of the original gallery and callus tissue the phloem layer (3) continued to grow seemingly uninterrupted by the damage caused by the EAB larvae...... 210!

xix

CHAPTER 1

LITERATURE REVIEW

INTRODUCTION

Species are considered invasive when they have the ability to establish populations in new areas and then spread uncontrollably, causing significant environmental and economic problems (Driesche and Driesche 2000). Some of the most devastating invasive species are alien forests pests that are constantly being introduced into new geographical locations because of increasing international commerce (Niemela and

Mattson 1996, Aukema et al. 2010). Invasive forests pests, such as the emerald ash borer

[(EAB), Agrilus planipennis Fairmaire (Coleoptera: Buprestidae)], have caused billions of dollars in damage and will continue to do so until suitable hosts are no longer available, or effective widespread management and containment of the pest become viable (Sydnor et al. 2007).

All of the major North American species of ash (Fraxinus spp.) that EAB has encountered are colonized and killed (Cappaert 2005, Poland and McCullough 2006), resulting in millions of dead ash trees in parts of Michigan and Ohio (Herms et al. 2004,

Cappaert 2005). It is estimated that 8 billion ash trees exist in the U.S., with a standing

value of more than $300 billion (Poland and McCullough 2006, Sydnor et al. 2007).

Thus, the impact from a purely economic standpoint is potentially highly significant, but the functional extirpation of the Fraxinus genus from North America would be equally devastating to the ecological health of urban and natural environments (Gandhi and

Herms 2010a).

EAB was first discovered in May 2002 infesting ash trees in various parts of

Southeastern Michigan (Haack 2002). Approximately 15 million ash trees were either dead or dying in 2004 in southeastern Michigan as a result of EAB attacks (Agius 2005).

EAB spreads an average of 20 km/year, but isolated infestations have been found in

Indiana, Iowa, Maryland, Minnesota, Missouri, New York, Ohio, Tennessee, Wisconsin, and parts of Canada (Prasad et al. 2010, emerald ash borer.info). Humans act as long- distance dispersal vectors by moving infested firewood or nursery stock, with two-thirds of EAB outbreaks occurring around major highways (McCullough 2004, Iverson 2006,

Petrice 2006).

EAB is a native of northeastern China, Korea, Japan, and parts of Russia and

Mongolia (Haack 2002). Its native host range is limited to Fraxinus chinensis subsp. chinensis, F. chinensis subsp. rhynchophylla, and F. mandshurica, with F. chinensis subsp. chinensis being the most susceptible to attack by EAB (Chinese Academy of

Sciences 1986, Yu 1992). The life cycle of EAB in its native range begins as an egg laid by a female, 7-10 d after initial mating, on a suitable host’s outer bark surface, from mid-

June to mid-July. Eggs are mostly laid in bark crevices with access to sunlight and at the base of the trunk. Larvae emerge 7-9 d after oviposition, begin to feed on the phellem

(cork) layer of the bark, and eventually bore into the phloem and cambium where they

feed and overwinter as larvae. Larval feeding in the phloem and outer sapwood destroys the hosts’ vascular system and leads to tree mortality. Pupation occurs in late April, peaks in mid-May, and ends in mid-June. Adults emerge from mid-May through mid-to- late July. Eggs begin to appear during this period, at which point the cycle repeats. One generation a year is typical, but some individuals may require two years (Chinese

Academy of Sciences 1986, Yu 1992).

While EAB colonizes healthy and stressed North American ash species, in its native range it functions primarily as a secondary invader (Liu et al. 2007). This is supported by the experimental finding that healthy Manchurian ash is largely resistant to attack by

EAB (Rebek et al. 2008). It is hypothesized that Manchurian ash resists EAB colonization by deploying targeted defenses that developed over time by virtue of its co- evolutionary history with the insect (Rebek et al. 2008). However, only a few studies have thus far focused on potential mechanisms of Manchurian ash resistance to EAB

(Cipollini et al. 2011, Eyles et al. 2007). Therefore, the goal of this dissertation research was to identify mechanisms of ash resistance to EAB. We investigated constitutive and induced mechanisms of resistance in ash using proteomic and metabolomics approaches to study physiologically altered trees belonging to resistant and susceptible species from diverse phylogenetic backgrounds.

POTENTIAL AND REALIZED IMPACTS OF THE EMERALD ASH BORER

INVASION

EAB has the potential to eliminate ash (Fraxinus spp.) from Ohio and all North

American forests (Herms 2004, Gandhi and Herms 2010a). Ash species are ubiquitous throughout much of the eastern half of the United States and play crucial roles in the economy of Ohio, other U.S. states, and Canada. Ash trees are integral parts of the natural and seminatural landscape, where they serve a wide range of ecological roles, and the urban landscape, where they serve aesthetic and functional purposes. To date, all

North American species of ash tested are susceptible to attack by EAB (Herms et al. unpublished;(Cappaert 2005, Rebek et al. 2008). As Herms et al. (2004) stated, “The

[emerald ash borer] threat cannot be overestimated.” This statement from 2004 has been realized many times over in the following years.

The initial infestation core of EAB was discovered in southeastern Michigan and

Ontario, Canada in the summer of 2002 (Haack 2002). The first infestation in Ohio was found in Lucas County in February 2003, followed by four more infestations in Defiance,

Paulding, Wood, and Franklin counties, respectively, by the end of summer 2004. The first satellite population of EAB was confirmed in September 2003 in Maryland, near

Washington, D.C. Satellite populations are the result of the movement of infested ash logs, firewood, and nursery stock to uninfested areas. The artificial spread of infested ash material by human vectors is greatly expediting the expansion of EAB (Herms et al.

2004). Since 2004 and the initial discoveries of those EAB populations, there has been a rapid spread of EAB to surrounding areas, with many subsequent discoveries of satellite

populations in the U.S. and Canada (Herms et al. 2004, emeraldashborer.info). Recent models, based on the currently known location of EAB populations, predict that, at the rate of natural EAB expansion, coupled with the artificial rate of expansion resulting from human vectoring, EAB will spread throughout the whole range of ash in the eastern half of the U.S. by 2019 (Kovacs et al. 2010).

To date, EAB has been found in 15 U.S. states and throughout Ontario and Quebec,

Canada (emeraldashborer.info). In the U.S., an area with a confirmed population of EAB is quarantined immediately. Any ash material within infested zones under quarantine cannot be moved to areas outside the quarantined zones (Agriculture 2006). Anyone found to be in violation of this law is subject to a fine up to $250,000 and/or imprisonment. This law is stated under the Plant Protection act of 2000 and is enforced by the USDA’s Animal and Plant Health Inspection Service’s (APHIS) Plant Protection and Quarantine (PPQ) program (Agriculture 2006). On September 8th, 2010 all of Ohio’s

88 counties were placed under quarantine (Agriculture 2010). The potential loss of economic activity as a result of these imposed quarantines is estimated to be $200 million annually (ODNR 2005).

An estimated 3.8 billion white ash (F. americana L.) trees, which include small saplings all the way to mature trees, were present in Ohio’s forests alone in 2005, with green ash (F. pennsylvanica Marshall) being almost as abundant (ODNR 2005). Overall, ash trees make up 10% of the total tree population in Eastern North American forests

(ODNR 2005). An analysis of data collected through the end of 2007, in just the primary infestation zones of southeastern Michigan, northeast Indiana, and northwest Ohio, estimated that 53.6 million ash trees (diameter at breast height - d.b.h. ! 5 cm)

wereinfested with EAB (including forest and urban sites) (Smith et al. 2009), while ash mortality in 2004 was estimated to be between 8 and 10 million ash trees in Ohio alone

(Herms 2004). Mortality of ash is assumed to expand linearly once a stand becomes infested, i.e. mortality within an infested stand will increase from 0-100% in a predictable manner (~5-6 years) (Knight and Long 2008, Kovacs et al. 2010).

Ash wood is used commercially in a number of different ways. White ash wood is the most valuable of all the ash species and is valued for its strength, hardness, heavy weight, and shock resistance. White ash wood is renowned for its use in baseball bats, but it is also used for furniture, doors, veneer, antique vehicle parts, railroad cars and ties, canoe paddles, snowshoes, boats, posts, ties, and fuel. Native Americans used it for tools and other implements, and it is still used for tool handles commercially today (Nesom

2000).

Another eastern North American ash species, black ash (F. nigra Marshall), is used by native North Americans to weave baskets. Basket making is a traditional Native art that, in the past, was a tool necessary for survival. Baskets were made to serve a multitude of purposes and because of their lightweight and portability were the perfect containers for gathering, harvesting, storing, cooking, and serving food. As a result of the availability of inexpensive kitchen and storage containers, baskets are now mainly used in Native American communities for ceremonies, decoration, or gifts. Seven native

American nation basket makers have long used ash not only for ornamental baskets, but for weaving lightweight and practical backpacks used by many hikers. Baskets are popular items commonly sold by native craftspeople from upstate New York and New

England as a means of generating income as well as a way of preserving traditions from their past (Giese 1996).

EAB is not only affecting the cultural and commercial use of ash, but it is also impacting nurseries and cities. In 2003 nursery growers in Ohio estimated that ashes contributed a wholesale value of $20 million to Ohio's economy each year. These costs do not take into account the additional money generated by retail garden centers and landscape contractors (ODA 2005). Ash trees are important components of city parks and have been widely planted as street trees. The average landscape value for park and street trees is estimated at $672 – $807 for 30 – 31 cm (d.b.h.) trees, respectively (Sydnor et al. 2007). Overall, costs incurred in the state of Ohio as a result of the EAB invasion in terms of landscape value of trees, removal and replacement costs, will equal or exceed

$7.5 billion dollars (Sydnor et al. 2007).

If EAB continues its spread through Ohio, private homeowners will be responsible for the removal of dead ash trees on their property. State foresters estimate that there are

3.4 million ash trees growing on private urban property statewide. In a 2005 study, the total cost to Ohio private property owners of removing EAB-killed ash trees was estimated to be approximately $1.37 billion (based on a conservative average removal cost of $400/tree) (ODNR 2005). More recent estimates of tree removal costs to homeowners ranged between $200 (2.5 – 30 cm d.b.h.) to $1,100 (d.b.h. > 61 cm) per tree (Kovacs et al. 2010). The result is that a total cost of $10.7 billion dollars in all areas with EAB infestations is projected for private homeowners between 2009 and 2019

(Kovacs et al. 2010).

Ashes are also very important functional components of eastern North American forests and EAB poses a multitude of direct and indirect threats to ecological processes and interactions (Gandhi and Herms 2010a, Gandhi and Herms 2010b). Ash species are important contributors to nutrient cycling in hardwood forests, a variety of animals feed on seed produced by ash trees, and in general serve a multitude of purposes for bird and mammal species (Reiners and Reiners 1970, Cappaert 2005, Dodds and Otis 2009). An in depth review of the literature focusing on North American arthropods that rely on

Fraxinus spp. for survival in some capacity identified 282 documented arthropod species that are associated with ashes (Gandhi and Herms 2010a). The study also identified 43 arthropod species that are completely dependent on ash for survival (Gandhi and Herms

2010a). The survival of these species is now threatened by the continued spread of EAB.

Previous widespread mortality events of important forest trees can serve as models for what to expect regarding EAB-induced mortality of ash. For instance, the widespread mortality of the American chestnut (Castanea dentata (Marsh) Borkh.), caused by the chestnut blight pathogen (Cryphonectria parasitica (Murril.) Barr.) has shown that several species that were once found in close association with chestnut can now be considered regionally extinct (Centre 1996).

In summary, the EAB invasion is already having, and will continue to have, devastating and lasting effects on the economy of afflicted regions; the cultural practices of Native Americans; the aesthetics, structure, and function of the urban landscape; and fundamental ecological processes, and interactions of ash trees with other species, in

North American forests.

ECOLOGY OF EAB AND ASH

Indigenous wood-borers (family Buprestidae and Cerambycidae) and bark beetles

(family Curculionidae: Scolytinae) persist at endemic levels by colonizing dead or dying trees, and outbreaks of these pests have been associated with specific host stress inducing events, such as prolonged drought (Wood 1982, Speight 1989, Anderson 1944, Barter

1957, Hines and Heikkenen 1977, Dunn 1986, Koricheva et al. 1998). This appears to be largely the case for EAB in its native range of China, where EAB mainly colonizes the more susceptible Asian species F. chinensis subsp. chinensis and F. chinensis subsp. rhynchophylla when growing in the open or along the margins of a forest (Chinese

Academy of Sciences 1986). However, severe infestations of these species can lead to the devastation of a whole forest stand (Yu 1992).

In other words, in its native range EAB appears to colonize only trees that are stressed

(e.g. drought, girdling, etc.), similar to the dynamics of native North American wood- borers and their host species (Anderson 1944, Gould et al. 2005, Liu et al. 2007,

Muilenburg 2010). However, colonization of all native ash species in the vicinity of colonized trees will eventually occur, depending on population densities (Poland and

McCullough 2006). The speed of EAB development is also determined in part by host vigor, with a more rapid life cycle in stressed trees compared to unstressed trees (Siegart

2006).

On the other hand, EAB has the ability to colonize and kill healthy North American trees on high-quality sites (Haack 2002, Herms 2004), while Manchurian ash seems to be resistant (Rebek et al. 2008). This pattern seems to fit the hypothesis that variation in

plant resistance is linked to co-evolutionary relationships between herbivores and their hosts within a geographic region (Bryant 1994, Swihart 1994, Rebek et al. 2008).

All North American species of ash that EAB has come into contact with have succumbed to infestation and eventual death (Cappaert 2005, Poland and McCullough

2006). However, evidence is mounting that EAB is preferentially attracted to certain

North American Fraxinus species over others (although EAB is still able to complete its lifecycle within all North American species successfully) (Anulewicz et al. 2006,

Anulewicz et al. 2008). The main North American species on which these studies have focused include the susceptible species, green ash (F. pennsylvanica), white ash (F. americana), black ash (F. nigra), and blue ash (F. quadrangulata Michaux). It appears that healthy green ash is the most attractive and heavily infested ash species when compared alongside other healthy ash species (Anulewicz et al. 2006, Anulewicz et al.

2008, Pureswaran and Poland 2009). Healthy white and black ash rank next, while healthy blue ash is the least preferred by EAB among all North American species. This has caused some to postulate that blue ash may be a source of potential North American resistance/tolerance, even though blue ash is still colonized by EAB in the field

(Anulewicz et al. 2008, Pureswaran and Poland 2009). Black ash appears to be highly susceptible to attack by EAB in the field (Smith et al. 2005).

Recent EAB-induced mortality of European ash (F. excelsior L.) in Moscow, Russia

(Baranchikov et al. 2008) suggests that EAB now poses a major threat to European ash trees in Eurasia and Europe, where they are main components of forest and urban ecosystems (FRAXIGEN 2005, Baranchikov et al. 2008).

PHYLOGENY AND PHYLOGEOGRAPHY OF THE GENUS FRAXINUS

The genus Fraxinus is one of 24 genera within the olive family (Oleaceae). Bentham

(1876) was the first to characterize the Oleaceae by placing 17 genera into four tribes based on morphological characteristics. Knoblauch (1895) further split the Oleaceae into two subfamilies (Oleoideae and Jasminoideae) and placed tribes as described previously within these subfamily classifications based upon reproductive characteristics of individual genera. The next major change was a re-arrangement of the genera within subfamilies based on chromosomal data and fruit morphology (Taylor 1945), followed by a review of the entire family that split several genera into completely different families

(Johnson 1957). The most recent classification of the Oleaceae abandoned completely the idea of subfamily within the Oleaceae, due to the paraphyletic nature of some genera contained with the subfamily Jasminoideae that could not be accurately characterized based on morphological and chromosomal characteristics alone (Wallander and Albert

2000). The reclassification of the Oleaceae was based on rps16 and trnL-F sequence data, and split the paraphyletic subfamily Jasminoidae into 4 tribes (Myxopyreae,

Fontanesieae, Forsythieae, and Jasmineae). The subfamily Oleoideae (Knoblach 1895) was abandoned and the monophyletic group contained within that subfamily was classified as the tribe Oleeae in which four subtribes were identified (Ligustrinae,

Schreberinae status novus, Oleinae, and Fraxininae status novus) (Johnson 1957,

Wallander and Albert 2000). The new classification of the Oleaceae was further supported by a review of the family based on the literature as it relates to the biosynthesis of iridoids and the distribution of caffeoyl phenylethanoid glycosides, in which

chemotaxonomic relationships were established and reinforced the classification of

(Wallander and Albert 2000, Jensen et al. 2002). The subtribe Fraxininae contains only the genus Fraxinus and encompasses a total of 43 species (Wallander and Albert 2000,

FRAXIGEN 2005). Species within the genus Fraxinus occur in North and Central

America (19 species) and also Europe, North Africa, and Asia (24 species) (FRAXIGEN

2005).

The genus Fraxinus was classified several times over the past century and has been divided into sub-genera, sections, and subsections (Lingelsheim 1920, Vassiljev 1952,

Nikolaev 1981, Wallander 2008). The most recent classification of the genus, and the classification this dissertation uses to understand and interpret proteomic and phytochemical data, has divided the genus into six tribes, and abandoned the previous classification schemes. Phylogenetic relationships were estimated on the basis of 106 nuclear ribosomal ITS sequences, supported by an independent set of two chloroplast regions representing 40 of the 43 recognized species (Wallander 2008). The three species not represented in the above analysis were categorized based on data generated from a previous molecular investigation of the genus (Jeandroz et al. 1997). Wallander (2008) identified several discrepancies in previous phylogenetic trees for the genus, and used previous classification schemes in congruence with in depth molecular data, supported by morphological characters for individual species, to develop a new phylogenetic classification. According to Wallander 2008 the genus Fraxinus is divided into six sections: Dipetalae, Fraxinus, Melioides, Ornus, Pauciflorae, and Sciadanthus.

Wallander (2008) contends that each of the six sections is clearly monophyletic and separates well morphologically with each having molecularly distinct lineages.

Sciad- Dipetalae Melioides Pauciflorae Ornus anthus Fraxinus F. nigra F. angustifolia F. platypoda F. mandshurica F. F. trifoliolata F. F. xanthoxyloides F. hupehensis F. F. papillosa F. latifolia F. caroliniana F. texensis F. pennsylvanica F. velutina F. apertisquamifera F. lanuginosa F. bungeana F. F. dubia F. raibocarpa F. paxiana F. F. cuspidata F. spaethiana F. F. longicuspischinensis F. excelsior F. F. profunda F. sieboldiana F. ornus F. floribunda F. micrantha F. F. purpusii F. rufescens F. F. dipetala F. anomala F. chiisanensis F. greggii F. F. uhdei F. F. berlandieriana F. F. quadrangulata F. americana F. gooddingii F.

dioecy no dioecy corolla no corolla no calyx rachis winged

androdioecy terminal infl. corolla rachis hermaphroditism winged

dioecy

geographic distribution pollination system Old World entomophily New World anemophily polygamy no corolla unwinged leaf rachis lateral inflorescences hermaphroditism corolla calyx

Figure 1.1. Phylogenetic tree of the genus Fraxinus. Species used in this study are outlined. Reproduced with permission from Eva Wallander and Springer publishing.

Therefore, the species investigated in this dissertation will be discussed in relation to their phylogenetic origins. Manchurian, black, and European ashes belong to the section

Fraxinus; green and white ashes are members of the Melioides; and blue ash is one of three members belonging to the section Dipetalae (Figure 1.1).

The phylogeography (the historical processes potentially responsible for the current geographical distribution of individuals) of the genus Fraxinus has been investigated and discussed (Jeandroz et al. 1997, Wallander 2008). Jeandroz et al. 1997 hypothesized that, based on the current geographical distribution of the genus, two potential biogeographic histories are possibly, with the more parsimonious scenario likely to be correct – a North

American origin for the genus, as opposed to an Asian origin. The North American origin of the genus is further supported by fossil evidence, and the phylogenetic findings of Wallander 2008 provide further support to this hypothesis (Call and Dilcher 1992).

The hypothesis put forth by Jeandroz et al. 1997 and modified slightly by Wallander

2008 suggests two dispersal events for the genus into Eurasia, and one dispersal event back to North America. The hypothesis suggests that the genus originated from a common ancestor in North America, with the earliest known evidence (~49-39 Myr) of the genus in fossilized samaras from Fraxinus wilcoxiana Berry identified in western

Tennessee (Call and Dilcher 1992).

The progenitor of the genus Fraxinus originated in North America and led to the paraphyletic sections Dipetalae (blue ash), Melioides (green and white ashes), and

Pauciflorae. The first dispersal event from North America to Eurasia occurred during the

Eocene (56-34 Myr), and led to several species treated by Wallander as incertae sedis and the ancestors of the sections Ornus, Fraxinus, and Sciadanthus (Jeandroz et al. 1997,

Wallander 2008). Following separation of the continents the progenitors for Manchurian and European ashes evolved in Europe and Asia as members of the section Fraxinus.

Then, during the 1st half of the Pleistocene (2.6 Myr – 12,000 yr), a dispersal event from

Eurasia back to North America occurred, and led to the introduction of the ancestor of black ash (section Fraxinus) into North America, presumably from a shared common ancestor of Manchurian and European ash. This scenario helps to explain why black ash is the only New World member of the section Fraxinus found in North America

(Wallander 2008).

The information above brings into focus several important facts: 1) the four main ash species in eastern North American forests are geographically close yet represent distinct phylogenetic origins; 2) the differences observed for blue ash when compared to the other

North American ash species can be easily explained when viewed from a phylogenetic perspective; 3) green and white ash, while very closely related to each other, have been evolutionarily isolated from blue and black ash since sometime during the Eocene (56-34

Myr); and 4) the similarities between black and European ash to Manchurian ash are a result of their more recent common ancestor. Having a working knowledge of the evolutionary origins of the genus Fraxinus will lead to more informed conclusions when investigating, selecting, and breeding for specific mechanisms of resistance in ash against

EAB.

PLANT DEFENSE AGAINST HERBIVORES

Plants and insects have co-existed for more than 350 million years. It is estimated that over one million insect species derived from multiple taxonomic groups rely on plants as a food source (Gatehouse 2002, Howe and Jander 2008). Insect feeding on plants has been a driving force behind the diversification of species for both herbivores and hosts (Ehrlich and Raven 1964). The dynamic interactions between plants and insects have not only driven species diversification, but have also driven the evolution of a myriad of specialized defense mechanisms that allow sessile plants to thwart insect attack (Fraenkel 1959, Janzen 1966, Walling 2000).

Plant defenses against insect attack are as diverse as are the counter-adaptations (i.e. detoxification systems) and feeding habits of different species of insects (Dowd et al.

1983, Harborne 1993). Traits that confer resistance to insects in plants can be classified according to how they are regulated, and are characterized as being constitutive (also known as static) and/or induced (active) defenses (Harborne 1993).

Constitutive and induced defenses are classified into two categories – physical and chemical. Physical defenses inhibit insect feeding and oviposition via a physical barrier that affects behavioral aspects of an insect’s ability to successfully utilize the plant for survival (Painter 1951, Gatehouse 2002). Chemical defenses such as secondary metabolites and defensive proteins can decrease the nutritive value of plant tissues, deter, or kill the insect invaders (Feeny 1969, Wittstock and Gershenzon 2002, Howe and

Jander 2008). Secondary metabolites include alkaloids, terpenoids, glucosinolates, cyanogenic glycosides, non-protein amino acids, and phenolics (Levin 1976, Harborne

1993, Bennett and Wallsgrove 1994, Wittstock and Gershenzon 2002, Mayer 2004).

Defensive proteins can be characterized based on the type of direct or indirect effect they exert on insect herbivores. Direct effects of defensive proteins can include: 1) inhibition of insect gut proteases by protease inhibitors that negatively impact growth and development of insects; 2) disruption of the peritrophic membrane of insect pests by plant proteases leading to death of the insect; 3) modification of dietary proteins via oxidative enzymes; and 4) modification of amino acids (primary metabolites) by amino acid- degrading enzymes that limit growth of the insect (Farmer and Ryan 1990, Felton et al.

1992, Pechan et al. 2002, Chen et al. 2005, Mohan et al. 2006). Defensive proteins with an indirect role in plant defense against herbivores include enzymes involved in the production of secondary metabolites (Pichersky and Gang 2000).

Induced responses to herbivore feeding are hypothesized to have evolved because they cost less in terms of resources than constitutive defenses (Karban et al. 1997,

Baldwin 1998, Bonello et al. 2006). The ability of plants to respond to attack from an insect pest constitutes a form of immunity. The current understanding of the mechanisms governing induced plant defenses, as well as the evolutionary basis of the recognition responses in plants, is based mainly on studies of plant-microbe interactions (Howe and

Jander 2008). A recent review of induced defense responses in trees to pests and pathogens has categorized induced mechanisms into five categories: 1) chemical defenses, 2) protein-based defenses, 3) anatomical defenses, 4) ecological or indirect defenses, and 5) inducible civilian defenses (tolerance) (Eyles et al. 2010).

We utilized proteomic and metabolomic techniques to investigate constitutive and induced resistance traits of ash species susceptible and resistant to attack by EAB. In our

approach to elucidate resistance mechanisms, we focused on changes in phenolic chemistry, lignin, and proteins with known or putative roles in defense against herbivores.

Phenolics

Phenolics (phenylpropanoids) are compounds produced in plants via several biochemical pathways, with the shikimate pathway being responsible for the production of a majority of metabolites (Strack 1997). Phenolics are ecologically important molecules because they are considered to be plant defensive molecules against herbivores and pathogens (Feeny 1969, Appel 1993, Witzell and Martin 2008). Phenolic compounds exert their toxic effects on herbivore pests by inhibiting digestion through the denaturation and aggregation of digestive enzymes and proteins in the insect midgut, and by reducing the quality of ingested plant proteins, the main nitrogen source for most insect pests (Feeny 1969, Felton et al. 1992, Appel 1993). In some cases, oxidation, quinylation, and removal of a sugar residue are required before phenolics become biologically active (Felton et al. 1992, Appel 1993).

Phenolic compounds are the dominant class of secondary metabolites in the genus

Fraxinus (Eyles et al. 2007, Kostova and Iossifova 2007). Classes of phenolic compounds previously described from the genus Fraxinus include phenolic acids, simple phenolics, coumarins, lignans, secoiridoids, phenylethanoids, and flavonoids (Kostova

2001, Iossifova et al. 2002, Eyles et al. 2007, Kostova and Iossifova 2007). Sterols, triterpenes, pyrocatechol tannins, and various plant hormones have also been found in

several ash species, although only a limited number of species and tissue types have been investigated (Perez-Castorena et al. 1997, Blake et al. 2002).

The rich diversity of phenolic compounds in Fraxinus species has only recently been studied in relation to potential for deterrence of, or toxicity against, EAB (Eyles et al.

2007, Cipollini et al. 2011). Several compounds identified in Fraxinus spp. have previously been associated with resistance mechanisms against herbivorous pests in related plant species (Eyles et al. 2007). For instance, the compound oleuropein is the main mechanism of defense in privet (Oleaceae) against herbivorous insects, but is only activated following the removal of its sugar moiety with an endogenous #-glucosidase

(Konno et al. 1999). However, the resistant Manchurian ash has low constitutive levels of this compound relative to susceptible ash species, thus reducing its potential importance in ash defense against EAB (Eyles et al. 2007). Conversely, another compound in Fraxinus, pinoresinol, is found in higher quantities in the phloem tissues of the resistant Manchurian ash (Eyles et al. 2007). Pinoresinol has known insecticidal properties in other plant systems, as well as growth inhibition and antifeedant activities, making it and related compounds candidates for further investigation (Cabral et al. 2000,

Garcia et al. 2000). The diversity of phenolic compounds associated with ash species merits further investigations into the role constitutive and induced phenolics play in the

Fraxinus – EAB interaction.

Lignin

Lignin is a heterogeneous polymer of phenylpropanoid subunits that form intimate associations with cell wall polysaccharides (cellulose, hemicellulose), and enable plants to develop into large freestanding structures (Strack 1997). Lignin is typically considered a constitutive, physical defense trait and its role in defense against fungal pathogens has long been recognized (Akai and Fukutomi 1980). In addition to being a constitutive, physical defense trait, lignin is induced upon attack by fungal pathogens in some plant species and contributes significantly to their defense (Vance et al. 1980).

While the effect of lignin against fungal pathogens in planta is rather well understood

(physically inhibiting growth of an invading pathogen), the effectiveness of lignin as a defense against herbivorous insects remains unclear (Brodeur-Campbell et al. 2006).

Lignin can act as an indirect chemical defense following degradation into individual constituents (e.g. dihydroconiferyl alcohol) in bark beetle frass, which act as antifeedants and prevent further colonization of hosts (Borg-Karlson et al. 2006). Lignin also has been observed to function as a dose-dependent physical defense by limiting colonization and reducing larval survival, growth rate, and weight of Dendroctonus micans Kugelann in Norway spruce (Picea abies (L.) Karst) and American Sitka spruce (Picea sitchensis

(Bong.) Carr.) (Wainhouse et al. 1990). Lignin contributes to ‘leaf toughness’, which is often cited as an important factor that affects folivores (Feeny 1970, Coley 1983).

Defensive Proteins

Defensive plant proteins contribute to constitutive defenses against herbivores, although the majority of studies have focused on the induction of defensive proteins by insect feeding (Ryan 1990, van Loon et al. 2006, Shivaji et al. 2010). Defensive proteins are classified based on the mode of action by which they affect insect pests and are classified in terms of anti-nutrition effects and/or toxicity (Chen 2008). Anti-nutritive proteins include protease inhibitors, oxidative enzymes, and amino acid deaminases

(Ryan 1990, Pechan et al. 2002, Chen et al. 2005, Chen 2008). Toxic plant proteins include cysteine proteases, chitinases, lectins, and leucine aminopeptidases (Dowd 1994,

Zhu-Salzman et al. 2008).

Anti-nutritive proteins negatively affect the nutritive value of ingested host material

(Chen 2008). Protease inhibitors produced by plants form complexes with proteases in the insect midgut and inhibit the ability of insects to effectively break down nutritive plant proteins (e.g. rubisco) into their individual components, which serve as the primary source of nitrogen, which is essential for growth and development of the insect (Ryan

1990). This phenomenon was first described as an induced mechanism of defense in potato and tomato leaves in response to insect feeding (Green 1972). Oxidative enzymes such as ascorbate oxidases, lipoxygenases, peroxidases, and polyphenol oxidases interact with plant phenolics to form reactive quinones that are capable of polymerizing or forming covalent adducts with certain chemical groups on proteins (Felton et al. 1992,

Dowd 1994). In addition to their ability to interact with various phenolic compounds, oxidative enzymes may function via several mechanisms that negatively impact herbivore

performance (Zhu-Salzman et al. 2008). Amino acid deaminases, such as arginase and threonine deaminase, are induced in tomato following insect feeding and degrade essential free amino acids in the insect midgut (Chen et al. 2004). Overexpression of arginase in tomato makes plants more resistant to attack by Manduca sexta L. larvae, whereas suppression of threonine deaminase in tobacco increases its susceptibility to M. sexta larvae (Chen et al. 2005, Kang and Baldwin 2006).

Toxic plant proteins affect insect pests by disruption of the midgut lining [peritrophic membrane (PM)] which interferes with digestion and absorption of nutrients, ultimately leading to death of the pest (Zhu-Salzman et al. 2008). Cysteine proteases permeabilize the PM of insect pests, most likely by directly degrading PM proteins (Mohan et al.

2006). Leucine aminopeptidases are thought to function in a similar manner by directly damaging the insect midgut (Felton 2005). Finally, lectins disrupt the PM by changing its morphology (Fitches and Gatehouse 1998), while the role of chitinases in herbivore defense is somewhat questionable, yet its effectiveness as an antiherbivore defense has been observed in some cases (Lawrence and Novak 2006, Zhu-Salzman et al. 2008).

MECHANISMS OF RESISTANCE TO WOODBORERS

The literature contains relatively few studies addressing resistance mechanisms of deciduous trees to buprestid beetles. Several studies have hypothesized that feeding- induced wound responses in phloem tissues are key to resistance (Barter 1957, Dunn

1990). In the case of phloem feeding buprestids, Eyles et al. 2007 proposed that host resistance of angiospermous trees is the result of a combination of constitutive and

induced defenses (Dunn 1990). These include, for example, the presence of the outer bark and pre-existing phenolics in the phloem (constitutive defenses) and the accumulation of phenolics and the formation of necrophylactic periderms (i.e. callus tissue) following an attack (induced defenses), which isolate the wound and inhibit spread of the invading pest (Mullick 1973).

Several specific mechanisms of resistance against wood-boring pests have been proposed in other systems. For example, resistance of Eucalyptus trees against the eucalyptus long-horned borer (Phoracantha semipunctata Fabricius) is the result of high moisture content in phloem tissues (Hanks et al. 1991). In the case of Agrilus bileneatus

Weber and Agrilus anxius Gory the rate of necrophylactic (wound) periderm formation in oak and birch is hypothesized to be the critical defense limiting the spread of actively feeding larvae (Anderson 1944, Dunn 1990). In ash, it has been proposed that differences in phenolic chemistry of North American and Asian species are responsible for differences in resistance (Eyles et al. 2007, Cipollini et al. 2011), although the studies reported in this dissertation suggest that this may not be the case. Based on these observations and the knowledge acquired over the course of the study for this dissertation, I propose that all three components outlined by Eyles et al. 2007 are necessary for ash to mount an effective resistance response against EAB (Chapter 7).

HYPOTHESES AND OBJECTIVES OF MY WORK

The goal of my dissertation research was to characterize mechanisms of resistance in ash. To address this goal I tested the central hypothesis that Manchurian ash, by virtue

of its co-evolutionary history with EAB, resists attack through a combination of constitutive and induced chemical defenses that are targeted against EAB, and which are not present in susceptible North American species of ash because they lack a shared history with EAB. To test the central hypothesis I used a variety of strategies and tools, as summarized below.

My objectives in addressing the central hypothesis were to: 1) compare the rates of

EAB colonization in different ash species that were altered physiologically by either girdling or priming with a defense phytohormone (methyl jasmonate – MeJA) (Chapter

2); 2) characterize constitutive proteomic and metabolomic differences between resistant and susceptible species of ash (Chapter 3 and Chapter 4); 3) describe the effect of exogenously applied MeJA on the phenolic chemistry of a resistant and a susceptible species of ash (Chapter 5); and 4) analyze the chemical composition of the oral secretion of actively feeding late 3rd and early 4th instar EAB larvae and assess its effects on the defense responses of susceptible green ash (Chapter 6).

STRATEGIES AND TOOLS USED TO STUDY DEFENSE RESPONSES IN ASH

To study the interactions of ash and EAB we artificially altered the physiology of host plants in order to quantify defense responses under different conditions. Investigations of healthy trees on high quality sites are not representative of the varying conditions plants endure in nature. While healthy Manchurian ash trees can withstand attack from EAB, in

China colonization of Asian ash species by A. planipennis occurs in stressed trees (Wei et al. 2004, Rebek et al. 2008). Similarly, two congeneric buprestids native to North

America, the bronze birch borer (A. anxius) and the two-lined chestnut borer

(A. bileneatus), only colonize stressed trees with which a co-evolutionary history is shared (Anderson 1944, Dunn 1990). Similarly to EAB, the bronze birch borer colonizes healthy European white birch (Betula pendula Roth.), an introduced species which shares no co-evolutionary history of the pest, quite readily, while colonization of an endemic birch species, paper birch (Betula papyrifera) is relegated to highly stressed individuals

(Miller 1991). Plant defense theory predicts that plants experiencing low levels of available resources will be limited in growth and photosynthetic capability resulting in reduced growth and moderate levels of secondary metabolites (Herms and Mattson

1992). Comparing defensive traits of healthy and physiologically compromised trees has the potential to yield information regarding what defensive characteristics are modified in planta that allow for increased susceptibility of resistant species of ash to EAB.

Stress predisposes trees to attack by weakening the hosts’ endogenous defense mechanisms (Anderson 1944, Barter 1957). We chose to impose an artificial stress on hosts (girdling) to investigate the influence of stress on the interaction between ash and

EAB (Chapter 2). Application of methyl jasmonate (MeJA), a well-known plant hormone that regulates defense responses against herbivoreous pests, was used to determine its potential role in EAB colonization (Chapter 2) and phenolic chemistry

(Chapter 5). Finally, we characterized EAB oral secretions using proteomic and metabolomic approaches to identify potential elicitors that may be involved in the induction of endogenous defense mechanisms in ash (Chapter 6).

Girdling

A tree is girdled when its phloem is completely severed by making a small incision or by removing a large band of bark around the whole tree, without causing damage to the tissue underneath. The technique has been invaluable for understanding basic plant physiology. Photosynthate accumulation above the girdle affects the normal cycle of seed production and flowering (Guinier 1886). Girdled black ash displays an increase in seed production, white ash accumulates carbohydrates immediately above girdled parts, and loblolly pine (Pinus taeda L.) produces double the amount of cones, although only half the number of viable seed, of ungirdled trees (Pond 1936, Zimmerman 1960,

Hansbrough and Merrifield 1963). The over production of cones above the girdle was suggested to be the result of the accumulation of photosynthate above the girdle, while the reduction in viable seed was attributed to a general loss of vigor and photosynthetic capability (Noel 1970). Below the girdle, tissues experience a shortage of photosynthate that eventually results in depletion of carbohydrate reserves and death (Zimmerman 1960,

Noel 1970).

Girdling is known to enhance the susceptibility of oak (Quercus spp.) and birch

(Betula spp.) to two native Agrilus species A. bileneatus and A. anxius, respectively

(Anderson 1944, Haack 1982, Dunn 1987, Muilenburg 2010). Girdled white and green ash trees in a forest setting were more attractive to EAB than untreated trees

(McCullough et al. 2009a). Tissues below a girdle have reduced concentrations of total phenolics and rate of wound periderm formation, resulting from increased carbon stress due to isolation from the flow of photosynthate produced in the foliage (Muilenburg

2010). In this dissertation, girdling was used as a means to enhance the susceptibility of ash to EAB (Chapter 2).

The Use of Phytohormones

Jasmonates, such as methyl jasmonate (MeJA), jasmonic acid (JA), jasmonyl- isoleucine (JA-Ile), and other JA-amino acid conjugates and jasmonate derivatives play central roles in regulating defense responses against herbivores that cause various types of tissue damage (Howe and Jander 2008). For instance, the role of proteinase inhibitors in insect defense was established prior to the discovery that they could be regulated through airborne MeJA via interplant communication (Green 1972, Farmer and Ryan

1990). Subsequent studies have found that MeJA, and its non-volatile active form jasmonic acid (JA), have the ability to regulate changes in the production of toxic secondary metabolites, morphological barriers, and vegetative growth rates (Gundlach et al. 1992, Yan et al. 2007, Koo and Howe 2009, Pauwels et al. 2009, Yoshida et al. 2009).

The dynamic range of responses regulated by jasmonates in planta suggests that these hormones play very significant roles in regulating tradeoffs between growth and defense

(Herms and Mattson 1992, Baldwin 1998, Koo and Howe 2009).

Wounds created by chewing insects, and necrosis of plant tissues caused by phytotoxins produced by necrotrophic pathogens, result in the rapid accumulation of JA at the site of wounding in plants (Glazebrook 2005, Howe and Jander 2008, Kliebenstein and Rowe 2008). MeJA is converted into JA via methyl jasmonate esterase (MJE).

Responses are regulated through a signaling cascade when JA accumulates in plant

tissues and is converted into jasmonate-amino acid conjugates through the jasmonate- resistant 1 enzyme (JAR1) and begins to accumulate in tissues, which causes the derepression of the JAZ (jasmonate ZIM-domain) repressor protein, allowing for the synthesis of defense related gene products (Thines et al. 2007, Howe and Jander 2008,

Pauwels et al. 2009, Ballare 2011). Derepression of the JAZ protein results in the activation of transcriptional regulons that lead to the induction of direct, indirect, and other defense responses, as well as developmental processes under certain conditions

(Howe and Jander 2008).

MeJA has the ability to prime plants through modifications of the transcriptome/metabolome that do not necessarily confer resistance directly, but prepare the plant to respond more quickly and with a more intense response following attack by an insect or pathogen (Ballare 2011). Exogenous application of MeJA to the outer bark of certain tree species has direct impacts on insect performance and induction of defense related traits. For instance, exogenous application of MeJA to the outer bark of Norway spruce (Picea abies L.) caused significant reductions in colonization by the bark-beetle

Ips typographus L., and resulted in significantly shorter parental galleries and fewer eggs deposited, than untreated control plants (Erbilgin et al. 2006). MeJA treated plants were found to have higher concentrations of defensive terpenes and an increased number of traumatic resin ducts (Erbilgin et al. 2006).

Characterization of Insect Oral Secretions

An increasing number of studies has demonstrated that insect feeding or application of insect oral secretions to plant wounds, elicit a stronger response from plants than just mechanical damage by itself (Arimura et al. 2004a, Arimura et al. 2004b, Philippe et al.

2010). Elicitors have been identified in the oral secretions of various insects and fatty acid-amino acid conjugates (FACs) are among the most important in the oral secretions and salivary extracts of several orders of insect (Howe and Jander 2008). In this dissertation I discuss the characterization of EAB oral secretions derived from 3rd and 4th instar EAB larvae using proteomic and metabolomic approaches. The ultimate goal of

EAB oral secretion characterization is to identify molecules that may be involved in the elicitation/suppression of defense responses in ash.

Proteomics and Metabolomics

To study the interactions between ash and EAB I used techniques that monitor global changes in the proteome and metabolome of ash. The term proteome refers to the entire set of proteins expressed by a genome, cell, tissue, or organism at a given time under defined conditions and was first coined by Marc Wilkins in 1994 (James 1997).

Techniques that monitor global changes in the proteome, referred to as proteomics, include quantitative and qualitative methods that utilize gel-based or gel-free techniques.

The metabolome is the complete set of metabolites synthesized by an organism (Oliver et al. 1998, Fiehn 2002). Metabolomics is the study of metabolic changes. Metabolomic

approaches include metabolite target analysis, metabolite profiling, metabolic fingerprinting, metabolic profiling, and metabonomics (Fiehn 2002).

In this dissertation, quantitative gel-based and qualitative gel-free proteomic approaches were utilized (Chapter 3 and Chapter 6, respectively). Difference gel electrophoresis (DIGE) offers both qualitative and quantitative information about protein differences or changes between total protein extracts (Ünlü et al. 1997). DIGE was developed as a means to overcome the inherently poor reproducibility of 2-dimensional gel electrophoresis (2-DE), which makes accurate quantitative comparisons difficult between protein sample runs on different gels. DIGE overcomes the limitations of conventional 2-DE by using two (or more) fluorescent labels with different excitation and emission spectra to label different samples on the same gel. DIGE reduces the number of gels that need to be run, allows samples to undergo the same technical conditions, and uses an internal standard comprising equal amounts of protein from each sample for globally reliable and reproducible quantitative and qualitative comparisons (Timms and

Cramer 2008). Differentially expressed proteins are excised from gels and subjected to a similar workflow as qualitative, gel-free approaches. Qualitative, gel-free approaches, such as shot-gun proteomics and MudPIT (Multidimensional Protein Identification

Technology), utilize the separation power of liquid chromatography (LC) coupled with the sensitivity of mass spectrometry (MS) to identify individual peptides resulting from digestion of the total protein extracts with trypsin or some other enzyme (Washburn et al.

2001). Once MS data are acquired, a bioinformatics workflow is used to assign identities and decipher potential biological functions and pathways of identified proteins (Barrell et al. 2009, Binns et al. 2009).

Metabolomics bridges the gap between the genotype and the phenotype (Fiehn 2002).

The use of LC, coupled with electrospray ionization MS (LC-ESI-MS), has emerged as a powerful tool for metabolomics investigations (Fiehn 2002). Workflows for the complete analysis of plant extracts have been developed and the analysis of an organism’s metabolome is now possible if the correct approach is taken and the data mining tools are available (Fiehn 2002, Ossipov et al. 2008). In this dissertation, I have taken a metabolite profiling approach toward characterizing the phenolic chemistry of ash and a metabolomics approach in order to characterize EAB oral secretions. Overall, by employing proteomics and metabolomics approaches, I was able to explore some of the links between the genotype of ash and its phenotype with respect to resistance to EAB.

CHAPTER 2

SPECIES-SPECIFIC EFFECTS OF GIRDLING, INSECTICIDE APPLICATION,

AND METHYL JASMONATE PRIMING ON EAB COLONIZATION OF ASH

ABSTRACT

The emerald ash borer is an invasive phloem-feeding, wood-boring insect that was introduced to southeastern Michigan from Asia and first discovered in the U.S. in 2002.

EAB larvae kill ash (Fraxinus spp.) trees by disrupting the flow of water and nutrients throughout the main stem of host plants. In this study, we tested the effects of girdling, exogenous application of methyl jasmonate (MeJA), and a protective insecticide on EAB colonization of, and larval survival in, susceptible green, white, black, and resistant

Manchurian ash. As expected, girdling significantly enhanced the susceptibility of all trees to the borer below the girdle. Furthermore, exogenous application of MeJA induced the same level of resistance in susceptible North American species as available constitutively in Manchurian ash, and provided the same level of protection to the susceptible ash species as that of a commercially available protective insecticide. These data suggest that MeJA could be developed into an effective EAB management tool for susceptible North American ash species in the future and would represent an ecologically friendly alternative to currently used insecticide treatments. 32

INTRODUCTION

The emerald ash borer (EAB) [Agrilus planipennis Fairmaire (Coleoptera:

Buprestidae)] was first discovered in SE Michigan in 2002 and has since become established in 15 U.S. states and parts of Canada, as well as all 88 counties in OH (Herms

2004, Poland and McCullough 2006, USDA 2011). In its native range of China, Korea,

Japan, and parts of Russia and Mongolia EAB attacks Fraxinus chinensis, F. chinensis var. rhyncophylla, and F. mandshurica (Yu 1992, McCullough 2004). The larval stage of the insect is responsible for the devastation of ash in North America. Around the end of

May/early June female EAB adults begin ovipositing in the cracks and crevices of the outer bark of a host tree. Neonates then hatch and bore into the outer-bark and secondary phloem tissues continuing to feed and develop in the secondary phloem, inner cambium, and outer xylem (Haack 2002, Cappaert 2005). Larvae overwinter as pre-pupae and emerge as adults the following spring to complete the cycle (Haack 2002, Wei et al.

2007).

All of the major eastern North American ash species are susceptible to attack by

EAB, including green (Fraxinus pennsylvanica), white (F. americana), black (F. nigra), and blue ash (F. quadrangulata) (Cappaert 2005, Poland and McCullough 2006), but differences in feeding preference have been documented in several studies. For instance, healthy green ash was the most attractive and heavily infested ash species when compared to healthy white and black ash, while blue ash was the least attractive in controlled studies (Anulewicz et al. 2007, Anulewicz et al. 2008, Pureswaran and Poland

2009). In the field, black ash was observed to be highly susceptible to attack by EAB

(Smith et al. 2005). Susceptibility of North American ash species is likely caused by the lack of a shared co-evolutionary history with the insect (Bryant 1994, Swihart 1994,

Rebek et al. 2008).

EAB does not cause widespread mortality of ash in its native landscape (Chinese

Academy of Sciences 1986, Yu 1992). Therefore, it was hypothesized that Manchurian ash, which has a co-evolutionary history with the insect, was resistant to EAB. Indeed, results from a common garden planting of several ash cultivars of different species (white cv. ‘Autumn Purple’, green cv. ‘Patmore’, Manchurian cv. ‘Mancana’, and a black x

Manchurian ash hybrid cv. ‘Northern Treasure’) under high EAB pressure showed that

Manchurian ash cv. ‘Mancana’ was resistant (Rebek et al. 2008). The application of an insecticide (imidacloprid) showed no effect when compared to untreated Manchurian ash, further supporting its high level of resistance (Rebek et al. 2008).

While EAB has the ability to attack and kill healthy and stressed ash trees in North

America, in Asia it appears to function as a secondary pest that colonizes only stressed or declining host trees (Yu 1992). Two North American native Agrilus species, A. bileneatus and A. anxius, attack oak (Quercus spp.) and birch (Betula spp.), respectively, but only when the trees have been stressed naturally (drought, injury, etc.) or artificially

(girdling) (Anderson 1944, Haack 1982, Dunn 1987, Muilenburg 2010). Girdled white and green ash trees in a forest setting were also more attractive to EAB than untreated trees (McCullough et al. 2009a). Girdling has been found to increase plant susceptibility to pests below the girdle point. Tissues below a girdle have reduced concentrations of total phenolics and show reduced rates of wound periderm formation, resulting from

increased carbon stress due to isolation from the flow of photosynthate produced in the foliage (Muilenburg 2010).

Multiple studies have focused on the protection of ash against EAB using insecticides

(McCullough 2004, McCullough et al. 2005, Smitley 2006). Insecticide options available for effective control of EAB include: 1) systemic insecticides applied as soil injections or drenches; 2) systemic insecticides applied as trunk injections; 3) systemic insecticides applied as lower trunk sprays; and 4) protective cover sprays that are applied to all parts of the tree (depending on the product) (Herms 2006). Imidacloprid is an effective systemic insecticide that can be applied as either a soil drench or a soil/trunk injection while bifenthrin is an effective protective bark and foliar treatment (Herms 2006).

However, widespread insecticide use is not the ideal means by which to control pests because insect resistance to traditional pesticides is on the rise, and the potential for environmental damage from pesticide sprays is causing increasing fear among the public

(Hallenbeck and Cunningham-Burns 1985, Roush and Tabashnik 1990, Margni et al.

2002). These general concerns regarding widespread pesticide use make the need of alternative pest management strategies more evident.

An alternative to the use of pesticides to control insect pests involves the exploitation of the endogenous resistance mechanisms in plants to combat insect attack. Plant defense elicitors, such as methyl jasmonate (MeJA), a plant stress hormone involved in the regulation of plant defense responses against chewing insects (Howe and Jander 2008), has the potential to be used as a means to prime the endogenous defense machinery in plants prior to insect attack. It has been shown in other systems to confer resistance against herbivore pests (Chen et al. 2004). However, the use of MeJA as a means to

control insect pests in the field has been limited, and MeJA for pest control is not currently available as a pest management option (Black et al. 2003). MeJA has been tested in the field as a means to enhance the attraction of ash trees to EAB (McCullough et al. 2009b), but no such effect was observed (McCullough et al. 2009b). However,

MeJA application did elicit changes in the volatile emission profiles of ash that were similar to profiles of trees being actively fed upon by EAB larvae, suggesting a potential role for MeJA in the EAB – Fraxinus interaction (Rodriguez-Saona et al. 2006). Also, exogenous application of MeJA to outer bark tissues of ash induced the accumulation of phenolics in phloem tissues, suggesting that the MeJA defense pathway is active in

Fraxinus spp. (Chapter 5). Therefore, conflicting evidence surrounding the effects of

MeJA on Fraxinus spp. requires further investigation into its role and potential benefits.

In this study, we quantified EAB colonization of susceptible green, white, and black ash, and resistant Manchurian ash that were: 1) treated with a protective cover spray insecticide (bifenthrin; OnyxTM) (positive control); 2) left untreated (negative control); 3) girdled; or 4) treated with MeJA. We hypothesized that application of MeJA to the main stems of ash trees would protect susceptible species against EAB attack via induction of endogenous defenses in phloem tissues. We also hypothesized that girdling would enhance the susceptibility of all species tested to attack by EAB. By girdling trees, we are able to artificially impose stress on the trees under investigation in a field setting, making them susceptible to attack by EAB. Comparison of defensive traits in plant tissues above and below the girdle may give insight into the essential mechanisms governing resistance against EAB.

MATERIALS AND METHODS

Experimental Design

Clonal individuals of F. mandshurica cv. ‘Mancana’, F. nigra cv. ‘Fallgold’, F. pennsylvanica cv. ‘Patmore’, and F. americana cv. ‘Autumn Purple’ and seedlings of F. excelsior were obtained from Bailey Nursery, Inc., St. Paul MN, USA. Seedlings of F. quadrangulata were obtained from Moon Dancer Farm, Lexington KY, USA, and seedlings of F. pennsylvanica were obtained from the USDA Forest Service, Northern

Research Station, Delaware OH, USA. All of the aforementioned groups, except F. nigra cv. ‘Fallgold’, were planted in a randomized complete block design (eight blocks) in a common garden established in November 2007 in Bowling Green, OH in an open field next to the water treatment plant for the city (41.38° N/83.61° W). A separate block of F. nigra cv. ‘Fallgold’ trees was planted adjacent to the common garden in a completely randomized design in May 2008. All trees were spaced 2.4 m apart within a block and blocks were spaced 3.6 m apart from each other. Each species was replicated four times in each block so each treatment could be equally replicated across species and blocks for a total of N = 8 for all species x treatment combinations. Likewise, the separate black ash block was composed of 32 individual trees so each treatment could be replicated 8 times within that block. Blue and European ash seedlings had mean stem diameters that significantly differed from the other species at the time of this experiment, and as a result exit hole data collected on blue and European ashes were not included in this analysis.

All treatments were randomly assigned within each block. Thus, a total of 160 trees were

used throughout the entire experiment.

In the spring of 2008 all trees in the plot were treated with a protective cover spray

(OnyxTM) to prevent local populations of EAB from colonizing trees prematurely. All tree stems were also covered in tree wrap to inhibit deer browsing and groundhog grazing. Tree wrap was then removed at the beginning of the 2009 summer. It was observed that the effectiveness of the 2008 protective insecticide spray was not 100%, and therefore all EAB exit holes were recorded in the fall of 2009 for each tree located within the plot.

The ash plots were planted in an area with high levels of EAB infestation. In addition to a surrounding EAB population, and to minimize escapes and homogenize attacks within the plots, EAB populations were artificially augmented by placing 11 ash bolts (each ~ 60 cm long), within each block on April 15, 2009. The bolts derived from

10 heavily infested green ash trees (50-75% canopy dieback was used to select trees).

Each tree was cut into 10 bolts, which were then labeled 1 – 10 starting from the bolt taken from the highest portion of the tree down to the base. Bolts were then randomized into 9 groups and each group was assigned to a block. The bolts had an average of 36.1

± 3.3 (s.e.m.) exit holes/bolt at the end of the 2009 growing season. Thus, we augmented the number of natural immigrants from surrounding infested areas by more than 3,500 active adults throughout the ash plot during the course of the 2009 summer. By using this approach we can reasonably assume that our design maximized the chance of trees coming under attack from EAB uniformly in our research plot.

Each tree within a species received one of the following treatments in the summer of

2009, with each treatment equally replicated within a block and species: 1) no treatment

(negative control), 2) insecticide (positive control), 3) methyl jasmonate (MeJA), and 4) girdling. Negative controls were left untouched throughout the course of the experiment.

Positive controls received a one-time application of OnyxTM (a.i. bifenthrin), using the recommended rate as labeled for protection against wood-borers, during the third week of

May 2009 by spraying and coating the entire stem of each tree.

For the MeJA treatment, a 1 M solution of MeJA in water, with 100 µl of Tween 20 to break surface tension, was applied evenly along the main stem of each tree using a foam paintbrush. MeJA was applied three times through the course of the 2009 summer

(May 27th, July 1st, and August 3rd).

Trees in the girdling treatment were girdled on May 27th, 2009 using a pipe cutter to remove a 1-inch strip of phloem halfway up the stem of each tree. At time of the first application of MeJA, Manchurian cv. ‘Mancana’, black cv. ‘Fallgold’, green ‘Patmore’, green seedlings, and white cv. ‘Autumn Purple’ ash trees had mean stem diameters of 3.2

± 0.06 (s.e.m.) cm, 2.7 ± 0.08 cm, 3.2 ± 0.05 cm, 2.8 ± 0.1 cm and 3.2 ± 0.1 cm, respectively, at 15 cm above the soil line. These trees were 6-years old by the time the experiment began. It has been shown that trees of this size are large enough to be colonized by EAB (Rebek et al. 2008).

Sampling

Emergence holes of EAB adults were counted on July 5th, 2010 by visually inspecting the main stem and branches of all trees. To evaluate the effectiveness of our treatments applied in the summer of 2009 on EAB larval survival, we subtracted emergence holes

counted in 2009 from emergence holes counted on the same trees in the summer of 2010 to remove any confounding factors not directly related to our experiment. Therefore, numbers of D-shaped emergence holes created by larvae emerging in the summer of 2010 were the result of EAB activities originating in the summer of 2009. Total exit holes per tree (exit holes/tree in 2010 – exit holes/tree in 2009 = Total exit holes/tree) were used for all subsequent statistical analyses.

Statistical Analysis

Since black ash was not part of the initial RCB design all emergence data for black ash were analyzed separately. A Student’s t-test was used to evaluate exit hole number differences above and below the girdle for black ash. Univariate ANOVA of square root- transformed exit hole data (to meet assumptions of normality and homogeneity of variance) was used to evaluate differences between all species and treatments. All statistical analyses were conducted using IBM SPSS Statistics v. 19 (SPSS Inc. 2010).

RESULTS

EAB Colonization of Treated Ash Species within the RCB Design

Frequencies of exit holes differed significantly by species and treatment, and the species x treatment interaction was also significant (Figure 2.1). MeJA treatment induced a 64.5% reduction in the number of exit holes across species when compared to control

trees (Figure 2.1). Our results also confirm that Manchurian ash has a high level of resistance to EAB, even when artificially stressed (girdled), which is in agreement with the findings of Rebek et al. 2008.

EAB Colonization of Treated Black Ash

EAB emergence holes on black ash trees were significantly affected by treatment

(Figure 2.1). MeJA treated black ash trees showed on average a 70.1% reduction in EAB emergence holes when compared to control trees (Figure 2.1).

DISCUSSION

EAB attack densities on control trees of both resistant and susceptible species in our study (Figure 2.1) are in agreement with the findings of Rebek et al. 2008. Also, while black ash was separate from the RCB experiment in our study, and thus cannot be compared directly to the other species in a statistical manner, it is still noteworthy that black ash control trees had a higher number of emergence holes (9.6 ± 1.9) than the control treatments for white (5.1 ± 2.2), green ‘Patmore’ (5.4 ± 1.4), green seedling (3.0 ±

1.5), and Manchurian ash (0.1 ± 0.1), suggesting a higher level of susceptibility of, and/or preference for EAB adult oviposition on, black ash. These data are in agreement with previous observations that black ash in the field succumbs quickly to EAB infestation

(Smith et al. 2005). Black ash may be particularly susceptible because, while having a very similar genetic background to Manchurian ash(one of the natural hosts in EAB’s

Figure 2.1. Effects of species and treatment on EAB attack rates (number of emergence holes per trees). Species (F3,127 = 13.526; P < 0.001) and treatment (F3,127 = 17.349; P < 0.001) had significant effects. The insecticide and MeJA treatments had statistically similar effects, but were both different from the other two treatments (protected LSD test, P < 0.05). The species x treatment interaction was also significant (F3,127 = 2.889; P < 0.01). Treatment effects for black ash were also significant (F3,31 = 13.864; P < 0.001) similar to those observed in the other species, including the statistical equivalence of the MeJA and insecticide treatments. Error bars are the standard error of the mean (s.e.m.) and different letters indicate significantly different means separated by the protected LSD test (! = 0.05).

native range) as a result of close phylogenetic relationships, it lacks the resistant traits that co-evolutionary history has conferred to Manchurian ash (Jeandroz et al. 1997,

Wallander 2008, Rebek et al. 2008, also see Chapter 3 and Chapter 4).

Over the course of a three-year period, Rebek et al. 2008 found that emergence hole frequency mirrored percent survival of ash in a field trial, and that there was a positive correlation between percent canopy dieback and EAB emergence hole frequencies. Our study was conducted in the summer of 2009 and ended with the counting of exit holes in the summer of 2010. While we collected mortality data in the summer of 2010, most trees were still surviving (data not shown). However, the findings of Rebek et al. 2008 identify a strong relationship between EAB emergence hole data and ash mortality, providing support to the measurement of emergence hole data as an indicator of tree health and survival over time as it relates to infestation by EAB. Likewise, work conducted on EAB congeners has found similar relationships between emergence holes and tree mortality (Anderson 1944, Barter 1957, Muilenburg 2010). Therefore, we conclude that emergence hole data is predictive, and therefore a good indicator, of EAB attack densities and tree mortality over time.

Girdling is a technique that has been used to artificially impose stress in trees by impeding the flow of photosynthate synthesized in the foliage above the girdling point to tissues below the girdling point (Noel 1970). The effect of girdling as inducer of susceptibility to wood-boring Agrilus spp. below the girdling point is a well-documented phenomenon (Anderson 1944, Barter 1957, Muilenburg 2010). We used girdling as a means to enhance susceptibility of both North American and Asian ash trees to attack by

EAB (Figure 2.1). Girdling is known to affect the normal physiology of host plants by

creating an imbalance in carbon allocation between above and below girdled plant parts.

Photosynthate accumulates in plant tissues above the girdle, while carbon pools below the girdle are quickly depleted. By employing a girdling treatment, we were attempting to make Manchurian ash cv. ‘Mancana’ trees susceptible to attack by EAB, and enhance susceptibility of North American ash species. However, Manchurian ash cv. ‘Mancana’ was still well protected, even when artificially stressed, suggesting that constitutive (pre- formed) defenses in Manchurian ash cv. ‘Mancana’ may be critical in defense against

EAB.

MeJA and insecticide treatments did not differ statistically for all species in the RCB design and black ash (Figure 2.1). MeJA is a plant hormone involved in the regulation of plant defenses against insect herbivores, more specifically, chewing insects such as

EAB (Farmer and Ryan 1990, Howe and Jander 2008). In planta, MeJA is converted into its non-volatile form, jasmonic acid (JA), by the enzyme MeJA esterase (MJE). JA then activates the signaling cascade that regulates plant defense responses against chewing herbivores (Glazebrook 2005, Howe and Jander 2008).

The few existing studies regarding the effects of MeJA on Fraxinus spp. approached

MeJA as a stress inducing compound, rather than a defense inducing compound

(Rodriguez-Saona et al. 2006, McCullough et al. 2009b, McCullough et al. 2009a, also see Chapter 5). Rodriguez-Saona et al. 2006 identified several volatile compounds that were induced upon treatment of Manchurian ash with MeJA, suggesting that the MeJA pathway is active in Fraxinus spp. In a separate experiment, MeJA was applied exogenously to green, white, and Manchurian ash trees in a common garden in Novi, MI at concentrations of 0.0, 0.1, 0.5, 1.0, and 2.0 M (see Chapter 5). Analysis of phloem

tissues in those trees showed that the concentration of MeJA ranged from 0 to 2.3 "mol/g

FW, indicating that MeJA was effectively absorbed into host tissues (Chakraborty et al. unpublished). That study also identified the 1.0 M treatment as the most effective dose because: 1) it was able to elicit changes in phenolic chemistry more effectively than the

0.1 and 0.5 M dosages and 2) it did not induce necrotic lesions around lenticels or on foliage as was observed for 2 M treatments. MeJA was found to accumulate in concentrations of 1.5, 1.2, and 0.8 µmol/g FW in Manchurian, green, and white ash phloem tissues, respectively. The same 1.0 M treatment was shown to induce accumulation of total and individual phenolics in the phloem tissues of white and

Manchurian ash in yet another study (see Chapter 5).

Based on these results, we hypothesize that MeJA could be used to induce the endogenous defense machinery of susceptible ash species in the field and thus inhibit or reduce attack by EAB. Our results indicate that the MeJA treatment offered the same level of protection as an insecticide (OnyxTM) formulated for the management of bark beetles (DeGomez et al. 2006). This suggests that MeJA could be developed into an environmentally friendly ‘green technology’ that could replace insecticides. The idea of using MeJA as a means to control insect pests is not novel, but its application in the real world has been limited (Ryan 1990, Black et al. 2003), although it has been previously recognized that host plant resistance is the ideal means of controlling insect pests

(Hanover 1975, Herms 2002). In any case, further experiments are needed to test the durability of MeJA application to manage EAB in the field.

We hypothesize that MeJA provided protection against EAB via one or a combination of the following mechanisms: 1) MeJA is directly toxic to EAB eggs and/or larvae; 2)

application of MeJA to the outer bark of ash induces changes in the volatile profiles of ash phloem tissues that deter oviposition selection by gravid EAB females; and/or 3)

MeJA induces the production of defensive metabolites and proteins in phloem tissues that are directly toxic to or detrimental to the growth and survival of EAB larvae in planta.

In conclusion, we have (1) demonstrated the ability to alter the physiology of ash trees in the field, which in turn had a direct effect on the colonization preferences/larval survival of EAB on susceptible and resistant Fraxinus spp., and (2) identified a potentially important management tool for EAB in the form of a well-characterized natural plant hormone, MeJA. Further work on the biochemical and molecular components of resistance in physiologically altered hosts will facilitate the identification of effective resistance mechanisms in ash against EAB.

ACKNOWLEDGEMENTS

We thank David S. Bienemann for help with establishing and maintaining ash trees in

Bowling Green, OH, providing infested ash logs, and for his suggestions on maintaining healthy trees in the field; Bryant Chambers for help with the application of insecticide and instructions on the proper care and planting for trees in the field; Diane Hartzler and the Herms lab for help in establishing the Bowling Green ash plot; Gerardo Suazo for technical help and experimental execution; and Karla Medina-Ortega and Ronald Batallas for assistance in the collection of EAB exit hole data.

CHAPTER 3

INTERSPECIFIC PROTEOMIC COMPARISONS REVEAL ASH PHLOEM

GENES POTENTIALLY INVOLVED IN CONSTITUTIVE RESISTANCE TO

THE EMERALD ASH BORER

ABSTRACT

The emerald ash borer (Agrilus planipennis) is an invasive wood-boring beetle that has killed millions of ash trees since its accidental introduction to North America.

All North American ash species (Fraxinus spp.) that emerald ash borer have encountered so far are susceptible, while an Asian species, Manchurian ash (F. mandshurica), which shares an evolutionary history with emerald ash borer, is resistant. Phylogenetic evidence places North American black ash (F. nigra) and Manchurian ash in the same clade and section, yet black ash is highly susceptible to the emerald ash borer. This contrast provides an opportunity to compare the genetic traits of the two species and identify those with a potential role in defense/resistance.

We used Difference Gel Electrophoresis (DIGE) to compare the phloem proteomes of resistant Manchurian to susceptible black, green, and white ash.

Differentially expressed proteins novel to the resistant Manchurian ash when compared to

the susceptible ash species were identified using nano-LC-MS/MS and putative identities assigned. Proteomic differences were strongly associated with the phylogenetic relationships between all four species. Proteins identified in Manchurian ash potentially associated with its resistance to emerald ash borer include a PR-10 protein, an aspartic protease, and a thylakoid-bound ascorbate peroxidase. Discovery of resistance related proteins in Asian species could facilitate approaches for introgressing resistance genes into North American species to generate ash genotypes for use in forest ecosystem restoration and urban plantings in the wake of the emerald ash borer invasion.

INTRODUCTION

The emerald ash borer (EAB), Agrilus planipennis Fairmaire (Coleoptera:

Buprestidae), is an invasive insect that has killed tens of millions of ash (Fraxinus spp.) trees in the U.S. and Canada (Cappaert 2005, Poland and McCullough 2006). Larvae feed on phloem and outer xylem of host trees, which disrupts translocation of water and nutrients resulting initially in canopy thinning and ultimately death within one to three years of first expression of symptoms (Chinese Academy of Sciences 1986, Yu 1992,

Herms 2004, Poland and McCullough 2006). All North American species of ash that emerald ash borer has encountered thus far are susceptible to colonization, even when growing on high quality sites and in the absence of obvious environmental stress (Herms

2004, Rebek et al. 2008, Pureswaran and Poland 2009).

Conversely, Asian species of ash, which share a coevolutionary history with emerald ash borer, appear to be colonized only when weakened by abiotic or biotic stress

(Gould et al. 2005, Liu et al. 2007). In a common garden experiment, Rebek et al. 2008 found Manchurian ash (F. mandshurica Ruprecht), which is a primary host in its endemic range, to be much more resistant to emerald ash borer than North American green (F. pennsylvanica Marsh) and white ash (F. americana L.). A Manchurian x black ash (F. nigra Marsh) cross was also highly susceptible, indicating the hybrid did not inherit emerald ash borer resistance from its Asian parent. North American black ash is also known to be highly susceptible. For example, in Michigan forests mortality of black ash proceeds at a faster rate than green and white ash (Smith et al. 2005).

Resistance of deciduous trees to wood-boring insects is hypothesized to be the result of a combination of constitutive and induced, physical and phytochemical defenses that deter or kill the insect (Dunn 1990). Constitutive traits that confer resistance to wood-borers, such as defensive phytochemicals or proteins, could serve as biomarkers for use in introgressing emerald ash borer resistance genes into North American ash species via hybridization or transgenic approaches.

While insightful information can be obtained at the level of gene sequence and gene expression (genomics and transcriptomics, respectively), it is ultimately the proteome and the products of enzymatic reactions that dictate the interaction between plant and herbivore. Proteins that mediate plant-insect interactions include those that confer resistance directly (e.g. cysteine proteases or proteinase inhibitors) or indirectly through their roles in defense pathways (e.g., enzymes involved in the biosynthesis of defensive phytochemicals) (Howe and Jander 2008). Therefore, proteins serve as a logical starting point in the search for putative resistance genes.

One approach to investigate putative constitutive resistance traits is to use high- throughput methods to compare susceptible and resistant hosts. Proteomic high- throughput methods include techniques, such as difference gel electrophoresis or DIGE, that provide qualitative and quantitative information on total proteomic differences between two or more experimental units (Ünlü et al. 1997). Information garnered from

DIGE studies can serve as the basis for functional experiments in which resistance genes can be characterized in planta using transgenic approaches or through the use of Asian x

North American ash hybrids coupled with in depth analyses of the interaction between the modified/hybrid plant and the pest.

In this study, the constitutive proteome of whole phloem tissue of Manchurian ash was compared to that of three susceptible North American ash species. A recent phylogenetic analysis of the genus Fraxinus, based on DNA sequences from the nuclear ribosomal ITS and two chloroplast regions of all 43 species, placed black and Manchurian ash in the same clade in the section Fraxinus (Figure 1.1) (Wallander 2008). The contrast between phylogenetic similarity and divergence in phenotype of resistant Manchurian and susceptible black ash to emerald ash borer provides an opportunity to investigate their genetic differences in order to identify potential resistance traits.

DIGE has not been used extensively to study interspecific variation (Davis et al.

2009). In order to identify biologically meaningful differences in Manchurian ash phloem tissues we first compared it to the black ash proteome followed by comparisons against the white and green ash phloem proteomes. Due to their phylogenetic dissimilarity to Manchurian and black ash, we further augmented our selection of potential resistance related gene candidates through the incorporation of green and white ash (Figure 1.1). The primary objective of the study was to identify genes in the phloem tissues of Manchurian ash associated with potential mechanisms of resistance to emerald ash borer.

MATERIALS AND METHODS

Experimental Design

Clonal individuals of F. mandshurica cv. ‘Mancana’, F. nigra cv. ‘Fallgold’, F. pennsylvanica cv. ‘Patmore’, and F. americana cv. ‘Autumn Purple’, were obtained from

Bailey Nursery, Inc., St. Paul MN, USA. Six-yr-old saplings of each species were planted in a common garden established in November 2007 in Bowling Green, OH.

Trees were planted in a randomized complete block design with eight blocks. One sapling per block of each species for a total of 8 biological-clonal replications per species were sampled on August 6, 2008. At sampling, Manchurian, black, green, and white ash trees had mean stem diameters of 3.4 ± 0.06 (s.e.m.) cm, 2.9 ± 0.1 cm, 3.6 ± 0.04 cm, and

3.3 ± 0.1 cm respectively, at 30 cm above the soil line. Second year branches were chosen for analysis. Branches were removed from trees, stripped of leaves, placed on ice, and then transported back to the lab where phloem tissue was immediately removed, frozen in liquid N2, and stored at -80° C until protein extraction.

Protein Extraction and Purification

Proteins were extracted according to Vâlcu and Schlink 2006, with minor modifications to accommodate for differences in scale, protein extract preparation, and cleaning for

DIGE experiments. Phloem tissue was ground in liquid nitrogen and 0.1 g was suspended in 500 $l of pre-cooled (-20° C) precipitation solution: 10% TCA (Sigma-

Aldrich; St. Louis, MO, USA) and 20 mM DTT (BioRad; Hercules, CA, USA) in acetone. Proteins were precipitated overnight at -20° C. Phloem tissue and precipitated proteins were then washed twice for 1 h each at -20° C with 1 ml of 20 mM DTT in acetone and pelleted by centrifugation for 30 min at 26 000 x g (4° C). Following removal of supernatant, the pellet was dried under vacuum for 10 min and re-extracted twice with 500 and 200 µl of extraction buffer [7 M urea (BioRad), 2 M thiourea (Sigma-

Aldrich), 4% CHAPS (BioRad), 50 mM DTT (BioRad), and 1x Complete Protease

Inhibitor cocktail (Roche; Indianapolis, IN, USA)]. Protein extracts were then subjected to two subsequent purification steps prior to DIGE in order to remove remaining contaminants (i.e. salts, sugars, and secondary plant compounds). For the first clean-up step, proteins were precipitated from the protein extracts using the Bio-Rad ReadyPrep!

2-D Clean-up Kit and re-suspended in 25 $l of a DIGE compatible buffer [7 M urea, 2 M thiourea, 4% CHAPS, 30 mM Tris (pH 8.5), and 1x Complete Mini EDTA-free Protease

Inhibitor Cocktail (Roche)]. Protein pellets were placed on a shaker at RT for 3 hrs to allow for complete re-solubilization of the pellet. Protein concentrations were measured with the Coomassie Plus! Protein Assay (Pierce; Rockford, IL, USA) and compared against a standard curve of BSA prepared in a DIGE compatible buffer.

1-D SDS-PAGE

A 1-D protein gel was run to ensure that the integrity of the proteins was not affected

(Figure 3.1). Twenty µg of ash phloem protein samples were separated under denaturing conditions in a discontinuous sodium dodecyl sulfate polyacrylamide gel (SDS-PAGE)

(stacking gel – 3.5%; resolving gel – 12%). Before separation on the gel, ash protein samples were diluted in Laemmli buffer (1:1). The SDS-PAGE gel was run in a Mini

Protein 3 Cell (Bio-Rad, CA, USA) at 75 V for 30 min and 150 V for 1 h. After completion of electrophoresis protein bands were visualized in the gel via Coomassie blue staining.

Sample preparation for DIGE

Prior to running DIGE, a second protein clean-up step was necessary to remove residual contaminants that were observed on a 2-D SDS-PAGE gel used to visualize proteins prior to the DIGE experiment (data not shown). Protein extracts were precipitated out of solution with a 4:1 (v:v) methanol : chloroform mixture in order to remove residual sugars. Following precipitation, protein pellets were re-suspended in lysis buffer [7 M urea, 2 M thiourea, 4% CHAPS, 30 mM Tris (pH 8.5)]. Proteins were then quantified via the Bradford assay using BSA as the standard and diluted to a concentration 1 $g/$l.

Each of the eight replicate protein extracts for each species were randomly labeled with 8 pM of either Cy3- or Cy5- minimal labeling dyes (GE Healthcare; Little Chalfont, UK) in

DMF (N, N – dimethylformamide) for every 1 $g of protein. Following calibration, the internal standard (IS), consisting of 22 $l from each sample, was pooled into one tube

(total 704 $l) and labeled with Cy2- minimal labeling dye in the same proportions as with the other samples. Labeled samples were allowed to react on ice for 30 min. After 30 min, 1 $l of 10 mM lysine was added to each sample to quench the labeling reaction.

Labeled samples (a total of 40 µg from each sample) were then combined as shown in

Figure 3.1. A 1-D SDS-PAGE gel of protein extracts (20 µg per lane) from Manchurian, black, green, and white ash phloem tissues showing the high quality of the extracts. Protein extracts are pools from eight biological replicates.

Table 3.1, vortexed, and diluted with sample buffer [7 M urea, 2 M thiourea, 2%

CHAPS, 65 mM DTT and 0.5% immobilized pH gradient (IPG) buffer (GE)] to a volume of 470 µl before loading on the IPG strip.

IPG Strip pI Range and First Dimension Isoelectric Focusing

A 2-D SDS-PAGE gel using an IPG strip with a pI range of 3-10 was used to determine the optimum range of proteins in our extracts (Figure 3.2). The majority of proteins were found to be concentrated in the 4-7 pI range. IPG strips (pI 4-7) were then used for all subsequent analyses in order to achieve better protein spot resolution (Figure 3.3).

Prior to isoelectric focusing, samples were applied to 24 cm IPG strips (pI 4-7), overlaid with mineral oil, and allowed to rehydrate passively overnight. Isoelectric focusing (IEF) was performed using an Ettan" IPGphor" II Isoelectric Focusing System (GE

Healthcare/ Amersham Biosciences). Isoelectric focusing took place using the following conditions: instrument temperature set at 20° C; maximum current set at 75 µA/strip; step

(1): step and hold at 500 V for 1 hr (500 total Vhr); step (2): rapid ramping up to 1000 V for ~1 hr (800 total Vhr); step (3): rapid ramping up to 10,000 V for 3 hr (16,500 total

Vhr); step (4): hold at 10,000 V up to a total of 52,500 Vhr for 3.5 hr. After focusing, strips were stored at -80° C until used in the second dimension analysis.

Figure 3.2. A 2-D SDS-PAGE gel (pI 3-10) of a pooled protein extract consisting of equal parts derived from 32 individual biological replicates (n = 8 each for Manchurian, black, green, and white ash). Most of the proteins are found in the 4-7 pI range, which was subsequently used in all DIGE analyses.

Figure 3.3. 2-D DIGE gel overlay. The gel image represents proteins (pI 4-7) derived from Manchurian ash (Cy3– red spots) and black ash (Cy5– green). The internal standard (composed of equal parts from all four ash protein extracts) is visible in this gel as Cy2- blue spots. Yellow indicates non-differentially expressed, overlapping spots common to all species. Numbered protein spots are associated with resistant Manchurian ash and susceptible black, green, and white ash (see Tables 3.3 and 3.4).

Strip Equilibration and Second Dimension

Prior to running the second dimension gels, IPG strips were thawed, reduced in equilibration buffer I (6 M urea, 30% glycerol, 75 mM Tris pH 8.8, 2% SDS, and 0.5%

DTT), and alkylated in equilibration buffer II (6 M urea, 30% glycerol, 75 mM Tris pH

8.8, 2% SDS, and 4.5% iodoacetamide) for 15 min in each. Following equilibration, IPG strips were rinsed in SDS running buffer (25 mM Tris, 192 mM glycine, 0.2% SDS), and sealed on top of 12% SDS-PAGE gels (26 x 20 x 0.1 cm). Gels were then overlaid with

0.5% w:v agarose and a trace amount of bromophenol blue in running buffer, and run at a constant 2 W at 20° C for 45 min and constant 15 W per gel for 4.5 hr in SDS running buffer (25 mM Tris, 192 mM Glycine, 0.1% SDS).

Gel Imaging and Statistical Analysis

Following SDS-PAGE, gels were scanned using a Typhoon Variable Mode Imager 9400

(GE Healthcare). Cy2-, Cy3-, and Cy5- labeled protein images were produced by excitation of gels at 488, 532, and 633 nm, respectively and emission at 520, 590, and

680 nm, respectively. Gel images were analyzed statistically using DeCyder v6.05.11 software (GE Healthcare). A total of eight biological replicates/species were used for all statistical analyses. Each image (32 analytical images and 16 IS images) was manually analyzed to exclude saturated spots, artifacts (dust spots) and noise. The IS image of gel

10 was chosen as a master gel because of the gel quality and total number of spots detected (2,434). The settings used to detect spots were optimized using the master gel as

a reference and a number was assigned to each spot. The same settings were then used for all other gel image analyses. Boundary and volume of protein spots were detected according to the Decyder spot detection algorithm. Statistical analyses were performed as follows: normalized protein volume ratios were calculated (DIA; differential in-gel analysis module) for each individual protein spot from Cy3- and Cy5- sample relative to

Cy2- (IS) corresponding to the same spot. These values were used for further statistical analyses and referred to as the standardized abundance. Differences in average standardized abundances between experimental groups show differential protein expression and are expressed as an average ratio. One-way ANOVA was performed using the Decyder software (which also simultaneously controls for false discovery rate

(FDR)) to evaluate differences in protein expression between all 4 experimental groups.

The same analysis was then used to evaluate differences between only two experimental groups (i.e. black vs. Manchurian, green vs. Manchurian, and white vs. Manchurian) as an equal variance, two-tailed Student’s t-test. Proteins of interest (POI) were defined as having abundances that were significantly different (P < 0.05) between: 1) Manchurian and black ash, 2) Manchurian and green ash, and 3) Manchurian and white ash experimental groups along with an absolute abundance ratio that was greater than 2-fold.

We chose to focus on the 355 POI from Manchurian and black ash meeting these criteria for sequencing and identification as black and Manchurian ash share the most recent common ancestor and are therefore closely related phylogenetically (Table 3.1) (Jeandroz et al. 1997, Wallander 2008). We further limited our selection by comparing the average ratios of green and white ash in order to remove proteins that did not differ significantly

(P < 0.05; 2-fold or greater) when compared to proteins identified from the Manchurian

Table 3.1. Details of DIGE in-gel comparisons: M = Manchurian, G = green, W = white, B = black ash. Numbers associated with each species represent biological replicates.

Gel Number Cy3- labeled protein, 40 Cy5- labeled protein, 40 µg each µg each

Gel 1 M1 W3 Gel 2 W1 G3 Gel 3 G1 B3 Gel 4 M2 B4 Gel 5 B1 M3 Gel 6 G2 W4 Gel 7 B2 G4 Gel 8 W2 M4 Gel 9 W5 B7 Gel 10 M5 W7 Gel 11 B5 W8 Gel 12 G5 M7 Gel 13 W6 G7 Gel 14 G6 M8 Gel 15 B6 G8 Gel 16 M6 B8

and black ash comparison. Principal component analysis (PCA) was used to separate experimental groups and was performed using the algorithm provided with the Decyder software (Figure 3.4).

Preparative Gel, Protein Spot-cutting, and Trypsin Digestion

Equal amounts of unlabeled individual biological replicates were pooled for each species and loaded on preparative gels for a total of 400 µg/gel. Two preparative gels were run using the above conditions for first and second dimensions. Preparative gels were fixed

(15% ethanol, 1% citric acid) overnight and stained with LavaPurple (FLUOROtechnics).

Gels were de-stained using 15% ethanol for 1 hr. LavaPurple stained gels were scanned using a Typhoon Variable Mode Imager 9400 (GE Healthcare) with 532 nm excitation wavelength and 610 nm emission filter and matched to the master gel image. The 355 proteins of interest meeting the previously mentioned criteria identified from Manchurian and black ash were subjected to further analysis, of which 264 were successfully matched to the master gel image and used to generate a picking list. Proteins of interest were excised from the gels using the Ettan Spot Picker in conjunction with the Ettan Spot

Handling Workstation (GE Healthcare). Spots were digested with trypsin using the Ettan

Digester robot (GE Healthcare) in preparation for peptide sequencing using Nano-LC-

MS/MS.

Nano LC-MS/MS, Protein Identification, and Data Deposition

Tryptic peptides were sequenced via capillary nano-LC-MS/MS on a Thermo Finnigan

LTQ mass spectrometer equipped with a nanospray source operated in positive ion mode.

Capillary nano-LC-MS/MS was performed using similar methods as described in

(Reddish et al. 2008). Solvent A (50 mM acetic acid in water) and solvent B

(acetonitrile) were used for all chromatographic separations. Samples (5 $l from each sample) were prepared in solvent A and injected onto a µ-Precolumn Cartridge (Dionex,

Sunnyvale, CA) and washed with 50 mM acetic acid. The injector port was switched to inject and the peptides were eluted from the trap onto the column. A 5 cm, 75 µm ID

ProteoPep II C18 column (New Objective, Inc. Woburn, MA) packed directly in the nanospray tip was used for all chromatographic separations. Peptides were eluted directly off the column into the LTQ system using a gradient of 2-80% solvent B over 45 minutes, with a flow rate of 300 nl/min. The total run time was 65 minutes. The nanospray source was operated with a spray voltage of 3 kV and a capillary temperature of 200˚ C was used. The scan sequence of the mass spectrometer was based on the

TopTen™ method; the analysis was programmed for full scan (recorded between 350 –

2,000 Da), and a MS/MS scan to generate product ion spectra to determine amino acid sequence in consecutive instrument scans of the ten most abundant peaks in the spectrum.

The CID fragmentation energy was set at 35%. Dynamic exclusion was enabled with a repeat count of 2 within 10 seconds, a mass list size of 200, an exclusion duration of 350 seconds, the low mass width was 0.5, and the high mass width was 1.5. The raw data files collected using the mass spectrometer were converted to mzXML and MGF files

using MassMatrix data conversion tools version 1.3

(http://www.massmatrix.net/download). For low mass accuracy data, tandem MS spectra that were not derived from singly charged precursor ions were considered as both doubly and triply charged precursors. The resulting MGF files were searched using Mascot

Daemon by Matrix Science version 2.2.2 (Boston, MA). The .mgf files were searched against the NCBI database version 20090710 limited to Viridaeplantae (green plants) as taxonomy (384,871 sequences). Trypsin was selected as the digest enzyme with up to two missed cleavages. Carbamidomethyl and oxidation were set as the fixed and variable modifications. Peptide and fragment mass tolerances were set to ± 2 Da and 0.8 Da respectively. Sequence data were also automatically searched against a decoy database in order to avoid false positives. MASCOT based probability scores were used to evaluate protein identities and were considered correct if the match had a score greater than 70, which indicates identity or significant (P < 0.01) similarity, and two peptide matches.

Identities were accepted in some cases if the above parameters were met and only one peptide match was found along with an expected value that was highly significant (P <

0.001) for the peptide match and the MASCOT protein score was 70 or greater for that single peptide match. All protein identities reported in this paper were checked manually to confirm –b and –y ion sequence tags in MS/MS spectra. Any protein spot to match two or more protein identities in the MASCOT database search (meaning more than one protein was potentially present in the cored spot), or did not pass any of the above criteria, was not used in further analyses. Filtered peptide data have been deposited with

Peptidome, National Center for Biotechnology Information. All information and raw data associated with peptides identified in Manchurian, black, green, and white ash are

accessible on the Peptidome NCBI Peptide Data Resource homepage

(http://www.ncbi.nlm.nih.gov/peptidome) via study accession number PSE148. Specific information about individual peptides identified from black, green, white, and

Manchurian ash are accessible directly through sample accession numbers PSM1313 and

PSM1314.

Protein Gene Ontology Annotation and Identification of Putative Defense and Increased

Susceptibility Genes

Gene Ontology (GO) annotations for biological processes were added to all proteins with an average fold-change of 2 and above (P < 0.05) across all species comparisons (Table

3.5 and Table 3.6). Gene ontologies were added by searching individual gi numbers

(obtained from the MASCOT search files) in the Protein Information Resource Database

(National Biomedical Research Foundation, pir.georgetown.edu, Washington, D.C.).

Proteins that were not found in the PIR database were subsequently searched in the NCBI databank to obtain a basic biological understanding of the protein. Most proteins were associated with multiple biological process GO terms and those not associated with a biological process GO term had a molecular function term that was recorded. Proteins with no biological process GO term but that did have a molecular function GO term was treated as miscellaneous proteins in this paper (Table 3.5 and Table 3.6). A single biological process category was selected for proteins associated with multiple GO terms by choosing the most specific and biologically relevant term to plant defense against herbivores. Black, green, white, and Manchurian ash specific biological process GO

categories were loaded onto the QuickGO (Barrell et al. 2009) annotation page in order to visualize relationships between biological processes for each species independently. A limited subset of high-level GO terms were used as parent categories in order to organize the proteins identified in this study and visualize the ontology distribution of proteins significantly differing (P < 0.05) with an absolute abundance ratio greater than 2-fold between Manchurian, black, green, and white ash (Table 3.5 and Table 3.6).

Potential resistance-related genes in Manchurian ash were selected by using similar criteria as described in (Davis et al. 2009). Potential constitutive resistance- related proteins were selected based on the following criteria: 1) an absolute abundance ratio greater than 5-fold when expression in Manchurian ash was compared to expression across all three susceptible species of ash, 2) a highly significant difference (P < 0.05) in expression was found when compared to the three susceptible ash species, and 3) the protein’s potential direct or indirect role in plant resistance based on its gene ontology annotation as it relates to the known literature of plant defense. Conversely, proteins potentially related to enhanced susceptibility to EAB were chosen based on: 1) an absolute abundance ratio greater than 5-fold when expression in black, green, and white ash was compared to Manchurian ash protein expression, and 2) a highly significant difference (P < 0.05) in expression was found when all three susceptible ash species were compared to Manchurian ash.

RESULTS

We used 2-D DIGE to compare the proteomes of Manchurian, black, green, and white ash. A total of 2,434 spots were detected in the master gel (Figure 3.2). The internal standard image of gel 10 (Figure 3.2 and Table 3.1) was chosen as the master gel image that consisted of equal parts of protein extracts from each individual biological replicate from all four species of ash. An average of 2,184 (± 71 spots [95% confidence interval]) spots were detected in all 16 gels. We were able to resolve 1,733 (± 100 spots [95% confidence interval]) of the 2,434 spots detected in the master gel (P < 0.01) for all 16 gels, each containing two separate biological replicates plus the internal standard (Figure

3.2). The 1,733 spots that were matched to the master gel image on each gel were used for all subsequent statistical analyses.

Principal Component Analysis

Principal component analysis (PCA) of proteins found to differ (P < 0.05) in abundance, and with an absolute abundance ratio greater than 2-fold (n = 8 for each species), revealed a clear separation between species (Figure 3.4). White and green ash co-localized in the same region of the PCA grid while individual replicates grouped together by species. Black and Manchurian ash co-localized in the same plane (PC1) but were separated by PC2. PC1 and PC2 combined to account for nearly 64% of the original variance, with 46.5% explained by PC1 and 17.3% explained by PC2. Results of

Figure 3.4. PCA analysis of ash phloem proteomic profiles. PC1 clearly separates black (!) and Manchurian (!) ash from white (!) and green (!) ash, while PC2 separates Manchurian and black ash but not white and green ash. Each color point represents a single biological replicate (n=8/species).

Table 3.2. Number of proteins differentially expressed between North American ashes and Manchurian ash. Differences between species reflect phylogenetic relatedness (Figure 3.1).

Number of Proteins Comparison Significantly Differenta White vs. green 215

Manchurian vs. black 355

White vs. black 545

White vs. Manchurian 580

Green vs. black 589

Green vs. Manchurian 610 a > 2-fold absolute difference and p < 0.01.

the principal component analysis suggest that DIGE is a robust methodology to make interspecific proteomic comparisons.

Proteomic Differences between Manchurian and Black, Green, and White Ash

A total of 355 proteins were found to differ (p < 0.01) between Manchurian and black ash with an absolute abundance ratio greater than 2-fold (Table 3.2). Of these, 178 proteins had a higher level of differential expression in black ash (131 proteins ranging between 2

- 5-fold and 47 proteins with a > 5-fold difference), while 177 had higher levels of expression in Manchurian ash (126 proteins ranging between 2 - 5-fold and 51 proteins with a > 5-fold difference). A total of 264 of the initial 355 proteins of interest could be reliably matched from the preparative gels to the master gel image and were included in the picking list generated for nano-LC-MS/MS analysis. Based on the following criteria, we excluded 147 of the 264 identified proteins of interest from the comparison between black and Manchurian ash based on: 1) multiple protein identities were detected for a single spot, 2) protein identities assigned to a protein spot did not meet our criteria for a reliable match based on MASCOT and statistical data, or 3) no protein identity was assigned for the spot in question due to the lack of a sufficient match in the MASCOT database search. We also incorporated data related to the comparison of Manchurian vs. green, Manchurian vs. white, green vs. Manchurian, and white vs. Manchurian ash species to further eliminate an additional 65 proteins based on no significant difference of differential expression when compared to these species. Using the comparisons of

Manchurian against green and white we identified 33 proteins that had a significantly

higher (P < 0.05; > 2-fold) expression when compared across all species (Table 3.5). We used similar criteria to identify 19 proteins present in the 3 susceptible species with a > 2

–fold level of differential expression (P < 0.05) when compared to Manchurian ash

(Table 3.6). These eliminations resulted in a total of 52 proteins from the initial 355 proteins of interest [33 significantly (P < 0.01; > 2-fold average ratio) more abundant in

Manchurian ash and 19 significantly more abundant in black, green, and white ash] were included in the final analysis (Table 3.5 and Table 3.6).

Protein Identification, Classification, and Gene Ontology Annotation

Proteins were classified according to gene ontology (GO) annotations derived from graphical representations of biological process information using GO Slim (Table 3.5 and

Table 3.6). A limited subset of high-level GO terms (9 total categories and a miscellaneous category based on information obtained from the QuickGO resource) was used to organize the proteins identified in this study as they related to one another based on their GO annotation relationships (Table 3.5 and Table 3.6). Manchurian ash proteins

502 and 1481 had no associated biological process GO term, but were associated with hydrolase and lactoylglutathione lyase activity for molecular function GO terms, respectively (Table 3.5). Black/green/white ash had two proteins (1032 and 1722) with no associated GO term for any category. The remaining miscellaneous protein (975) was associated with ATP binding activities (Table 3.6).

Identification of Putative Defense and Susceptibility-Related Genes

The criteria used to compare ash species were similar to those of a previous DIGE study that compared the proteomes of two Entamoeba species (Davis et al. 2009). An absolute abundance ratio cutoff of 5-fold (with P < 0.05) between Manchurian and black, green, and white ash resulted in the selection of a total of 13 proteins present in

Manchurian ash phloem as putative constitutive resistance related proteins. Information relating to GO annotations for biological process were used to identify proteins with a putative role in plant defense, based on the current literature and reduced the selection to four proteins (1510, 1537, 1627, and 2096) (Table 3.3 and Table 3.4). Conversely, two proteins (430 and 1962) fit these criteria for the green, white, and black ash/Manchurian ash comparison (Table 3.4 and Table 3.6). These proteins were selected for further discussion below. The four proteins we identified (PR-10, aspartic protease, phenylcoumaran benzylic ether reductase, and ascorbate peroxidase) with potential for a direct or indirect role in resistance to emerald ash borer are discussed below (Table 3.3).

There was very little variation in protein expression among individual biological replicates of Manchurian, black, green, and white ash for these four proteins with

Manchurian ash having significantly higher quantities that when compared to the three susceptible ash species (p < 0.001) (Figure 3.5).

DISCUSSION

In this study we describe the first analysis of phloem tissues from a tree species using an interspecific comparative proteomics approach for the purpose of identifying potential constitutive resistance/susceptibility-related genes against an invasive insect pest.

Overall, interspecific variation in the phloem proteome corresponded strongly to phylogenetic relationships between species (Figure 1.1, Figure 3.4, and Table 3.2). Using this approach, we identified four proteins that may play a role in the resistance of

Manchurian ash to emerald ash borer: a major allergen (PR-10), a putative aspartic protease, a phenylcoumaran benzylic ether reductase, and a thylakoid-bound ascorbate peroxidase. Using the same criteria to select proteins of interest from black ash, we identified two putative susceptibility-related genes.

Proteomic Analysis of Non-Model Plants and Inter-species Comparisons Using DIGE

Protein sequences are highly conserved across organisms. This offers a major advantage for the high-throughput identification of gene products of non-model plant species via comparison to well known protein orthologs (Liska and Shevchenko 2003).

Furthermore, monitoring changes in global gene expression is emerging as an important tool for dissecting the molecular basis of plant interactions with other organisms

(Bhadauria et al. 2009). Therefore, studying proteins on a global scale using a proteomic approach can give insight into specific biological processes taking place in an organism or tissue at any one point in time (Carpentier et al. 2008). Proteomic analysis of non-

model plants is necessary to understand specific features and processes that are unique to certain plant systems which cannot be answered by model organisms (Carpentier et al.

2008).

Proteomic analysis has been applied successfully to several woody plant species in order to understand fundamental processes about wood formation, mechanisms governing fruit ripening, and responses to abiotic stress (Gion et al. 2005, Kieffer et al. 2008, Nilo et al. 2009). However, only a few studies have used proteomic methods (conventional 2-

DE or DIGE) to compare different species (Brobey et al. 2006, Davis et al. 2009). As others have noted, the challenges associated with cross-species analysis using a two- dimensional approach are not trivial. Amino acid sequence substitutions resulting from genomic differences between two closely related species, as well as post-translational modifications, splice variants, insertions, etc. could affect the location of a protein on a two dimensional plane via differences in their molecular weight and isoelectric point

(Davis et al. 2009).

Further challenges in comparing the proteomes of ash species emerge from their lack of genomic information. None-the-less, we did find that the total proteomic differences between ash species (Table 3.2) bear a strong correspondence with phylogenetic relationships (Figure 1.1) (Wallander 2008). Furthermore, PCA analysis revealed a clear separation between species that also corresponded with their evolutionary relationships

(Figure 3.4). In order to detect differences between species, a more conservative approach to visualizing differential expression is required. For instance, we focused on proteins that showed at least a 2-fold difference in expression between species for PCA

(Figure 3.4) and total proteomic analysis (Table 3.2), as compared to the 1.5-fold

difference criterion employed in a typical DIGE study (Kieffer et al. 2008). Clearly,

DIGE efficiently detected proteomic differences between species. However, ascribing functional roles to individual proteins is more difficult. Proteins known to contribute to virulence in humans were identified by comparing the constitutive proteomic differences between a virulent and a closely related non-virulent protozoan species using DIGE

(Davis et al. 2009). The identification of these differentially expressed proteins was accomplished consistently when the authors focused only on proteins with much higher levels of expression (5-fold difference and above, P < 0.01) in the pathogenic species when compared to the non-pathogenic relative. By using the same criteria, we were able to identify four unique genes from Manchurian ash with a potential role in defense against the emerald ash borer and two in susceptible ash species that may contribute to their susceptibility.

Putative Resistance-Related Genes in Manchurian Ash

Major Allergen

The major allergen (PR-10) had the highest average ratio for any protein found in

Manchurian ash when compared to the susceptible ash species. The major allergen from apple (Malus domestica L. Borkh.), Mal d 1 which best matched the protein from

Manchurian ash, is related to the birch family of allergens (Bet v 1). Mal d 1 is a pathogenesis-related (PR) protein, which suggests a potential role in plant defense against microbial attack and stress tolerance (Breiteneder et al. 1989, van Loon et al. 2006).

While classification of proteins as PR requires induction in response to microbial attack or related phenomena, many PR proteins are expressed constitutively in plant tissues, i.e. with no association to biotic attack (Liu and Ekramoddoullah 2006, van Loon et al.

2006). While PR-10 proteins are induced in response to a wide range of plant pathogens

(Puhringer et al. 2000, Robert et al. 2001, Park et al. 2004), their role in host resistance to insects has not been studied. However, PR-10 protein expression appears to be regulated by methyl jasmonate (MeJA) in several plant species (Wang et al. 1999, Rakwal et al.

2001). MeJA, and ultimately jasmonic acid-isoleucie conjugates (JA-Ile), regulate an important signaling pathway in the elicitation of induced resistance to herbivorous insects

(Howe and Jander 2008). In Manchurian ash, MeJA also has been shown to mediate emission of volatile compounds, suggesting the MeJA pathway is active and potentially regulates defense responses in this species (Rodriguez-Saona et al. 2006). PR-10 proteins have a diverse array of biological functions that include antimicrobial (Flores et al. 2002), ribonuclease (Koistinen et al. 2002), ligand-binding activities (Puehringer 2003), and involvement in secondary metabolism (Koistinen et al. 2005). The very high differential expression of constitutive PR-10 in Manchurian ash phloem relative to black ash suggests a potential role, either direct and/or indirect, in resistance of Manchurian ash to emerald ash borer.

Phenylcoumaran Benzylic Ether Reductase

Phenylcoumaran benzylic ether reductases (PCBER) are enzymes involved in neo- lignan biosynthesis (Gang et al. 1999). Lignoids (lignans and neo-lignans) are a class of

phenolic metabolites found throughout the plant kingdom with documented roles in plant defense (Strack 1997, Garcia et al. 2000). In relation to insects, lignans are known to have feeding and growth inhibition activities as well as toxicity against insects (Cabral et al. 2000, Garcia et al. 2000, Miyazawa et al. 1994). The presence of lignans in Fraxinus spp. has been documented extensively (Kostova 2001, Eyles et al. 2007, Kostova and

Iossifova 2007). More specifically, lignans were found to be much more highly concentrated in the phloem tissues of Manchurian ash when compared to green and white ash (Eyles et al. 2007, Cipollini et al. 2011). PCBER accumulates in the cambial region of young stems of Forsythia intermedia (a member of the Oleaceae, and relative to the genus Fraxinus) and has been implicated as serving dual functions as synthesizing key components for plant growth and active defense (Hillis 1987, Burlat et al. 2001). PCBER also acts directly as a food allergen and as a result is classified as being related to pathogenesis-related proteins (Karamloo et al. 2001). Based on the high level of expression of PCBER in Manchurian ash phloem and the current literature regarding the direct and indirect functions in plant defense make PCBER a very good candidate for future functional characterization as it relates to resistance against emerald ash borer.

Aspartic Protease

The putative aspartic protease we identified was consistently more highly expressed in Manchurian ash phloem than in the North American species of ash. Aspartic proteases have been found in all kingdoms of life, but our understanding of their biological roles derives mostly from microbes and animals (Davies 1990, Rawlings and Barrett 1995). In

plants, only serine proteases are more abundant than aspartic proteases (Simoes and Faro

2004). The latter have been found in monocots, dicots, and gymnosperms (Bourgeois and Malek 1991, Sarkkinen et al. 1992, Guevara et al. 1999), where they are typically associated with distinct organs, depending on species (Cordeiro et al. 1994, Guevara et al.

2001). Aspartic proteases in plants are involved in protein processing and degradation, senescence, stress responses, programmed cell death, reproduction, and antimicrobial defenses (Davies 1990, Rawlings and Barrett 1995, Mutlu and Gal 1999, Guevara et al.

2004, Simoes and Faro 2004, Ge et al. 2005). In potato (Solanum tuberosum L.) tubers and leaves, aspartic proteases display dose-dependent antimicrobial activity and are induced in response to infection by Phytophthora infestans and mechanical wounding

(Guevara et al. 1999, Guevara et al. 2001, Guevara et al. 2002, Guevara et al. 2004). In corn, cysteine protease has been shown to confer resistance to fall armyworm

(Spodoptera frugiperda) via degradation of the peritrophic membrane of this chewing insect, which interferes with nutrient acquisition, ultimately killing the insect (Jiang et al.

1995, Pechan et al. 2002, Mohan et al. 2006). Aspartic proteases are therefore extremely diverse in their roles in other biological systems and may participate in ash defense against emerald ash borer with mechanisms similar to those played by cysteine proteases in corn, or others yet to be characterized.

Thylakoid-Bound Ascorbate Peroxidase

Resistance of Manchurian ash to the emerald ash borer may also be mediated by ascorbate peroxidase. In plants, ascorbate peroxidases scavenge radical oxygen species

during photosynthesis. Herbivory is known to induce accumulation of H2O2, which can play a role in defense (Orozco-Cardenas and Ryan 1999), e.g. through direct toxicity (Bi and Felton 1995), or indirectly by serving as a secondary signaling molecule in the induction of defense genes (Alvarez et al. 1998). Enzymes that scavenge H2O2, e.g. catalases and peroxidases, can also be induced to higher levels upon attack by insect herbivores (Maffei et al. 2006). Ascorbate peroxidases reduce H2O2 to water (Asada

1999), and can simultaneously oxidize phenolic compounds to quinones. This process, known as the browning reaction (Urs and Dunleavy 1975), is thought to inhibit insect feeding (Felton et al. 1992, Dowd 1994). Quinones can also cross-link with other compounds such as proteins (Markwalder and Neukom 1976), rendering them less digestible to insects (Felton et al. 1992). Therefore, high levels of constitutive peroxidase activity in phloem tissue of Manchurian ash phloem may predispose it to respond more effectively when exposed to emerald ash borer attack.

Conversely, significant underexpression of the proteins listed in Table 3.4 in resistant

Manchurian ash compared to susceptible North American ash species may contribute to susceptibility. However, based on the current literature, we cannot hypothesize what mechanisms these genes may govern that would result in enhanced susceptibility to the emerald ash borer, and therefore require further investigation.

Conclusions

Plants resist herbivores through complex combinations of constitutive and induced defenses (Rasmann and Agrawal 2009). This is the first study to identify constitutive

proteins (a PR-10 protein, a phenyl-coumaran benzylic ether reductase, an aspartic protease, and a thylakoid-bound ascorbate peroxidase) that are strongly associated with the resistant Manchurian ash. Relative expression levels of the four proteins of interest from Manchurian ash show very little variation among individual biological replicates

(Figure 3.5). Functional analysis of these genes will be the next step to fully characterize their potential role in resistance against the emerald ash borer. This can be achieved with two separate but complementary approaches. First, the exact genetic sequences must be identified through genomic or transcriptomic approaches coupled with information on protein sequence data obtained from this study. Functional characterization of these genes can then be accomplished by transforming susceptible ash species with the gene of interest. (Transformation protocols are now available for green ash (Du and Pijut 2009)).

Second, development of an artificial diet for emerald ash borer (Keena et al. 2009) will provide a vehicle to test putative defense molecules, including the proteins identified in this study, directly against emerald ash borer larvae. Ultimately, discovery of genes associated with emerald ash borer resistance in coevolved Asian ash species will accelerate the development of resistant North American ash trees, similarly to the development of blight resistant American chestnuts (Castanea dentata Marsh), which have been produced using both conventional breeding methods and transgenics (Merkle et al. 2007). Introgression of Asian resistance genes into susceptible North American ash species, either via hybridization or transgenics, will accelerate the generation of resistant genotypes for restoration of forested and urban ecosystems that have been severely impacted by the emerald ash borer invasion.

ACKNOWLEDGEMENTS

Funding for this study was provided jointly by the USDA Forest Service and the USDA

Animal and Plant Health Inspection Service (08-8100-1288-CA) and by State and

Federal funds appropriated to the Ohio Agricultural Research and Development Center,

The Ohio State University. We thank David S. Bienemann for help with establishing and maintaining ash trees in Bowling Green, OH; Nathan Kleczewski, Karla Medina-Ortega,

Gerardo Suazo, and Duan Wang for technical assistance; and Sourav Chakraborty for expertise on mass spectral data analysis and a pre-submission review. We are grateful to

Dr. Don Cipollini, Dr. David Denlinger, Dr. Omprakash Mittapalli, Dr. Eva Wallander, and Dr. Guo-liang Wang for pre-submission reviews of this paper.

Table 3.3. Phloem proteins putatively related to defense identified in Manchurian ash.

Protein Average Ratio (p-value)* MASCOT NCBI Protein ID Annotation‡ Master Score Accession Gel M/B M/G M/W [No. of Number† Number peptides]

2096 49.85(7.10E-11) 56.74 (2.1E-10) 35.31 (3.0E-9) 76 [1] gi|886683 Major Allergen [Malus x Defense Response domestica]

1510 33.57 (1.6E-9) 25.6 (1.4E-9) 27.44 (6.0E-9) 435 [9] gi|7578895 Phenylcoumaran benzylic ether Metabolic Process reductase Fi1 [Forsythia x intermedia]

1537 14.52 (6.50E-11) 31.42 (1.3E-11) 16.28 (6.3E-9) 78 [2] gi|13897888 Putative Aspartic Protease Proteolysis [Ipomoea batatas]

-09 -11 -9 83 1627 10.15 (2.60E ) 10.03 (6.5E ) 9.18 (2.5E ) 110 [2] gi|25992557 Thylakoid-Bound Ascorbate Hydrogen Peroxide Catabolic

Peroxidase [Triticum aestivum] Process

*Proteins present in higher abundances (> 5-fold) in Manchurian ash relative to black, green, and white ash. †Searching NCBI Peptidome using the NCBI accession number of the matched protein will lead to detailed information about the peptides identified in this study. Manchurian ash peptide information can be obtained through Peptidome sample accession number PSM1314. ‡Annotations are Gene Ontology annotations for Biological Process.

Table 3.4. Phloem proteins putatively related to susceptibility identified in black, green, and white ash.

Protein Average Ratio (p-value)* MASCOT NCBI Protein ID Annotation‡ Master Score Accession Gel B/M G/M W/M [No. of Number† Number peptides]

1962 15.13 (1.2E-8) 7.06 (1.3E-7) 5.34 (2.0E-7) 108 [3] gi|147815877 Hypothetical protein [Vitis Protein Folding vinifera]

430 20.36 (1.3E-8) 15.57 (1.5E-9) 11.53 (2.0E-8) 755 [14] gi|147809607 Hypothetical protein [Vitis Proteolysis vinifera]

*Proteins present in higher abundances (> 5-fold) in black, green, and white ash relative to Manchurian ash. †Searching NCBI Peptidome using the NCBI accession number of the matched protein will lead to detailed information about the peptides identified in this study. Black, green, and white ash peptide information can be obtained through Peptidome sample

84 accession number PSM1313. ‡Annotations are Gene Ontology annotations for Biological Process.

Table 3.5. Proteins identified by MS and MASCOT analysis from Manchurian ash with an average ratio of 2 or greater when compared to black, green, and white ash.

d e Protein NCBI Protein Average Ratio (P-value) MASCOT Score Master Accession name/Speciesc a b Number Number M/B M/G M/W Biosynthetic Processf 686 gi|121044462 Granule-bound 24.2 (2.10E-8) 8.78 (4.8E-9) 6.84 (3.5E-9) 224 starch synthase I [Lycium pumilum]

Folic Acid and Derivative Biosynthetic Processg 1381 gi|157358403 Unnamed protein 10.95 (4.60E-7) 25.51 (2.6E-10) 8.4 (4.4E-7) 133 product [Vitis vinifera]

Purine Ribonucleoside Salvage 1291 gi|147812626 Hypothetical 9.16 (5.70E-11) 7.17 (1.6E-8) 8.6 (1.8E-9) 169 protein [Vitis vinifera]

Carbohydrate Metabolic Process 322 gi|157341193 Unnamed protein 2.92 (1.20E-5) 7.98 (5.6E-10) 6.75 (7.2E-10) 120 product [Vitis vinifera]

Glycolysis 769 gi|3023685 Enolase (Alnus 2.97 (1.40E-5) 4.77 (1.2E-8) 3.44 (3.8E-7) 742 glutinosa)

787 gi|3023685 Enolase (Alnus 12.05 (6.50E-11) 33.21 (3.3E-13) 11.56 (6.6E-12) 908 glutinosa)

Mannose Metabolic Process

326 gi|157343878 Unnamed protein 4.37 (2.60E-8) 11.76 (2.4E-10) 9.74 (2.0E-10) 124 product [Vitis vinifera]

Polysaccharide Catabolic Process 662 gi|147785379 Hypothetical 4.7(1.2E-4) 6.73 (1.0E-6) 4.65 (9.5E-5) 263 protein [Vitis vinifera] Cellular Process Cell Redox Homeostasis 440 gi|225459587 Hypothetical 3.71 (1.40E-5) 3.05 (1.1E-6) 5.58 (3.7E-7) 85 protein [Vitis vinifera]

546 gi|145666464 Protein disulfide 4.72 (3.40E-9) 6.02 (3.6E-11) 5.77 (9.6E-11) 268 isomerase [Zea mays]

86 Malate Metabolic Process

1345 gi|126896 Malate 4.25 (1.10E-5) 2.64 (1.1E-6) 3.52 (2.0E-8) 347 dehydrogenase

434 gi|228412 Malic enzyme 2.18 (5.9E-3) 5.98 (6.7E-7) 3.55 (3.3E-7) 433

437 gi|228412 Malic enzyme 4.91 (6.10E-8) 5.13 (3.4E-7) 4.37 (6.5E-7) 366

444 gi|228412 Malic enzyme 9.05 (8.70E-6) 13.22 (1.6E-9) 12.46 (1.5E-10) 411

Cellular Amino Acid Metabolic Process L-Serine Biosynthetic Process 536 gi|15235282 EDA9 (embryo 2.16 (4.10E-8) 4.9 (1.4E-11) 5.63 (3.5E-9) 169 sac development arrest [Arabidopsis thaliana]

Methionine Biosynthetic Process 958 gi|118488207 Unknown 5.62 (8.70E-11) 2.04 (1.1E-7) 2.64 (9.8E-9) 74 [Populus trichocarpa]

Metabolic Process 1485 gi|7578895 Phenylcoumaran 4.78 (3.40E-9) 8.6 (1.2E-8) 4.38 (8.3E-8) 148 benzylic ether reductase homolog Fi1 [Forsythia x intermedia]

1508 gi|4731376 Isoflavone 7.82 (1.80E-5) 2.51 (0.002) 2.01 (0.01) 145 reductase homolog Bet v 6.0101 [Betula pendula] 87

1510 gi|7578895 Phenylcoumaran 33.57 (1.60E-9) 25.6 (1.4E-9) 27.44 (6.0E-9) 435 benzylic ether reductase homolog Fi1 [Forsythia x intermedia]

1818 gi|225440390 Hypothetical 12.29 (5.30E-9) 4.86 (5.4E-9) 3.85 (7.4E-8) 323 protein [Vitis vinifera]

Oxidation Reduction 1045 gi|71793966 Alcohol 2.31 (4.10E-5) 3.6 (6.7E-9) 4.2 (7.7E-9) 142 dehydrogenase [Alnus glutinosa]

1232 gi|157352052 Unnamed protein 2.31 (7.30E-5) 3.34 (7.8E-7) 5.23 (6.4E-7) 96

product [Vitis vinifera]

991 gi|117067068 Monodehydroasco 2.33 (9.90E-8) 3.58 (2.6E-11) 4.33 (2.5E-8) 540 rbate reductase [Acanthus ebracteatus]

Photosynthesis 1758 gi|131385 Oxygen-evolving 3.26 (8.70E-6) 2.52 (2.4E-6) 13.4 (1.6E-10) 322 enhancer protein 1

Reductive Pentose-Phosphate Cycle 1225 gi|1173347 Sedoheptulose- 6.92 (1.70E-7) 9.07 (8.2E-10) 5.7 (4.2E-11) 300 1,7- bisphosphatase, chloroplastic

Protein Metabolic Process Proteolysis 1537 gi|13897888 Putative aspartic 14.52 (6.50E-11) 31.42 (1.3E-11) 16.28 (6.3E-9) 78 protease [Ipomoea batatas]

Response to Stress Defense Response 2096 gi|886683 Major allergen 49.85 (7.10E-11) 56.74 (2.1E-10) 35.31 (3.0E-9) 76 [Malus x domestica]

1317 gi|2765081 G5bf [Arabidopsis 2.45 (4.70E-6) 3.72 (4.2E-8) 2.7 (4.9E-7) 70 thaliana]

Hydrogen Peroxide Catabolic Process 1627 gi|25992557 Thylakoid-bound 10.15 (2.60E-9) 10.03 (6.5E-11) 9.18 (2.5E-9) 110 ascorbate

peroxidase [Triticum aestivum]

Response to Cold 1409 gi|157358403 Unnamed protein 10.08 (1.30E-9) 13.2 (9.3E-11) 10.78 (4.7E-10) 170 product [Vitis vinifera]

1469 gi|5923684 Putative 60S 8.61 (5.70E-11) 9.84 (1.1E-7) 26.76 (2.4E-10) 135 acidic ribosomal protein, 5' partial [Arabidopsis thaliana]

Miscellaneous 502 gi|157333098 Unnamed protein 2.99 (3.70E-8) 8.53 (4.9E-10) 5.28 (1.8E-8) 171 product [Vitis 89 vinifera]

1481 gi|2213425 Hypothetical 6.67 (1.50E-10) 5.98 (3.1E-10) 6.89 (4.7E-9) 127 protein [Citrus x paradisi]

a) Protein spot number matches the number to which the protein matches to the master gel spot map. b) Accession number corresponds to the protein identification obtained through the MASCOT database search. Searching NCBI Peptidome using the NCBI accession number of the matched protein will lead to detailed information about the peptides identified in this study. Manchurian ash peptide information can be obtained through Peptidome sample accession number PSM1314. c) Name of the protein identified through the MASCOT database search. d) Average ratio of protein abundance for Manchurian (M)/black (B), M/green (G), and M/white (W) ash and P-value of the two-tailed Student’s t-test for each protein spot comparison between individual species comparisons. e) MASCOT database score for peptide fragment matches to the database. f) GOA (gene ontology annotation) parent class appears in boldface type for broad categorization of overall protein biological function. g) GOA child terms appear in italics and refer to a specific biological function for certain proteins that also group under a specific parent term.

Table 3.6. Proteins identified by MS and MASCOT analysis from black, green, and white ash with an average ratio of 2 or greater when compared to Manchurian ash.

Protein NCBI Protein Average Ratio (P-value)d MASCOT Scoree) Master Accession name/Speciesc) Numbera) Numberb) B/M G/M W/M

Carbohydrate Metabolic Processf) Glycolysisg) 1176 gi|1022803 Phosphoglycerate kinase 3.04 (7.40E-5) 3.67 (1.1E-5) 2.75 (3.9E-5) 85 [Arabidopsis thaliana]

gi|533474 2-phospho-D-glycerate 2.15 (3.9E-4) 12.85 (2.2E-9) 2.13 (4.0E-4) 164 hydrolase

90 Cellular Amino Acid Metabolic Process Asparaginyl-tRNA Aminoacylation

386 gi|15223302 ATP binding / aminoacyl- 2.47 (1.4E-4) 2.94 (3.1E-6) 2.53 (1.0E-5) 138 tRNA ligase [Arabidopsis thaliana] Cellular Process Malate Metabolic Process 443 gi|1561774 Malate dehydrogenase [Vitis 3.33 (9.60E-7) 16.82 (3.3E-12) 9.77 (1.5E-11) 473 vinifera] Metabolic Process 1871 gi|38112662 Triose phosphate isomerase 18.07 (5.70E-11) 3.63 (2.4E-8) 11.26 (3.1E-11) 336 cytosolic isoform [Solanum chacoense] Oxidation Reduction 1572 gi|10334991 NADPH-dependent 2.47 (1.80E-5) 7.82 (1.2E-11) 2.81 (1.0E-6) 171 mannose 6-phosphate reductase [Orobanche ramosa]

Photosynthesis 1804 gi|21283 Unnamed protein product 4.56 (1.40E-15) 2.2 (5.8E-8) 2.36 (1.5E-7) 108 [Spinacia oleracea]

1687 gi|147791852 Hypothetical protein [Vitis 4.02 (5.30E-9) 52.53 (3.3E-13) 4.48 (1.7E-7) 496 vinifera]

Reductive Pentose-Phosphate Cycle 429 gi|1750348 Ribulose-1,5-bisphosphate 2.27 (2.80E-9) 6.33 (4.3E-7) 2.55 (6.5E-4) 86 carboxylase/oxygenase large subunit [Cosmelia rubra]

Photosynthesis, Light Harvesting 1992 gi|19184 Type I (26 kD) CP29 2.1 (6.80E-14) 6.72 (5.4E-8) 10.58 (4.7E-8) 304 polypeptide [Solanum lycopersicum]

91 1914 gi|671737 Chloropyll a/b binding 2.37 (1.10E-8) 3.05 (1.2E-7) 4.79 (1.1E-7) 117

protein [Amaranthus hypochondriacus]

Protein Metabolic Process Protein Folding 1962 gi|147815877 hypothetical protein [Vitis 15.13 (1.20E-8) 7.06 (1.3E-7) 5.34 (2.0E-7) 108 vinifera]

Proteolysis 360 gi|147797811 Hypothetical protein [Vitis 4.32 (3.40E-12) 4.32 (2.1E-6) 4.61 (8.2E-7) 597 vinifera]

363 gi|147797811 Hypothetical protein [Vitis 3.33 (4.10E-11) 2.22 (1.3E-4) 3.56 (1.9E-6) 445 vinifera]

430 gi|147809607 Hypothetical protein [Vitis 20.36 (1.30E-8) 15.57 (1.5E-9) 11.53 (2.0E-8) 755 vinifera]

Response to Stress Hydrogen Peroxide Catabolic Process 1935 gi|15223049 APX1 (ascorbate peroxidase 2.04 (3.9E-3) 2.77 (2.3E-6) 3.37 (2.6E-6) 113 1) [Arabidopsis thaliana]

Miscellaneous 975 gi|68565781 Ribulose bisphosphate 10.2 (6.30E-6) 2.17 (7.5E-3) 2.66 (3.0E-3) 646 carboxylase/oxygenase activase 2

1032 gi|146432257 GDP-mannose-3',5'- 2.16 (8.6E-4) 2.15 (6.2E-6) 2.12 (2.6E-5) 435 epimerase [Vitis vinifera]

1722 gi|83283979 Protein transport SEC13- 2.95 (1.80E-6) 4.4 (8.1E-9) 3.92 (2.7E-8) 107 92 like protein [Solanum

tuberosum]

a) Protein spot number matches the number to where the protein is located on the master gel spot map. b) Accession number corresponds to the protein identification obtained through the MASCOT database search. Searching NCBI Peptidome using the NCBI accession number of the matched protein will lead to detailed information about the peptides identified in this study. Peptide information can be obtained through Peptidome sample accession number PSM1313. c) Name of the protein identified through the MASCOT database search. d) Average ratio of protein abundance for black (B)/Manchurian (M), green (G)/M, and white (W)/M ash and P-value of the two-tailed Student’s t-test for each protein spot for individual comparisons against Manchurian ash. e) MASCOT database score for peptide fragment matches to the database. f) GOA (gene ontology annotation) parent class appears in boldface type for broad categorization of overall protein biological function. g) GOA child terms appear in italics and refer to a specific biological function for certain proteins that also group under a specific parent term.

CHAPTER 4

INTERSPECIFIC COMPARISONS OF ASH CONSTITUTIVE PHLOEM

PHENOLIC CHEMISTRY

ABSTRACT

Larvae of the emerald ash borer (Agrilus planipennis) feed on phloem of ash

(Fraxinus spp.) trees. Previous surveys and studies found that an Asian species,

Manchurian ash (F. mandshurica), is resistant to EAB while North American white (F. americana), green (F. pennsylvanica) and black ash (F. nigra), and the European species

F. excelsior are highly susceptible and experience high rates of EAB-induced mortality.

However, North American blue ash (F. quadrangulata) was colonized at much lower rates, comparable to that of Manchurian ash (Herms et al. unpublished). Divergence in resistant and susceptible phenotypes between ash species may be due in part to differences in phytochemical profiles of their phloem tissues, specifically phenolic chemistry, and provide an opportunity to explore the relationships between phloem phenolic metabolism and ash resistance to EAB.

The objective of this study was to characterize the constitutive phloem soluble phenolic chemistry and measure lignin levels (a phenolic polymer) of North American

(green, white, black, and blue), European (F. excelsior), and Asian (Manchurian) ashes

using RP-HPLC-PDA and MSn at two timepoints during the growing season that coincide with specific events in the EAB larval life cycle. We found qualitative similarities in the constitutive phloem phenolic chemistry of black cv. ‘Fallgold’, European ash seedlings, and Manchurian ash cv. ‘Mancana’ while individual compounds differed quantitatively.

Green ash cv. ‘Patmore’ and white ash cv. ‘Autumn Purple’ were similar in composition and quantity. The phenolic profile of blue ash, comprised mainly of hydroxycoumarins, was unique and showed only marginal similarity to the other species included in this study, as well as significantly higher concentrations of lignin at both time points. The similarities and differences in phenolic profiles of the ash species analyzed strongly correspond to their molecular phylogenetic relationships. The implications of this study on phenolic chemistry as it relates to resistance of Fraxinus spp. against EAB are discussed, with an emphasis on phylogenetic relationships.

INTRODUCTION

Resistance of deciduous trees to wood-boring, phloem feeding insects is hypothesized to be the result of a combination of constitutive and induced defense responses (Dunn

1990, Eyles et al. 2007). Constitutive traits comprise physical (e.g. outer bark texture or thickness) and chemical defenses (e.g. defense-related proteins and phytochemicals) which inhibit initial attack by the insect, in the latter case perhaps via direct toxicity or feeding deterrence (Mayer 2004, Wittstock and Gershenzon 2002, Koul 2008).

Allelochemicals such as soluble phenolic compounds and phenolic polymers (e.g. lignin) can have detrimental effects on growth and reproduction of insect herbivores that result in subpar performance or death for the invader (Wainhouse et al. 1990, Appel 1993,

Ruuhola et al. 2007, Swain 1979).

In recent studies comparing susceptible North American and resistant Asian ashes, constitutive phloem phenolic compounds identified in dormant stems of the resistant

Manchurian ash were found to be distinct from green and white ash. Specifically, the presence of hydroxycoumarins and phenylethanoid glycosides (calceolariosides A and B) set Manchurian ash apart (Eyles et al. 2007). A recent study (Cipollini et al. 2011) looking at the phenolic phloem profiles of green (wild and cultivated genotypes), white, and Manchurian ashes during phenologically relevant windows was essentially in agreement with the findings of Eyles et al. 2007.

These studies have identified a diverse array of constitutive phenolic metabolites in phloem tissues of Fraxinus species. Furthermore, (Kostova 2001, Kostova and Iossifova

2007) summarized the literature on Fraxinus chemistry published through the end of

2004. The review considered studies on all plant parts derived from any Fraxinus species and identified six main compound classes that are consistently associated with the genus, including: 1) phenolic acids/simple phenolics, 2) coumarins, 3) secoiridoids, 4) phenylethanoids, 5) lignans, and 6) flavonoids. Also of note is the presence of sterols, triterpenes, plant hormones, and pyrocatechol tannins (Perez-Castorena et al. 1997, Blake et al. 2002, Kostova and Iossifova 2007).

Such rich and diverse nature of chemical components of Fraxinus spp. requires a guided approach to studying the phytochemical differences between susceptible and resistant species in order to make biologically relevant predictions about potential defensive roles of compounds identified from the resistant Manchurian ash (Rebek et al.

2008). In addition to the potential defensive role of low-molecular weight phenolic metabolites, the phenolic polymer lignin is considered to be a quantitative defense trait and has been correlated with, or directly implicated in, resistance of some woody plants to certain insect pests (Coley 1986; Bryant et al. 1987; Wainhouse et al. 1990; Kurokawa et al. 2004).

We used four susceptible North American ashes and one susceptible European ash species– black ash cv. ‘Fallgold’, blue ash seedlings, green ash cv. ‘Patmore’, green ash seedlings, white ash cv. ‘Autumn Purple’, and European ash seedlings – and the resistant

Asian Manchurian ash cv. ‘Mancana’, to investigate the constitutive phloem phenolic profiles of healthy ash trees growing in the field. Trees were sampled at two separate time points corresponding to defined stages of the EAB larval life cycle. The first time point (early June) corresponds to when EAB eggs are laid on the outer bark of trees and neonate larvae begin to feed in phloem tissues. The second time point (early August)

corresponds to when EAB larvae are in their late 3rd to early 4th instars and have thus become established within host plants (Haack 2002). We used a metabolite profiling approach in which HPLC-PDA was coupled with LC-MS to compare phloem phenolics between and within all species and time points.

MATERIALS AND METHODS

Experimental Design and Sampling

For a detailed description of the research plot, please refer to Chapter 2. One sapling per block of each species (for a total of 8 biological-clonal replications per species) was sampled on June 2nd, 2008 and August 6, 2008. On June 2nd, 2008 Manchurian, black, green ‘Patmore’, green seedling, blue, white, and European ash trees had mean stem diameters of 3.1 ± 0.06 (s.e.m.) cm, 2.6 ± 0.06 cm, 3.0 ± 0.05 cm, 2.5 ± 0.1 cm, 1.7 ±

0.09 cm, 3.0 ± 0.1 cm, and 1.8 ± 0.08 cm respectively, at 15 cm above the soil line.

Phloem tissues from second year branches were chosen for all analyses. Branches were removed from trees, stripped of leaves, placed on ice, and then transported back to the lab where phloem tissue was immediately removed, frozen in liquid N2, and stored at -80°C until sample extraction. Samples were taken at times that correlate phenologically to the larval life cycle when EAB neonates are just beginning to feed on phloem tissues (June) and when larvae are in their late 3rd to early 4th instar (August) (Loerch and Cameron

1983).

Phenolic and Lignin Extraction

Phenolics were extracted according to Eyles et al. 2007. 100 mg of homogenized phloem tissue were extracted twice overnight in the dark at 4ºC with 500 µl of 100%

HPLC grade methanol (Fisher). Extracts were pooled and transferred into a 1.5 ml microcentrifuge tube and centrifuged at 12,000 g for 5 min to remove solids. Samples were then stored at -20ºC and used in subsequent HPLC analyses within three weeks following extraction. Lignin was extracted according to the methods of (Bonello and

Blodgett 2003). Pellets obtained from the soluble phenolics procedure were washed once each with 1 ml of Millipore water and HPLC grade methanol, and were then washed with

900 µl of tert-butyl methyl ether (Sigma) and dried overnight. Pellets were hydrolyzed at

40° C for 21 h on a shaker in 200 µl of 1 N NaOH. The reaction mixture was acidified with 200 µl of 1.5 M formic acid, followed by 400 µl of HPLC grade methanol. Pellets were then washed once with 1 ml of Millipore water. Pellets were re-suspended in 800 µl of 2 N HCl plus 300 µl of thioglycolic acid (Sigma). Tubes containing the solutions were incubated at 86°C for 4 h and the supernatants were discarded. Pellets were rinsed twice with 1 ml of water, resuspended in 1 ml of 0.5 M NaOH, and the tubes were placed on a shaker for 18 h. The supernatants were saved in a 2 ml tube and the pellets were again resuspended in 0.5 ml of 0.5 M NaOH for 18 h and the supernatants were pooled. The supernatants were acidified with 0.3 ml of concentrated HCl at 4°C for 4 h, the samples were centrifuged, and supernatants were discarded. The resulting pellets were air dried to complete dryness, and then dissolved in 1 ml of 0.5 M NaOH. In all steps, Eppendorf tubes were vortexed after adding solutions, and supernatants and pellets were separated

by centrifugation. Lignin concentration was determined spectrophotometrically against a standard of spruce lignin (Sigma-Aldrich, St Louis, MO).

Selection of Phenolic Compounds Using HPLC-UV for Qualitative, Quantitative, and

Statistical Analyses

HPLC-UV analyses of methanolic phenolic extracts were performed on an Alliance

2690 separation module (Waters, Milford, MA, USA) equipped with an autosampler and a 996 Photodiode Array Detector (Waters). The autosampler and column temperatures were set to 4 and 30ºC, respectively. Chromatographic separations were accomplished using a Waters Xterra! RP18, 5 µm, 4.6!150 mm column and a Waters Xterra! RP18,

3.9 µm, 3.0 ! 20 mm guard column. The binary mobile phase consisted of water/acetic acid (A) (98:2, v/v) and methanol/acetic acid (B) (98:2, v/v), with a flow rate of 1 ml/min. The elution program followed that of Eyles et al. 2007. The injection volume for all samples was 10 µl. Samples were passed through a Photodiode Array Detector

(PDA) (scanning range, 200-400 nm). Individual peak areas at 280 nm for all species and time points were used for all subsequent statistical analyses. Individual peaks were selected from each species at both time points by selecting peaks: 1) with a minimal absorbance greater than 0.02 AU based on a pooled sample from each species and time point (sample pools consisted of equal volumetric aliquots of methanolic extract from 8 individual biological replicates within a species), 2) that had a discernable UV spectrum, and 3) that were consistently detected in at least three individual biological replicates.

Analysis and Identification of Soluble Phenolics with HPLC-ESI-MS-PDA

Sample pools of phenolic extracts were analyzed on an HPLC-ESI-MS (Varian 500

MS; Palo Alto, CA, USA) in parallel with a PDA detector (Varian ProStar 335).

Chromatographic separations were accomplished using a Waters Xterra! RP18, 5 µm,

4.6!150 mm column. The binary mobile phase consisted of water/acetic acid (99.9:0.1 v/v) (eluant A) and methanol/acetic acid (99.9:0.1 v/v) (eluant B) with a flow rate of 1 ml/min. The gradient was as follows (percentages refer to proportions of eluant B): 5-

15% (0-15 min); 15-30% (15-35 min); 30-40% (35-40 min); 40-60% (40-50 min); 60-

90% (50-55 min); 90-100% (55-60 min). The injection volume for all samples was 10 µl.

Compound separation was achieved by post-column splitting (1:1) where half of the LC effluent was passed through the PDA Detector (scanning range, 200–400 nm) and the other half through an electrospray source (parameters below). The ESI was operated in both the negative and positive ion modes. The MS detector was optimized to obtain maximum yields of [M-H]- ions of esculin, luteolin, rutin, oleuropein, tyrosol, and verbascoside standards. The optimized MS parameters were: capillary voltage, -80 V; needle voltage, -5 kV (for the negative ion mode; the parameters remained the same in the positive mode except the polarity was reversed). The atmospheric pressure ionization

(API) parameters were as follows: API nebulizing gas (air) 25 psi; API drying gas

(nitrogen) 15 psi at 350° C. Survey scan was set to detect molecules between 50 – 1000 m/z. All species pools from both timepoints were run using both the TurboDDS and full scan mode in negative and positive ion modes. The TurboDDS trigger threshold was set to 5,000 counts (parent ion counts). In the TurboDDS scan mode, MS scan parameters

100

were set on-the-fly by the instrument to detect the most abundant parent ions and obtain maximum yields of those compound fragments. If TurboDDS was triggered, the mode temporarily switched to MSn mode to perform daughter scans of the putative parent ions.

The TurboDDS trigger threshold for daughter ions was 5,000, 500, and 50 ions for MS2,

MS3, and MS4 respectively. Full scan parameters were set following the conditions mentioned above. Full scan chromatograms were overlaid with PDA chromatogram

- traces at 280 nm to match [M-H] parent ions to "max of individual phenolic compounds.

In some cases compounds were not detected in the negative ion mode, and therefore the positive ion mode was used for compound identification (Tables 4.1 and 4.2). PDA data of individual compounds run on the HPLC-ESI-MS-PDA were analyzed using PolyView software (Varian) and then matched to the "max data generated from the UV analysis conducted on the Waters HPLC system in order to definitively identify peaks. Phenolic compounds were identified based on the congruence of parent and daughter ions, "max, retention time, and order in which compounds eluted from the stationary phase (Tables

4.1 and 4.2). Data acquisition and processing were all performed using MS Workstation

6.9.2 (Varian).

Identification and Quantification of Phenolics

The identities of esculin (peak 12, Table 4.1 and 4.2), esculetin (13), syringin (15), fraxin (18), fraxetin (22), 3-caffeoyl quinic acid (chlorogenic acid) (23), pinoresinol (41), verbascoside (47), oleuropein (53), quercetin diglycoside (rutin) and quercetin glycoside

(isoquercetin) (57), luteolin (67), and apigenin (68) were unequivocally confirmed by

101

spiking samples with the respective standards and matching retention times and UV maxima. Due to the lack of an external standard for the other compounds, samples were characterized and quantified following the methods of Eyles et al. 2007. Quantitation was achieved using HPLC-PDA peak areas at 280 nm. Following identification, individual compounds were quantified against external standards as mg g-1 FW. Exact matching external standards were not available for all compounds. Therefore, contents of individual compounds without an exact standard were expressed as: hydroxytyrosol equivalents for hydroxytyrosol hexoside (1); tyrosol equivalents for tyrosol hexoside (3) and tyrosol hexoside pentoside (11); vanillic acid equivalents for vanillic acid hexoside acetate adduct A (2), vanillic acid hexoside acetate adduct B (5), and unknown phenolic acid 1 (4); 3-caffeoyl-quinic acid (chlorogenic acid) equivalents for caffeoyl-quinic acid

A (8); fraxetin equivalents for fraxetin related compound (6); esculetin equivalents for esculetin A (7), esculetin B (10), esculetin C (13), and unknown coumarin 1 (14); esculin equivalents for escuside (49); fraxidin equivalents for fraxidin A (16), fraxidin B (19), and mandshurin (21); fraxin equivalents for fraxin related compound (56), unknown coumarin 2 (61), and unknown coumarin 3 (62); syringin equivalents for unknown monolignol 1 (17); pinoresinol equivalents for unknown lignan 1 (24), pinoresinol diglucoside + 2H2O (25), (+)-1-hydroxypinoresinol-4’-O-glucoside + 2H2O (28), unknown lignan 2 (30), unknown lignan 3 (32), and pinoresinol A (39); oleuropein equivalents for ligustroside A (27), oleuropein related compound 1 (29), ligustroside B

(31), unknown secoiridoid 1 (33), ligustroside C (35), 10-hydroxyoleuropein (38), dimethyloleuropein (40), oleuropein A (46), oleuropein B (50), demethylligstroside (51), oleuropein related compound 2 (58), ligustroside (59), and unknown secoiridoid 2 (64);

102

verbascoside equivalents for lugrandoside A (26), calceolarioside C (34), "- methoxylferrunginoside (42), calceolarioside A (44), lugrandoside B (45), verbascoside

A (48), calceolarioside B (52), verbascoside B (55), unknown phenylethanoid 1 (65), and verbascoside C (66); apigenin for apigenin glucoside (60); and luteolin for kaempferol galactoside (63). We were unable to quantify elenolic acid derivative 1 (9), caffeoylshikimic acid (37), and elenolic acid derivative 2 (54) due to a lack of relevant external standards. Comparisons of UV-spectral data with full scan fragmentation patterns led to the identification of 3 co-eluting compounds for peaks (11), (43), and (57)

(Tables 4.1 and 4.2).

Statistical Analyses

Principal component analysis (PCA) was used to investigate the overall relationships between phenolic metabolites and species through dimensionality reduction and feature extraction (Johnson and Wichern 1998). The PCA was performed using R (R

Development Core Team, 2010) for both species and metabolites. PCA were run twice

(Figures 4.1 and 4.5) in order to better visualize relationships between species. The second PCA was based on data obtained from a cluster analysis of species and metabolites following the first PCA. Cluster analysis was run to support the grouping of variables obtained in the PCA. The cluster analyses were carried out using the “pvclust” routine in the pvclust package in R (Suzuki and Shimodaira 2006). The pvclust package generates probability values for the clusters using the bootstrap resampling technique to assess their reliability (Efron et al. 1996a). The approximately unbiased (AU) confidence

103

value was computed using R (Suzuki and Shimodaira 2006). A total of 1000 bootstrap replications were generated for each cluster, and the AU confidence values were used to assess the uncertainty. The significance level of the clusters was set to 95. A larger confidence value is indicative of a “true” cluster (Efron et al. 1996b). In this study we also used biplots (common two dimensional display methods for exploring the data structure) to identify the relationship between species and metabolites. Biplots were also constructed using R.

Individual peak areas and mg g-1 FW concentrations for lignin were analyzed using univariate analysis of variance (ANOVA). Exploratory analyses of data and Levene’s test were used to evaluate variance equality and normality requirements of residuals.

Square-root and logarithmic transformations were made to meet normality requirements of residuals and homogeneity of variance. Any significant F-tests were followed by a protected LSD test (# = 0.05). All data were analyzed with IBM SPSS Statistics v. 19

(SPSS Inc. 2010).

RESULTS

Selection of Phenolic Compounds Using HPLC-UV for Qualitative, Quantitative, and

Statistical Analyses

We selected a total of 68 individual phenolic compounds from all species (including the green ash cultivar and seedlings) from the June and August samplings for detailed analyses (Tables 4.1 and 4.2, respectively). Comparison of ESI-MS/MS and PDA data

104

with the published literature, combined with confirmation of compounds using matching standards, revealed a diverse range of phenolic compounds present in the crude phloem extracts of the different ash species under investigation. Phenolic compounds identified in this study could be grouped into the following compound classes: phenolic acids/simple phenolics, hydroxycoumarins, lignans, secoiridoids, phenylethanoids, flavonoids, and coumarin-secoiridoids. All identities assigned in this paper are supported by the literature as it pertains to phenolic metabolites previously described for the genus

Fraxinus (Kostova and Iossifova 2007, Eyles et al. 2007). Most of the compounds described have previously been characterized – therefore we only present brief explanations regarding structural analysis for each compound and cite relevant studies with more detailed chemical information.

Simple Phenolics and Phenolic acids

Three compounds were classified as simple phenolics (1, 3, 11) and five compounds were classified as phenolic acids (2, 5, 8, 23, and 37). Compound (23) was unequivocally confirmed as 3-caffeoyl-quinic acid (chlorogenic acid). Simple phenolics hydroxytyrosol hexoside (1) and tyrosol hexoside (3) exhibited a [M-H]-, fragmentation patterns, and UV maxima consistent with identities from Eyles et al. 2007. Compound (11) exhibited a

[M-H]- of m/z 431 and UV spectral maximum at 275.3 nm. Fragmentation of [M-H]- resulted in a fragment of m/z 299 corresponding to the loss of a pentose moiety (132 Da).

Further fragmentation of m/z 299 resulted in a pattern matching that of tyrosol hexoside.

Therefore compound (11) was classified as tyrosol hexoside pentoside (Eyles et al. 2007,

105

Vallverdú-Queralt et al. 2010). Compounds (2) and (5) both exhibited a [M-H]- of m/z

389 and similar UV maxima (276.5 and 280.1 nm, respectively). Fragmentation of m/z

389 resulted in a dominant fragment of m/z 329, indicating a neutral loss of 60 Da, corresponding to an acetate adduct. Fragment ions of m/z 167 and m/z 161, combined with a dominant fragment of m/z 329, identify the compound as vanillic acid hexoside acetate adduct (Eyles et al. 2007, Jimenez et al. 2010). Compound (8) had a [M-H]- of m/z 353 and UV maximum at 326.4 and shoulder (sh) at 295 nm. Fragmentation of m/z

353 led to daughter ions of m/z 191 (162 Da loss of caffeoyl moiety) and m/z 179, which is consistent with caffeoyl-quinic acid (Eyles et al. 2007, Cardoso et al. 2005). We identified compound (8) as caffeoyl-quinic acid A (Cardoso et al. 2005). Compound (37) had a [M-H]- of m/z 335 and UV maximum at 327.6 and sh at 300 nm. Daughter ions of m/z 179 and m/z 135 were produced from the fragmentation of m/z 335. These data are consistent with those for caffeoylshikimic acid (Lin and Harnley 2008, Fang et al. 2002).

Coumarins

Twelve compounds were classified as coumarins at both time points across all species of ash (6, 7, 10, 12, 13, 16, 18, 19, 20, 21, 22, and 56). The coumarins esculin (12), esculetin (13), fraxin (18), and fraxetin (22) were unequivocally confirmed by spiking samples with standards and comparing [M-H]-, retention time, and UV maxima. As some previous studies have noted, the positive ion mode offers better sensitivity than the negative ion mode when identifying and detecting coumarin compounds from crude plant extracts (Yang et al. 2010). In our plant extracts, we were able to identify compounds (7,

106

10, 12, 13, 18, 20, and 56) in the negative ion mode, while compounds (6, 16, 19, 21, and

22) were only detected in the positive ion mode. We utilized both the negative and positive ion modes to maximize detection of coumarins in all of our crude extracts.

In the negative ion mode, we named compounds (7), (10), and (20) as esculetin A, esculetin B, and esculetin C, respectively. Each compound had a [M-H]- of m/z 177, fragment ions of m/z 133 and m/z 89, as well as UV maxima consistent with or similar to the esculetin standard, but with differing retention times from the esculetin standard.

Compound (56) gave a [M-H]- of m/z 545, which, following fragmentation, displayed a dominant signal of m/z 369 (neutral loss of 176). Fragmentation of m/z 369 gave fragments of m/z 207 and m/z 192 which are consistent with the fragmentation of the fraxin standard. The loss of 176 Da can correspond to several functional groups, therefore we identified compound (56) as a fraxin related compound.

In the positive ion mode, compound (6) exhibited a [M-H]- of m/z 385, with a dominant MS2 fragment of m/z 209. Further fragmentation of m/z 209 was consistent with the fraxetin standard. Therefore, compound (6) was identified as fraxetin related compound. Compound (16) and (19) were identified as fraxidin A and B due to a [M-H]- of m/z 223, fragmentation pattern, and UV spectra that were all consistent with the fraxidin standard, except for a difference in retention time. Compound 21 was identified as mandshurin as all data regarding compound structure were consistent with the identification proposed by Eyles et al. 2007.

107

Lignans

Four compounds were classified as lignans (25, 28, 39, and 41) and each displayed similar UV maxima consistent with those published in the literature and the pinoresinol

(41) standard. Compound (25) had a [M-H]- of m/z 717. While no compound matched to m/z 717 in our literature search, we did note a mass difference of 36 Da between compound (25) and pinoresinol dihexoside (Eyles et al. 2007), which likely corresponds to the presence of two water residues. Further investigation of compound (25) revealed similar RT and UV spectra as that of Eyles et al. 2007, as well as identical MS2 spectra

(m/z 519 and m/z 357) corresponding to the successive losses of two hexoside (162 Da) residues from the parent ion. Based on this evidence we tentatively identified compound

- (25) as pinoresinol dihexoside + 2 H2O. Compound (28) displayed a [M-H] of m/z 571 with daughter ions of m/z 373, 535, and 343. The daughter ions matched to (+)-1- hydroxypinoresinol-4’-O-glucoside (Guo et al. 2007) along with similar UV spectra. We noted a mass difference of 36 Da, corresponding to the loss of two water residues, between the parent ion (m/z 571) and (+)-1-hydroxypinoresinol-4’-O-glucoside, therefore we called compound (28) (+)-1-hydroxypinoresinol-4’-O-glucoside + 2 H2O. Compound

(39) had a [M-H]- of m/z 357 and dominant daughter ion fragments of m/z 151 and m/z

136 along with a UV maximum at 276.5 nm. These data were consistent with the pinoresinol standard except for a slight shift in the retention time and difference in UV maxima. We therefore tentatively identified this compound as pinoresinol A.

108

Monolignols

Compound (15) was classified as a monolignol and its identity was confirmed by comparison against the syringin standard. Compound (15) was detected in positive ion mode but not in negative ion mode, although it was previously detected in that mode by

Eyles et al. 2007. In the positive ion mode, compound (15) had a [M-H + Na]+ of m/z 395 and UV maximum at 264.7 nm. Fragmentation of the parent ion led to daughter ions of m/z 233 and m/z 217 and correspond to sodium adducts of the same fragments detected by Eyles et al. 2007 in the negative ion mode.

Secoiridoids

Thirteen compounds were classified as secoiridoids (9, 27, 31, 35, 38, 40, 46, 50, 51,

53, 54, 58, and 59). Oleuropein (53) was identified by comparison with the standard.

Compound (9) gave a [M-H]- of m/z 601 and was found to co-elute with compound (11).

Fragmentation of m/z 601 led to dominant daughter ions of m/z 403 and m/z 223 which gave the identification as elenolic acid derivative 1 and was previously described by

Eyles et al. 2007. Compounds (27), (31), and (35) all shared similar UV maxima, [M-H]- and fragmentation patterns, and matched to ligustroside (59) (Eyles et al. 2007).

Therefore compounds (27), (31), and (35) were tentatively named ligustroside A, ligustroside B, and ligustroside C, respectively. Compound (38) gave a [M-H]- of m/z

555 and dominant daughter ion of m/z 393 (loss of 162 Da) corresponding to the aglycone resulting from the loss of a hexose moiety. Compound (40) had a [M-H]- of m/z

109

525 and a dominant, decarboxylated MS2 fragment of m/z 481. MS2 fragments of m/z

389 and m/z 319 combined with the parent ion of m/z 525 were previously identified as being characteristic of dimethyoleuropein (Savarese et al. 2007). Compounds (46) and

(50) displayed UV maxima, [M-H]-, and fragmentation patterns that matched the oleuropein standard but with different retention times, and were therefore identified as oleuropein A and oleuropein B. Compound (51) exhibited a [M-H]- of m/z 509 and fragmentation gave a dominant [M-H]- of m/z 347, corresponding to the loss of a hexose

(162 Da). A shoulder at 278 nm was in agreement with the UV spectra observed for ligustroside and led to the putative identification of compound (51) as demethylligstroside (Takenaka et al. 2000). Compound (54) gave a [M-H]- of m/z 793.9 and fragmentation led to dominant daughter ions of m/z 403 and m/z 223, corresponding to an elenolic acid derivative similar to compound (9), which led to the identification of compound (51) as elenolic acid derivative 2. Compound (58) had a [M-H]- ion of m/z

553 and dominant daughter fragments of m/z 391, m/z 321, and m/z 289. These fragments display a mass difference of 14 Da from the [M-H]- (m/z 539) and main fragments (m/z 377, m/z 307, and m/z 275) of oleuropein, indicating the addition of a methyl group to oleuropein. Compound (58) also had a similar UV maximum to oleuropein and we therefore gave it a tentative identification of oleuropein related compound 2. Compound (59) was identified as ligustroside and was previously described by Eyles et al. 2007.

110

Phenylethanoids

A total of twelve compounds were identified as phenylethanoids (26, 34, 36, 42, 43,

44, 45, 47, 48, 52, 55, and 66). Verbascoside (47) was confirmed by direct comparison with a standard. Compounds (26) and (45) each exhibited a [M-H]- of m/z 639 and fragmentation led to successive fragments of [M-H – 162]- with m/z 477, which yielded characteristic fragments (m/z 315 and m/z 135) of another phenylethanoid, calceolarioside. Compounds (26) and (45) also had UV maxima at 329.9 nm and a shoulder at 295 nm, which led to the tentative identification of compounds (26) and (45) as lugrandoside A and B, respectively. Compounds (44) and (52) were previously described by Eyles et al. 2007 and were identified as calceolarioside A and B, respectively. Compound (34) had similar UV spectrum (maximum at 326.4 nm and a shoulder at 290 nm), [M-H]- (m/z 477), and dominant daughter fragments (m/z 315 and m/z 135) to compounds (44) and (52) and was therefore identified as calceolarioside C.

The characteristics of these three compounds were consistent with those found in the literature for calceolariosides (Eyles et al. 2007, Guo et al. 2007). Compounds (36) and

(43) exhibited a [M-H]- of m/z 785 and dominant daughter fragments of m/z 623 and m/z

461, which are consistent with the previously described forsythoside A-O-glucoside (Guo et al. 2007). Therefore we identified compounds (36) and (43) as forsythoside A-O- glucoside – A and forsythoside A-O-glucoside – B, respectively. Compound (42) had a

[M-H]- of m/z 507 along with notable daughter fragments (m/z 475, 323, 179, and 161) and UV spectrum that matched to the published description of $-methoxylferruginoside

(Guo et al. 2007). Finally, we identified compounds (48), (55), and (66) as verbascoside

111

A, B, and C, respectively, as each compound displayed similar UV spectra, [M-H]-, and fragmentation patterns as the verbascoside standard, with varying retention times.

Flavonoids

We identified five flavonoids in our species: (57, 60, 63, 67, and 68). The identities of compounds (57), (67), and (68) were confirmed by comparison with standards.

Compound (60) had a [M-H]- of m/z 431 with a dominant daughter ion of m/z 269.

Subsequent fragmentation of m/z 269 and the UV spectrum were identical to the apigenin standard. Therefore, we identified compound (60) as apigenin glucoside. Compound

(63) gave a [M-H]- of m/z 447 and daughter fragments of m/z 285, 255, 327, 227, and

211. UV maxima at 350 and 264 nm led to the tentative identification of compound (63) as kaempferol galactoside (Ye et al. 2005).

Coumarins-secoiridoids

Compound (49) was the only compound identified as a coumarin-secoiridoid in our study. The compound had a [M-H]- of m/z 725 and a dominant daughter ion [M-H –

386]- of m/z 339. A neutral loss of 386 Da indicates a secoiridoidal group. The daughter ion of m/z 339 fragmented in the same manner as the esculin standard. Therefore, compound (49) was identified as escuside, which is a combination of a secoiridoid and a coumarin (Iossifova et al. 2002).

112

Unknowns

Ten of the initial 68 compounds, selected based on HPLC-PDA chromatograms, could not be reliably matched to known species using MS data, or a putative identity was not found in our literature search and therefore did not receive a tentative identification.

However, each compound could be sorted into a respective compound class based on its

UV spectrum and retention time. Therefore compound (4) was identified as unknown phenolic acid 1; (14), (61), and (62) as coumarin 1, 2, and 3; (17) as unknown monolignol

1; (24), (30), and (32) as unknown lignans 1, 2, and 3; (33) and (64) as unknown secoiridoids 1 and 2; and (65) as unknown phenylethanoid 1. Grouping compounds into their respective compound classes allowed us to provide quantification based on equivalent standards based on UV spectra.

Qualitative and Quantitative Differences in Phenolic Profiles

We detected a total of 30 (black ash cv. ‘Fallgold’), 20 (blue ash seedlings), 22

(European ash seedlings), 18 (green ash cv. ‘Patmore’), 23 (green ash seedlings), 27

(Manchurian ash cv. ‘Mancana’), and 23 (white ash cv. ‘Autumn Purple’) compounds across both time points. Species differences were reflective of the phylogenetic relationships of the taxa (Tables 4.1-4.4). Compounds that were only detected in June include compounds: (4), (6), (8), (11), (17), (26), (30), (31), (32), (39), (42), (43), (45),

(50), (51), (57), and (66). Compounds only detected in August include: (9), (14), (20),

(27), (28), (48), (49), (54), (55), (56), (58), (61), (62), (64), and (65).

113

Ash species belonging to the section Fraxinus, i.e. black and European ashes, shared

77.8% and 63.0% of their compounds, respectively, with Manchurian ash. When compared with blue ash, a member of the section Dipetalae, Manchurian ash shared

40.7% of its compounds, but only 29.6% of compounds in Manchurian ash were shared with white, green ‘Patmore’, and green seedling, all of which belong to the section

Melioides. In comparison, blue ash only shared 20% of its compounds (1, 15, 38, and

41) with the Melioides section species. When comparing green ‘Patmore’ with green seedling and white ash, 89% and 83% of their compounds, respectively, were shared.

The difference between green ‘Patmore’ and green seedling was due to the presence of six metabolites [(5), (54), (55), (57), (64), and (67)] that fit our criteria for further analysis in green seedling, but were not in high enough quantities to merit further investigation in green ‘Patmore’. LC-MS full scan mode data did detect the presence of

[M-H]- corresponding to compounds (5), (55), and (57) in green ‘Patmore’ extracts, although at much lower quantities than in green seedling.

Blue ash was unique, localizing to its own PCA quadrant (Figures 4.1, 4.4 and 4.5) and forming its own cluster (Figure 4.2). Blue ash phloem is characterized primarily by the presence of hydroxycoumarins (Table 4.1 and 4.2), of which esculin (12) is particularly characteristic (Tables 4.3, 4.4 and Figures 4.3, 4.4). Only a few compounds were unique to the resistant Manchurian ash cv. ‘Mancana’, and concentrations of individual phenolics tended to be similar to those of susceptible black ash cv. ‘Fallgold’

(Tables 4.3 and 4.4). Compounds (25), pinoresinol dihexoside, and (62), unknown coumarin 3, were only identified in Manchurian ash cv. ‘Mancana’ phloem. Trees belonging to the Melioides, green ash cv. ‘Patmore’, green ash seedlings, and white ash

114

cv. ‘Autumn Purple’ were very different from the other species under investigation, but were virtually indistinguishable from one another qualitatively (Tables 4.1-4.4, and

Figures 4.1 and 4.5). Overall, compound concentrations tended to increase from June to

August across all species (Tables 4.3 and 4.4).

Principal Component, Cluster, and Biplot Analyses

The PCA for June showed similar trends and relationships as the August analysis, therefore, we focused on and discuss the analysis of August metabolites (Tables 4.2 and

4.4). The first two principal components accounted for 91.77% of the variation and revealed four clusters, one in each of the quadrants (Figure 4.1). Bootstrap analysis generally supported the four clusters (Figure 4.2). A separate cluster analysis based on the individual 68 compounds revealed that seven metabolites, esculin (12), syringin (15), fraxin (18), mandshurin (21), calceolarioside A (44), calceolarioside B (52), and oleuropein (53), clustered together with low AU values (Figure 4.3). Furthermore, a biplot analysis exploring the relationships among the species and the metabolites that were responsible for the individual clusters (Figure 4.4) showed that esculin (12) is the major variable separating blue ash from the other species; syringin (15) separates green

‘Patmore’, green seedling, and white ash from the other species; fraxin (18), mandshurin

(21), and oleuropein (53) separate European from Manchurian and black ash; and calceolarioside A (44) and calceolarioside B (52) are responsible for the separation of

Manchurian and black ash. Based on the results of the PCA, cluster analysis, and biplot analysis we explored the effect of removing the seven major metabolites on the species

115

Figure 4.1. PCA plot of phenolic data in six species of taxa. Numbers represent individual ash species: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon.

116

Figure 4.2. Approximately unbiased (AU) confidence value test from bootstrapping for clusters obtained using the first 2 PC scores from the six ash taxa. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group.

117

Figure 4.3. AU confidence value test for clusters obtained using the first 2 PC scores from metabolites. Numbers at the end of lines prefaced with A correspond to individual phenolic metabolites as listed in Table 4.1 and 4.2. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group.

118

Figure 4.4. Relationship between PC1 and PC2 scores obtained for six species of ash and 68 metabolites. Black dots with black numbers refer to individual ash taxa:1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon. Red arrows ending with an ‘A’ followed by a number correspond to individual phenolic metabolites as listed in Table 4.1 and 4.2. The biplot analysis visualizes the compounds that are the main drivers of the differentiation of the various taxa and clusters.

119

Figure 4.5. Relationship between PC1 and PC2 scores for individual peak areas for 61 compounds (excluding the most characteristic compounds that were associated with individual species – see Figure 4.2 and 4.4) identified in six species of ash. Numbers correspond to individual ash taxa: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Black dots represent the mean of eight biological replicates within a given taxon.

120

Figure 4.6. Approximately unbiased (AU) confidence value test from bootstrapping for clusters obtained using the first two PC scores from the six ash species based on 61 phenolic metabolites. Numbers at the end of a line correspond to individual ash species: 1 – black ash; 2 – blue ash; 3 – green ash seedling; 4 – European ash; 5 – green ash ‘Patmore’; 6 – Manchurian ash; and 7 – white ash. Numbers at nodes are AU values. Values > 95 indicate a significant cluster or group.

121

groupings. A second PCA was thus run which revealed results similar to the first PCA, but with some important differences (Figure 4.5). The first two principal components accounted for 90.75% of the variance and produced three clusters (Figure 4.5), which were consistent with the three sections of Melioides, Dipetalae, and Fraxinus. To test the statistical reliability of the clusters found in this PCA plot (Figure 4.5) we again used bootstrap analysis and obtained three highly supported clusters, again representing the three Fraxinus sections (Figure 4.6).

Lignin

Lignin concentrations were analyzed for all species at both time points (Tables 4.3 and 4.4). Lignin tended to increase in phloem tissues from June to August. Lignin was consistently more abundant in blue ash phloem than in the other species in the study

(Tables 4.3 and 4.4). European ash had the lowest concentration of lignin of any species in both June and August (Tables 4.3 and 4.4). White and Manchurian ash had consistently higher lignin concentrations than green ‘Patmore’ and green seedling. Black ash concentrations tended to be higher than the other species, with concentrations similar to blue ash.

DISCUSSION

Although our species comparisons were based on only one (nominal) genotype or cultivar for several species (black ash cv. ‘Fallgold’, green ash cv. ‘Patmore’, white ash

122

cv. ‘Autumn Purple’, and Manchurian ash cv. ‘Mancana’), multiple genotypes were represented in our seedling lots for blue, European, and green seedlings. Our data show that quantitative variation of individual compounds among different genotypes within species was similar to the variation observed within ash cultivars. Thus, we posit that all interspecific comparisons were sound. Detailed examination of qualitative and quantitative differences and similarities between species, along with our multivariate analyses (PCA, cluster analysis, and biplot analysis) based on the 68 phenolic compounds, show that patterns in the accumulation of constitutive phenolic compounds in Fraxinus spp. can be explained by their phylogenetic relationships (Figures 4.1-4.6 and

Tables 4.1-4.4).

Manchurian ash has a shared co-evolutionary history with EAB in Northeastern Asia and is resistant to infestation by EAB in the field (Rebek et al. 2008). In this study we have hypothesized, as have others, that the phenotypic differences in resistance of

Manchurian and susceptible North American ashes is the result of qualitative and quantitative differences in defense chemistry, specifically phenolics (Eyles et al. 2007,

Cipollini et al. 2011). Indeed Eyles et al. 2007 identified significant differences regarding constitutive phenolic profiles of dormant stem material from green, white, and

Manchurian ash. Hydroxycoumarins and two phenylethanoids, calceolariosides A and B, were only found in the phloem tissues of Manchurian ash and this led to the speculation that these compounds may be responsible for resistance against EAB. The results of another study (Cipollini et al. 2011) confirmed the uniqueness of certain Manchurian ash constitutive phloem phenolic metabolites compared to white and green ash cultivars, and green ash seedlings sampled during the growing season. While these studies have

123

thoroughly examined the constitutive phloem phenolic chemistry of Manchurian, green, and white ashes, more meaningful comparisons may have been overlooked. For instance, we observed qualitative differences among all ash species, but when viewed in the context of phylogenetic relationships the results were even more informative. A total of

27 compounds were identified in Manchurian ash phloem in this study, and the least number of shared metabolites was found between Manchurian ash and species belonging to the Melioides section (green and white ash). Black and European ash contain an equally diverse array of hydroxycoumarins previously thought to be unique to

Manchurian ash (Eyles et al. 2007). The phenylethanoid, calceolarioside A (44), was approx. 3-fold more concentrated in black than in Manchurian ash (Table 4.3) and was also detected in European and blue ash, albeit at much lower levels (Table 4.3).

Calceolarioside B (52) was also more highly concentrated in black than in Manchurian ash (> 2-fold higher; Table 4.3).

Overall, only two compounds identified in this study were found to be unique to

Manchurian ash phloem, (25) and (62) (Tables 4.1 and 4.2). Compound (25), pinoresinol dihexoside + 2H2O, was previously thought to be significantly more abundant in (Eyles et al. 2007), or unique to (Cipollini et al. 2011), Manchurian ash phloem. As these previous studies have noted, pinoresinol dihexoside is a type of lignan (phenylpropane dimers) (Strack 1997). The aglycone, pinoresinol, has been shown to have antifeedant and growth/molt inhibiting activities against several insect species (Miyazawa et al. 1994,

Cabral et al. 2000, Garcia et al. 2000). Compound (62) was tentatively identified as unknown coumarin 3 based on its UV spectrum. Further structural information is needed

124

before we can draw any conclusions regarding the potential role of this compound in

Manchurian ash resistance to EAB.

A recent proteomic analysis of ash phloem tissues showed that phenylcoumaran benzylic ether reductase (PCBER, an enzyme involved in the directed synthesis of lignans) is more highly expressed in constitutive phloem tissues of Manchurian ash compared to black, green, and white ash (Chapter 3). Thus, our approach to analysis of phenolics, and the separate but complementary proteomic approach, have resulted in the identification of lignan-related products that are unique to Manchurian ash phloem, suggesting a potential role of these compounds in resistance against EAB. Furthermore, a recent study described 531 phloem-specific ESTs, derived from five species of ash

(blue, black, green, white, and Manchurian ash), with functions related to the biosynthesis of phenylpropanoids (Bai et al. 2011). The database represents a rich resource to investigate actual function of some of the potential resistance traits we have identified.

Resistance of Manchurian ash is hypothesized to be the result of shared co- evolutionary history with EAB in its native China. However, no co-evolution can be hypothesized for North America native blue ash, even though resistance of this species to

EAB, or at least unpalatability, has been observed (Herms et al. unpublished; Anulewicz et al. 2007). EAB adults have also been observed to have a low feeding preference for blue ash foliage, similar to feeding rates observed on Manchurian ash foliage

(Pureswaran and Poland 2009). The potential of blue ash as germplasm for resistance genes has caused excitement and led some to hail blue ash as the logical choice to use in breeding for the introgression of resistance genes into the more susceptible North

125

American congeners (Anulewicz et al. 2007). However, based on the results of in depth phylogenetic analyses (Wallander 2008), phylogeograhical investigations on the origins of the genus Fraxinus (Jeandroz et al. 1998), and now our own phenolic metabolite analysis, it can be concluded that introgression of blue ash resistance traits into other

North American species may not yield the intended results. The putative resistance of blue ash may be the result of a complex mixture of traits that will not be easily introgressed into other native ash species because conventional breeding techniques will likely yield unsuccessful crosses, and transgenic approaches to introduce defense-related genes from blue ash into its North American congeners will not be possible until they are identified. Phylogenetic and phylogeographical studies have suggested that blue ash is either ancestral to all other ash species or split very early on as a paraphyletic section from all other sections in the genus (Wallander 2008, Jeandroz et al. 1998). Therefore, it is likely that blue ash resistance is a case of allopatric resistance, while Manchurian ash resistance represents sympatric resistance (Harris 1975). We suggest that the interpretation of ash resistance against EAB is better informed when phylogenetic relationships are taken into account. In practice this means that comparing Manchurian ash to phylogenetically similar, but susceptible congeners, such as black or European ash, is a more focused approach that will lead more quickly to biologically relevant conclusions, and that the introgression of resistance genes into North American black and

European ash from Manchurian ash may be achieved more easily by conventional or transgenic methods because they share a more recent common ancestor. Conversely, species belonging to the Melioides (green and white ash) are primarily found in North

America (Wallander 2008). Thus, the identification of allopatrically resistant individual

126

green and white ash genotypes will be important for combating EAB. Recent findings of

‘lingering ash’ suggest resistant individuals may exist, but further work is needed on the phenotypic, molecular, and biochemical differences between these individuals before any definitive conclusions can be made (Koch et al. 2010 Purdue symposium).

The identification of Manchurian ash-specific lignans and lignan derivatives as potential resistance traits (Eyles et al. 2007, Cipollini et al. 2011, Chapter 3) is now informing new investigations, including bioassays that will test the effect of this class of compounds directly on EAB larval performance.

ACKNOWLEDGEMENTS

We thank David S. Bienemann for help with establishing and maintaining ash trees in

Bowling Green, OH. We also thank Karla Medina-Ortega, Nathan Kleczewski, Duan

Wang, and Gerardo Suazo for technical assistance and Stephen Opiyo for statistical consultation.

127

Table 4.1. Phenolic compounds identified by HPLC-UV and LC-MS-PDA from samples collected on June 2nd, 2008 from susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash grown in a common garden in Bowling Green, OH.

g h Master Species in which RT - [M-H]- or Fragments m/z (in order ! max Putative ID Reference Peak peak detectedb HPLC-UVc [M-H]+d of decreasing (nm)f Numbera abundance)e 1 B, Q, E, G, GS, M, W 5.38 315 MS2: 135, 153, 179; MS3: 281.2 Hydroxytyrosol hexoside Cardoso et al. 107, 91, 117, 93, 79 2005; Eyles et al. 2007 2 B, M 6.5 389 MS2: 329, 167, 161 276.5 Vanillic acid hexoside acetate Eyles et al. adduct A 2007; Jimenez et al. 2010 3 G, E, GS, M, W 7.61 299 MS2: 179, 119, 143, 113, 275.3 Tyrosol hexoside Kammerer et 161, 131 al. 2005; Eyles et al. 128 2007

4 W 8.39 ND by LC- ND by LC-MS PDA 276.5 Unknown Phenolic Acid 1 - MS PDA 5 GS 9.08 389 MS2: 329, 167, 161 280.1 Vanillic acid hexoside acetate Eyles et al. adduct B 2007; Jimenez et al. 2010 6 B 9.24 385* MS2: 209; MS3: 149, 163, 280, shi Fraxetin related compound Yasuda et al. 181, 194; MS4: 121 327 2006 7 B, Q 9.65 177 MS2: 133 MS3: 89 257, Esculetin A Eyles et al. 289.5, 2007 340.7 8 E 9.59 353 MS2: 191, 179; MS3: 85, 327.6, sh Caffeoyl-quinic acid A Cardoso et al. 127, 173, 93 295 2005 10 Q 10.14 177 MS2: 133 MS3: 89 340, sh Esculetin B Eyles et al. 300, 2007 258.8 11 G, GS, W 10.1 431 MS2: 299, 191, 233, 251; 275.3 Tyrosol hexoside pentoside Eyles et al. MS3: 119, 143, 179, 131, 2007;

128

g h Master Species in which RT - [M-H]- or Fragments m/z (in order ! max Putative ID Reference Peak peak detectedb HPLC-UVc [M-H]+d of decreasing (nm)f Numbera abundance)e 149 Vallverdú- Queralt et al. 2010 12 B, E, M, Q 10.65 339 MS2: 177; MS3: 133, 105, sh 295, Esculin Eyles et al. 89, 149 334.7 2007; Parejo et al. 2004 13 B, E, M, Q 11.15 177 MS2: 133 MS3: 89 339.5, Esculetin C Eyles et al. 289.5, 2007 258 15 GS, G, W 11.91 395* MS2: 233, 185; MS3: 217, 264.7 Syringin Eyles et al. 203 2007; Kostova and Iossifova 2007 129 16 B, E, M 12.76 223* MS2: 208, 163, 190, 107; 289.5, Fraxidin A Yasuda et al.

MS3: 190, 180; MS4: 162, 341.9 2006 134 17 G, W, GS 12.7 ND by LC- ND by LC-MS PDA 265.9 Monolignol 1 - MS PDA 18 B, Q, E, M 13.37 369 MS2: 207; MS3: 192; 341.9, sh Fraxin Eyles et al MS4: 108, 164, 120, 175 300 2007; Godecke et al. 2005 19 B, Q, E, M 14.31 223* MS2: 208, 163, 190, 107; 293.1, Fraxidin B Yasuda et al. MS3: 190, 180; MS4: 162, 339.5 2006 134 21 B, E, M 15.66 385* MS2: 223; MS3: 209, 190; 328.7 Mandshurin Eyles et al. MS4: 190 2007; Terazawa 1986 22 B, M 17.28 209* MS2: 149, 163, 181, 194; 339.5 Fraxetin Eyles et al. MS3: 121 MS4: 93, 65 2007 23 B, E, M 18.57 353 MS2: 191, 179, 161 326.4, sh 3-Caffeoyl-quinic acid Cardoso et al. 298 (Chlorogenic acid) 2005

129

g h Master Species in which RT - [M-H]- or Fragments m/z (in order ! max Putative ID Reference Peak peak detectedb HPLC-UVc [M-H]+d of decreasing (nm)f Numbera abundance)e

24 B 19.72 573 MS2: 537, 375; MS3: 375; 277.7 Unknown Lignoid 1 - MS4: 179, 327 25 M 23.79 717 MS2: 519, 357; MS3: 357; 276.5 Pinoresinol dihexoside + 2H2O Eyles et al. MS4: 151, 136 2007; Terazawa 1986 26 B 26.05 639 MS2: 477; MS3: 315; 283, sh Lugrandoside A Baudouin et MS4: 135 327 al. 1987; Kostova and Iossifova 2007 29 W 26.97 525 MS2: 363, 301, 345; MS3: 278.9 Oleuropeain Related Compound Eyles et al. 301, 181, 199, 283; MS4: 1 2007;

130 283 Cardoso et al. 2007

30 Q 26.58 ND by LC- ND by LC-MS PDA 277.7 Unknown Lignoid 2 - MS PDA 31 G, GS 26.66 523 MS2: 361; MS3: 165, 179, 278.9 Ligustroside B De la Torre- 313; MS4: 147 Carbot et al. 2005 32 E 26.73 ND by LC- ND by LC-MS 277.7 Unknown Lignoid 3 - MS 33 Q 27.55 ND by LC- ND by LC-MS PDA 278.9 Unknown Secoiridoid 1 - MS PDA 34 B, M 28.29 477 MS2: 315, 203, 179, 341, 326.4, sh Calceolarioside C Eyles et al. 397; MS3: 135; MS4: 107 290 2007 35 Q 28.33 523 MS2: 361; MS3: 165, 179, 278.9 Ligustroside C Ryan et al. 313; MS4: 147 2002 36 G, GS, W 28.48 785 MS2: 623; MS3: 477, 461, 331.1, sh Forsythoside A O-glucoside - A Guo et al. 315; MS4: 315 290 2007 37 E 30.08 335 MS2: 179, 135; MS3: 135; 327.6, sh Caffeoylshikimic acid Lin and MS4: 107, 117, 91, 79 300 Harnley 2008;

130

g h Master Species in which RT - [M-H]- or Fragments m/z (in order ! max Putative ID Reference Peak peak detectedb HPLC-UVc [M-H]+d of decreasing (nm)f Numbera abundance)e Fang et al. 2002 38 GS, G, W, M, B 30.25 555 MS2: 393, 273, 307, 361, sh 280 10-Hydroxyoleuropein Hosny 1998; 375, 357, 419; MS3: 273, Kostova and 307, 357, 343, 361; MS4: Iossifova 137 2007 39 Q 30.18 357 MS2: 151, 136, 311; MS3: 276.5 Pinoresinol A Eyles et al. 136; MS4: 92, 108 2007 40 B 31.67 525 MS2: 481, 195, 389, 319, 281.2 Dimethyloleuropein Savarese et al. 345; MS3: 345; MS4: 165 2007 41 E, G, GS, M 32.99 357 MS2: 151, 136, 311; MS3: 278.9 Pinoresinol Eyles et al. 136; MS4: 92, 108 2007 42 Q, B 33.11 507 MS2: 475; MS3: 323, 203, 281.2, "-Methoxylferruginoside Guo et al. 233, 341; MS4: 161, 179, 327.6 2007

131 135 43 W 33.32 785 MS2: 623; MS3: 477, 459, 331.1, sh Forsythoside A O-glucoside - B Guo et al.

461; MS4: 315 290 (co-eluting peak with 41) 2007; Eyles et al. 2007 44 B, Q, E, M 34.15 477 MS2: 315, 179, 203, 323, 328.7, sh Calceolarioside A Eyles et al. 297, 341; MS3: 135; MS4: 290 2007 107 45 M, B 35.11 639 MS2: 477; MS3: 315; 329.9, sh Lugrandoside B Baudouin et MS4: 135 295 al. 1987; Kostova and Iossifova 2007 46 G, GS 36.02 539 MS2: 377, 307, 275, 291, sh 280 Oleuropein A Eyles et al. 359; MS3: 275, 291, 359, 2007; 179; MS4: 111, 155 47 E, G, GS, M, W 36.6 623 MS2: 461; MS3: 315, 297; 331.1, sh Verbascoside Eyles et al. MS4: 135 290 2007; 50 B 38.12 539 MS2: 377, 307, 275; MS3: sh 280 Oleuropein B Cardoso et al. 307, 275, 345; MS4: 275, 2005

131

g h Master Species in which RT - [M-H]- or Fragments m/z (in order ! max Putative ID Reference Peak peak detectedb HPLC-UVc [M-H]+d of decreasing (nm)f Numbera abundance)e 139 51 B 38.97 509 MS2: 347; MS3: 277, 233, sh 278 Demethylligstroside Takenaka et 179; MS4: 233 al. 2000; Kostova and Iossifova 2007 52 B, M 39.81 477 MS2: 315, 281, 251, 179, 328.7, sh Calceolarioside B Eyles et al. 221; MS3: 135; MS4: 107 290 2007 53 E, G, GS, M, B, W 40.21 539 MS2: 377, 307, 275; MS3: sh 281.2 Oleuropein Eyles et al. 307, 275, 345; MS4: 275, 2007; 139 Cardoso et al. 2005 57 GS 42.04 609a, 463b MS2a: 301; MS3a: 271, 257.6, Quercetin diglycoside (rutin)/ Eyles et al. 151, 179, 255; MS4a: 227, 356.2 Quercetin glucoside 2007; Ryan et

132 243, 215, 199: MS2b: 301; (isoquercetin) al. 2002 MS3b: 151, 179, 271, 255; MS4b: 107 59 E, G, GS, M, W 44.01 523 MS2: 361, 291, 259; MS3: sh 280 Ligstroside Eyles et al. 291, 259, 255; MS4: 139, 2007; 111, 171, 259, 143 60 GS, G 44.96 431 MS2: 269; MS3: 225, 117, 339.5, Apigenin glucoside Ryan et al. 240, 197 268.2 2002 63 E 46.57 447 MS2: 285, 255, 327, 227; 350, sh Kaempherol galactoside Ye et al. 2005 MS3: 255 MS4: 227, 211 295, 264 66 W 50.06 623 MS2: 461; MS3: 315 290.7, sh Verbascoside C Ryan et al. 332 1999 67 GS 52.35 285 MS2: 175, 199, 217, 241, 354, sh Luteolin Ryan et al. 243, 151; MS3: 147, 131, 290, sh 1999 133, 119, 146; MS4: 119 267, 249 68 GS, G, W 54.57 269 MS2: 225, 201, 149, 151; 268.2, sh Apigenin Eyles et al. MS3: 181, 183, 197; MS4: 290, 2007 155, 141, 154 339.5

132

aNumber corresponding to a single individual peak identified using HPLC-UV. Peak numbers were assigned according to their retention time. bLetters correspond to the species in which the compound was detected via HPLC-UV (B – black; E – European; G – Green; GS – Green seedling; Q – blue; M – Manchurian; W – white). cRT-HPLC-UV = the average retention time for all individual biological replicates for each compound from all species. dThe dominant molecular ion [M-H]- detected via LC-MS. Compounds were run in parallel with a PDA detector and full scan and PDA chromatograms were overlaid in order to match mass data. In some instances a [M-H]- could not be detected, in which case the positive ion mode was utilized to detect and identify the compound under investigation. For compounds identified using the positive ion mode, the corresponding [M-H]+ is denoted with an *. eCorresponding fragmentation of the dominant molecular ion. Mass spectra for MS2, MS3, and MS4 are shown where applicable. Fragments are ordered according to decreasing abundance, with bolded ions representing the main ions fragmented in subsequent fragmentations. f! max associated with each compound. gTentative compound identity assigned based on literature match. h 133 Corresponding reference for which a compound has been previously described. i

sh = shoulder

133

Table 4.2. Phenolic compounds identified by HPLC-UV and LC-MS-PDA from samples collected on August 8th, 2008 from susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash grown in a common garden in Bowling Green, OH.

g h Master Species in which RT - [M-H]- or Fragments m/z (in order of ! max Putative ID Reference Peak peak detectedb HPLC [M-H]+d* decreasing abundance)e (nm)f Numbera -UVc 1! B, Q, E, GS, M, W! 5.22! 315! MS2: 135, 179; MS3: 107, 91, 281.2! Hydroxytyrosol hexoside Cardoso et al. 117, 93, 79! 2005; Eyles et al. 2007! 2! B, GS, E, M! 6.2! 389! MS2: 329, 167, 161! 276.5! Vanillic acid hexoside acetate Eyles et al. adduct A 2007; Jimenez et al. 2010! 3! GS, E, W, G, M! 7.37! 299! MS2: 179, 119, 143, 113, 161, 275.3! Tyrosol hexoside! Kammerer et 131! al. 2005; Eyles et al. 2007! 5! GS! 8.59! 389! MS2: 329, 167, 161! 280.1! Vanillic acid hexoside acetate Eyles et al. 134 adduct B 2007; Jimenez

et al. 2010! 7! B, Q, E, M! 9.24! 177! MS2: 133; MS3: 89! 289.5, Esculetin A! Eyles et al. 340.7! 2007! 9! G, GS, W! 9.76! 601! MS2: 403, 223; MS3: 223, 371, 264.7, Elenolic acid derivative 1 Eyles et al. 179; MS4: 121! shi 300! (m/z 431; 275 UV max)! 2007! 10! M, Q! 10.02! 177! MS2: 133; MS3: 89! 258.8, Esculetin B! Eyles et al. sh 300, 2007! 340! 12! B, Q, E, M! 10.22! 339! MS2: 177; MS3: 133, 105, 89, 334.7, Esculin! Eyles et al. 149! sh 295! 2007; Parejo et al. 2004! 13! B, Q, E, M! 10.77! 177! MS2: 133; MS3: 89! 339.5, Esculetin C! Eyles et al. 289.5, 2007! 258! 14! B! 11.38! ND-LC- ND-LC-MS! 334.7, Unknown Coumarin 1 -! MS! sh 293, ! 257.6! 15! GS, W, Q, G! 11.8! 395*! MS2: 233, 185; MS3: 217, 203! 264.7! Syringin! Eyles et al.

134

g h Master Species in which RT - [M-H]- or Fragments m/z (in order of ! max Putative ID Reference Peak peak detectedb HPLC [M-H]+d* decreasing abundance)e (nm)f Numbera -UVc 2007; Kostova and Iossifova 2007! 16! B, E, M! 12.44! 223*! MS2: 208, 163, 190, 107; MS3: 289.5, Fraxidin A Yasuda et al. 190, 180; MS4: 162, 134! 341.9! 2006! 18! B, Q, E, M! 12.99! 369! MS2: 207; MS3: 192; MS4: 108, 341.9, Fraxin! Eyles et al. 164, 120, 175! sh 300! 2007; Godecke et al. 2005! 19! B, E, M! 13.94! 223*! MS2: 208, 163, 190, 107; MS3: 293.1, Fraxidin B Yasuda et al. 190, 180; MS4: 162, 134! 339.5! 2006! 20! Q! 13.94! 177! MS2: 133; MS3: 89 345.4, Esculetin! Eyles et al. 298, sh 2007! 260! 21! B, E, M! 15.25! 385*! MS2: 223; MS3: 208, 190; MS4: 328.7! Mandshurin! Eyles et al. 135 190! 2007;

Terazawa 1986! 22! B! 16.17! 209*! MS2: 149, 163, 181, 194; MS3: 339.5! Fraxetin! Eyles et al. 121; MS4: 93, 65! 2007! 23! B, E! 17.64! 353! MS2: 191, 179, 161! 326.4, 3-Caffeoyl-quinic acid Cardoso et al. sh 298! (Chlorogenic acid)! 2005! 24! B! 18.71! 573! MS2: 537, 375; MS3: 375; MS4: 277.7! Unknown Lignoid 1 "! 179, 327! ! 25! M! 23.64! 717! MS2: 519, 357; MS3: 357; MS4: 276.5! Pinoresinol dihexoside + 2H2O! Eyles et al. 151, 136! 2007; Terazawa 1986! 27! GS, G! 26.17! 523! MS2: 361; MS3: 165, 179, 313 278.9! Ligustroside A Hosny 1998! 28! Q! 26.31! 571! MS2: 373, 535, 343; MS3: 343; 281.2! (+)-1-Hydroxypinoresinol-4’-O- Guo et al. 2007! MS4: 285, 313, 207 glucoside + 2H2O! 29! W! 26.44! 525! MS2: 363, 301, 345; MS3: 301, 278.9! Oleuropein Related Compound 1 Eyles et al. 181, 199, 283; MS4: 283 2007; Cardoso et al. 2007 ! 33! Q! 27.24! ND by ND by LC-MS PDA 278.9! Unknown Secoiridoid 1 -! LC-MS

135

g h Master Species in which RT - [M-H]- or Fragments m/z (in order of ! max Putative ID Reference Peak peak detectedb HPLC [M-H]+d* decreasing abundance)e (nm)f Numbera -UVc PDA! 34! B, M! 27.69! 477! MS2: 315, 203, 179, 341, 397; 326.4, Calceolarioside C! Eyles et al. MS3: 135; MS4: 107 sh 290! 2007! 35! Q! 28.02! 523! MS2: 361; MS3: 165, 179, 313 278.9! Ligustroside C! Ryan et al. 2002! 36! G, GS, W! 28.08! 785! MS2: 623; MS3: 477, 461, 315; 331.1, Forsythoside A O-glucoside! Guo et al. 2007! MS4: 315 sh 290! 37! E! 28.9! 335! MS2: 179, 135 MS3: 135 MS4: 327.6, Caffeoylshikimic acid Lin and 107, 117, 91, 79 295! Harnley 2008; Fang et al. 2002! 38! B, Q, E, G, GS, M, W! 29.73! 555! MS2: 393, 273, 307, 361, 375, sh 280! 10-Hydroxyoleuropein Hosny 1998; 357, 419; MS3: 273, 307, 357, Kostova and 343, 361; MS4: 137 Iossifova 2007! 136 40! B! 30.43! 525! MS2: 481, 195, 389, 319, 345; 281.2! Dimethyoleuropein Savarese et al. MS3: 345; MS4: 165 2007!

41! B, Q, E, G, GS, M, W! 32.5! 357! MS2: 151, 136, 311; MS3: 136; 278.9! Pinoresinol! Eyles et al. MS4: 92, 108 2007! 44! B, Q, E, M! 33.5! 477! MS2: 315, 323, 179, 203, 341; 328.7, Calceolarioside A! Eyles et al. MS3: 135 MS4: 107 sh 290! 2007! 46! G, W! 35.55! 539! MS2: 377, 275, 291, 359; MS3: sh 280! Oleuropein A! Eyles et al. 275, 291, 359, 179; MS4: 111, 2007! 155 47! E, G, GS, M, W ! 36.04! 623! MS2: 461; MS3: 315, 297; MS4: 331.1, Verbascoside! Eyles et al. 135 sh 290! 2007! 48! W! 36.72! 623! MS2: 461; MS3: 315, 297; MS4: 281.2, Verbascoside A! Eyles et al. 136 331.1! 2007! 49! Q! 37.89! 725! MS2: 339; MS3: 177; MS4: 133 335.9, Escuside! Iossifova et al. sh 295! 2002;! 52! B, E, Q, M! 39.11! 477! MS2: 315, 281, 251, 179, 221; 328.7, Calceolarioside B! Eyles et al. MS3: 135; MS4: 107 sh 290! 2007! 53! B, E, G, GS, M, W ! 39.68! 539! MS2: 377, 307, 275; MS3: 307, 281.2! Oleuropein! Eyles et al. 275, 345; MS4: 275, 139 2007; Cardoso

136

g h Master Species in which RT - [M-H]- or Fragments m/z (in order of ! max Putative ID Reference Peak peak detectedb HPLC [M-H]+d* decreasing abundance)e (nm)f Numbera -UVc et al. 2007! 54! GS! 41.27! 793.9! MS2: 403, 589, 743, 748, 595; 265! Elenolic acid derivative 2! Eyles et al. MS3: 223, 371 2007! 55! GS, M, W! 41.75! 623! MS2: 461; MS3: 315, 297; MS4: 327.6, Verbascoside B! Eyles et al. 135 sh 288! 2007! 56! B! 42.02! 545! MS2: 369; MS3: 207, 192; MS4: 329.9, Fraxin Related Compound! Eyles et al. 192 sh 295! 2007; Godecke et al. 2005! 58! B, M! 43.09! 553! MS2: 391, 321, 289; MS3: 321, 280.1! Oleuropein Related Compound 2! Tanahashi et 289, 223; MS4: 139, 171, 167, al. 1996; 143, 261 Hosny 1998! 59! B, E, G, GS, M, W! 43.6! 523! MS2: 361; MS3: 291, 259; MS4: sh 280! Ligustroside! Eyles et al. 111, 139, 171, 143, 259 2007; Tanahashi et

137 al. 1996! 60! G, GS, W! 44.36! 431! MS2: 269; MS3: 225, 197, 117, 339.5, Apigenin glucoside! Ryan et al. 151 268.2! 2002! 61! B, M ! 44.77! 501! No fragmentation data available! 328.7, Unknown Coumarin 2 -! sh 295! ! 62! M! 45.48! ND by ND by LC-MS PDA! 327.6, Unknown Coumarin 3 -! LC-MS sh 298! ! PDA! 63! E! 46.01! 447! MS2: 285, 255, 327, 227; MS3: 350, sh Kaempherol galactoside! Ye et al. 2005! 255, 227; MS4: 227, 211! 295, 264! 64! GS! 46.03! ND LC- ND LC-MS PDA! sh 280! Unknown Secoiridoid 2 -! MS PDA! ! 65! W! 46.86! ND LC- ND LC-MS! 322.8, Unknown Phenylethanoid 1 -! MS! sh 282! ! 67! GS, W! 51.82! 285! MS2: 175, 199, 217, 241, 243, 254, sh Luteolin! Ryan et al. 151; MS3: 147, 131, 133, 119, 267, sh 1999! 146; MS4: 119 290, 349.0!

137

g h Master Species in which RT - [M-H]- or Fragments m/z (in order of ! max Putative ID Reference Peak peak detectedb HPLC [M-H]+d* decreasing abundance)e (nm)f Numbera -UVc 68! G, GS, W! 54.19! 269! MS2: 225, 149, 151, 201; MS3: 339.5, Apigenin! Eyles et al. 181, 183, 197; MS4: 155, 141, sh 290, 2007! 154 268.2! aNumber corresponding to a single individual peak identified using HPLC-UV. Peak numbers were assigned according to their retention time. bLetters correspond to the species in which the compound was detected via HPLC-UV (B – black; E – European; G – Green; GS – Green seedling; Q – blue; M – Manchurian; W – white). cRT-HPLC-UV = the average retention time for all individual biological replicates for each compound from all species. dThe dominant molecular ion [M-H]- detected via LC-MS. Compounds were run in parallel with a PDA detector and full scan and PDA chromatograms were overlaid in order to match mass data. In some instances a [M-H]- could not be detected, in which case the positive ion mode was utilized to detect and identify the compound under investigation. For compounds identified using the positive ion mode, the corresponding [M-H]+ is denoted with an *. e

138 Corresponding fragmentation of the dominant molecular ion. Mass spectra for MS2, MS3, and MS4 are shown where applicable.

Fragments are ordered according to decreasing abundance, with bolded ions representing the main ions fragmented in subsequent fragmentations. f! max associated with each compound. gTentative compound identity assigned based on literature match. hCorresponding reference for which a compound has been previously described. ish = shoulder

138

Table 4.3. Contents of individual phenolic compounds and lignin in susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash in samples collected on June 2nd, 2008. Compounds are separated into groups by phenolic compound class, while species are separated into the sections to which they belong within the genus Fraxinus (Wallander 2008). Contents are expressed in mg g-1 FW ± SEM (N = 8). Different letters within a row indicate significantly different means by the protected LSD test (! = 0.05). A solid black line separates black ash data from the other species because it was not part of the original experimental design, but can be visually compared to the other species.

Section Section Melioides Section Fraxinus Dipetalae Green Green Peak # Compound Name White Blue European Manchurian Black ‘Patmore’ Seedling

Phenolic Acids/Simple Phenolics 139

1 Hydroxytyrosol hexoside 0.2 ± 0.04 b 0.2 ± 0.1 b 0.4 ± 0.1 ab 0.6 ± 0.2 a 0.5 ± 0.1 ab 0.6 ± 0.1 a 0.6 ± 0.1

2 Vanillic acid hexoside acetate adduct A NDa ND ND ND ND 0.4 ± 0.1 0.4 ± 0.1 3 Tyrosol hexoside 0.7 ± 0.1 b 0.6 ± 0.1 b 0.6 ± 0.1 b ND 1.4 ± 0.3 a 0.4 ±0.1b ND 4 Unknown Phenolic Acid 1 0.1 ± 0.02 ND ND ND ND ND ND 5 Vanillic acid hexoside acetate adduct B ND ND 0.3 ± 0.1 ND ND ND ND 8 Caffeoyl-quinic acid A ND ND ND ND 3.5 ± 0.5 ND ND 11 Tyrosol hexoside pentoside 0.4 ± 0.1 a 0.5 ± 0.04 a 0.4 ± 0.1 a ND ND ND ND 23 3-Caffeoyl-quinic acid ND ND ND ND 0.5 ± 0.1 a 0.3 ± 0.04 a 0.4 ± 0.02

Coumarins

6 Fraxetin related compound ND ND ND ND ND ND 0.9 ± 0.1 7 Esculetin A ND ND ND 0.7 ± 0.2 ND ND 0.7 ± 0.02 10 Esculetin B ND ND ND 0.6 ± 0.2 ND ND ND

139

Section Section Melioides Section Fraxinus Dipetalae Green Green Peak # Compound Name White Blue European Manchurian Black ‘Patmore’ Seedling 12 Esculin ND ND ND 32.9 ± 1.7 a 0.7 ± 0.1 b 0.8 ± 0.1 b 16.3±0.4 13 Esculetin C ND ND ND 1.1 ± 0.1 a 0.1 ± 0.01 b 0.1 ± 0.01 b 0.7 ± 0.03 16 Fraxidin A ND ND ND ND 0.6 ± 0.1 a 0.6 ± 0.1 a 1.8 ± 0.2 18 Fraxin ND ND ND 8.5 ± 1.9 a 8.1 ± 0.8 a 11.0 ± 0.6 a 13.1 ± 0.7 19 Fraxidin B ND ND ND 0.3 ± 0.1 b 0.5 ± 0.1 ab 0.6 ± 0.04 a 0.03 ± 0.1 21 Mandshurin ND ND ND ND 0.8 ± 0.04 b 3.0 ± 0.3 a 5.7 ± 0.4 22 Fraxetin ND ND ND ND ND 2.3 ± 0.2 4.3 ± 0.13

140 Monolignols

15 Syringin 1.9 ± 0.6 b 4.4 ± 0.6 ab 5.6 ± 1.3 a ND ND ND ND 17 Monolignol 1 0.2 ± 0.1 a 0.7 ± 0.1 a 0.9 ± 0.2 a ND ND ND ND

Lignans

24 Unknown Lignoid 1 ND ND ND ND ND ND 1.9 ± 0.2 25 Pinoresinol dihexoside ND ND ND ND ND 1.0 ± 0.3 ND 30 Unknown Lignoid 2 ND ND ND 0.6 ± 0.1 ND ND ND 32 Unknown Lignoid 3 ND ND ND ND 1.4 ± 0.3 ND ND 39 Pinoresinol A ND ND ND 2.5 ± 0.5 ND ND ND 41 Pinoresinol ND 1.9 ± 0.2 b 3.0 ± 0.9 b ND 8.3 ± 1.5 a 8.4 ± 0.4 a ND

140

Section Section Melioides Section Fraxinus Dipetalae Green Green Peak # Compound Name White Blue European Manchurian Black ‘Patmore’ Seedling Phenylethanoids

26 Lugrandoside A ND ND ND ND ND ND 0.7 ± 0.1 34 Calceolarioside C ND ND ND ND ND 0.7±0.1 1.3 ± 0.3 36 Forsythoside A O-glucoside - A 2.2±0.4 a 2.2±0.2 a 2.4±0.5 a ND ND ND ND 42 "-Methoxylferruginoside ND ND ND 1.8 ± 0.3 ND ND 3.5 ± 0.5 43 Forsythoside A O-glucoside - B 0.8 ± 0.2 ND ND ND ND ND ND 44 Calceolarioside A ND ND ND 3.9 ± 0.8 b 3.1 ± 0.8 b 11.7 ± 2.3 a 33.0 ± 2.1

141 45 Lugrandoside B ND ND ND ND ND 1.8 ± 0.4 5.3 ± 0.4

47 Verbascoside 4.5 ± 0.4 a 3.6 ± 0.8 ab 1.9 ± 1.1 ab ND 2.6 ± 1.1 ab 1.5 ± 0.3 b ND 52 Calceolarioside B ND ND ND ND ND 5.0 ± 0.8 12.9 ± 0.7 66 Verbascoside C 0.1 ± 0.1 ND ND ND ND ND ND

Secoiridoids

29 Oleuropeain Related Compound 1 2.7 ± 0.2 ND ND ND ND ND ND 31 Ligustroside B ND 1.8 ± 0.2 a 1.6 ± 0.4 a ND ND ND ND 33 Unknown Secoiridoid 1 ND ND ND 1.2 ± 0.2 ND ND ND 35 Ligustroside C ND ND ND 0.7 ± 0.5 ND ND ND 38 10-Hydroxyoleuropein 2.2 ± 0.3 b 4.6 ± 0.3 a 4.4 ± 0.9 a ND ND 5.1 ± 0.2 a 3.9 ± 0.4 40 Dimethyloleuropein ND ND ND ND ND ND 4.8 ± 0.7 46 Oleuropein A ND 3.8 ± 0.4 a 2.8 ± 0.8 a ND ND ND ND

141

Section Section Melioides Section Fraxinus Dipetalae Green Green Peak # Compound Name White Blue European Manchurian Black ‘Patmore’ Seedling 50 Oleuropein B ND ND ND ND ND ND 1.4 ± 0.2 51 Demethylligstroside ND ND ND ND ND ND 2.3 ± 0.5 53 Oleuropein 2.2 ± 0.3 c 5.3 ± 1.8 bc 10.9 ± 3.1 a ND 2.8 ± 0.5 c 8.3 ± 0.9 ab 8.8 ± 1.4 59 Ligstroside 9.0 ± 1.1 a 8.9 ± 1.1 a 7.7 ± 1.4 ab ND 6.3 ± 0.6 ab 5.0 ± 0.5 b ND

Flavonoids

Quercetin diglycoside (rutin)/ 57 ND ND 0.7 ± 0.2 ND ND ND ND Quercetin glucoside (isoquercetin) 60 Apigenin glucoside ND 0.1 ± 0.01 a 0.1 ± 0.03 a ND ND ND ND

142 63 Kaempherol galactoside ND ND ND ND 0.3 ± 0.1 ND ND

67 Luteolin ND ND 0.1 ± 0.03 ND ND ND ND 68 Apigenin 0.01 ± 0.01 b 0.1 ± 0.01 b 0.1 ± 0.03 a ND ND ND ND

Phenolic Polymer

Lignin 10.0 ± 1.9 ab 7.2 ± 0.8 b 7.6 ± 1.0 b 11.7 ± 1.9 a 3.1 ± 0.2 c 9.0 ± 0.9 ab 15.1 ± 1.7

aND = Not detected

142

Table 4.4. Contents of individual phenolic compounds and lignin in susceptible white, green ‘Patmore’, green seedling, blue, European, black, and the resistant Manchurian ash in samples collected on June 2nd, 2008. Compounds are separated into groups by phenolic compound class, while species are separated into the sections to which they belong within the genus Fraxinus (Wallander 2008). Contents are expressed in mg g-1 FW ± SEM (N = 8). Different letters within a row indicate significantly different means by the protected LSD test (! = 0.05). A solid black line separates black ash data from the other species because it was not part of the original experimental design, but can be visually compared to the other species.

Section Section Melioides Section Fraxinus Dipetalae

Green Peak # Compound Name White Green ‘Patmore’ Blue European Manchurian Black Seedling

Phenolic Acids/Simple Phenolics

143 1 Hydroxytyrosol hexoside 0.4 ± 0.03 b NDa 0.4 ± 0.1 b 1.1 ± 0.1 a 0.5 ± 0.1 b 1.2 ± 0.1 a 1.0 ± 0.04

Vanillic acid hexoside 2 ND ND 0.1 ± 0.02 b ND 0.1 ± 0.01 b 1.0 ± 0.04 a 0.8 ± 0.1 acetate adduct A 3 Tyrosol hexoside 0.6 ± 0.1 b 0.5 ± 0.04 bc 0.4 ± 0.1 c ND 1.2 ± 0.2 a 0.5 ± 0.05 bc ND Vanillic acid hexoside 5 ND ND 0.4 ± 0.1 ND ND ND ND acetate adduct B 3-Caffeoyl-quinic acid 23 ND ND ND ND 1.2 ± 0.4 ND 0.2 ± 0.03 (Chlorogenic acid)

Coumarins 7 Esculetin A ND ND ND 0.4 ± 0.1 b 1.2 ± 0.3 a 0.1 ± 3e-3 c 0.9 ± 0.1 10 Esculetin B ND ND ND 0.4 ± 0.2 a ND 0.1 ± 3e-3 b ND 12 Esculin ND ND ND 63.0 ± 5.0 a 1.0 ± 0.2 c 1.3 ± 0.1 b 15.0 ± 0.3 13 Esculetin C ND ND ND 1.5 ± 0.3 a 0.2 ± 0.1 b 0.2 ± 0.01 b 1.0 ± 0.1 14 Unknown Coumarin 1 ND ND ND ND ND ND 0.1 ± 0.04

143

Section Section Melioides Section Fraxinus Dipetalae

Green Peak # Compound Name White Green ‘Patmore’ Blue European Manchurian Black Seedling 16 Fraxidin A ND ND ND ND 1.6 ± 0.3 a 1.3 ± 0.2 a 2.0 ± 0.2 18 Fraxin ND ND ND 8.7 ± 0.9 b 8.6 ± 1.7 b 22.5 ± 1.6 a 16.0 ± 0.6 19 Fraxidin B ND ND ND ND 0.7 ± 0.1 a 0.7 ± 0.1 a 0.9 ± 0.1 20 Esculetin ND ND ND 0.2 ± ND ND ND ND 21 Mandshurin ND ND ND ND 1.1 ± 0.1 b 5.0 ± 0.5 a 6.9 ± 0.8 22 Fraxetin ND ND ND ND ND ND 1.6 ± 0.5 56 Fraxin Related Compound ND ND ND ND ND ND 0.7 ± 0.1 61 Unknown Coumarin 2 ND ND ND ND ND 1.1 ± 0.1 1.2 ± 0.1

144

Unknown Coumarin 3

62 ND ND ND ND ND 0.7 ± 0.1 ND

Monolignol 15 Syringin 7.3 ± 1.2 b 7.6 ± 0.6 a 11.4 ± 1.4 a 0.6 ± 0.4 c ND ND ND

Lignans 24 Unknown Lignoid 1 ND ND ND ND ND ND 1.6 ± 0.2 Pinoresinol dihexoside + 25 ND ND ND ND ND 1.7 ± 0.3 ND 2H2O (+)-1-Hydroxypinoresinol- 28 ND ND ND 1.3 ± 0.4 ND ND ND 4’-O-glucoside + 2H2O 41 Pinoresinol 3.5 ± 0.7 b 2.7 ± 0.3 b 2.5 ± 0.4 b 3.4 ± 0.6 b 9.5 ± 1.7 a 10.1 ± 0.7 a 7.0 ± 0.4

144

Section Section Melioides Section Fraxinus Dipetalae

Green Peak # Compound Name White Green ‘Patmore’ Blue European Manchurian Black Seedling Phenylethanoids 34 Calceolarioside C ND ND ND ND ND 3.7 ± 0.4 3.0 ± 0.2 36 Forsythoside A O-glucoside 7.2 ± 2.6 a 3.9 ± 0.2 a 3.8 ± 0.6 a ND ND ND ND 10.1 ± 1.4 44 Calceolarioside A ND ND ND 6.5 ± 1.6 b 26.3 ± 1.8 a 34.6 ± 1.1 b 47 Verbascoside 4.3 ± 1.2 a 4.2 ± 0.3 a 1.7 ± 0.2 b ND 5.0 ± 1.4 a 1.7 ± 0.2 b ND 48 Verbascoside A 1.2 ± 0.3 ND ND ND ND ND ND 52 Calceolarioside B ND ND ND 3.0 ± 1.4 b 0.7 ± 0.2 c 25.2 ± 2.6 a 21.2 ± 1.3

145 55 Verbascoside B 3.9 ± 0.5 a ND 0.3 ± 0.1 c ND ND 1.1 ± 0.1 b ND Unknown Phenylethanoid 1 65 0.4 ± 0.1 ND ND ND ND ND ND

Secoiridoids

27 Ligustroside A ND 2.5 ± 0.2 a 1.6 ± 0.3 b ND ND ND ND 35 Ligustroside C ND ND ND 2.6 ± 0.3 ND ND ND Unknown Secoiridoid 1 33 ND ND ND 1.7 ± 0.2 ND ND ND

Oleuropein Related 29 3.0 ± 0.4 ND ND ND ND ND ND Compound 1 38 10-Hydroxyoleuropein 3.4 ± 0.3 d 9.4 ± 0.8 a 5.2 ± 0.5 c 7.5 ± 0.9 b 4.1 ± 0.6 cd 5.4 ± 0.3 c 4.5 ± 0.2 40 Dimethyoleuropein ND ND ND ND ND ND 3.0 ± 0.5 46 Oleuropein A 3.2 ± 0.2 b 5.4 ± 0.7 a ND ND ND ND ND 53 Oleuropein 1.8 ± 0.1 d 5.1 ± 0.5 c 19.2 ± 2.7 b ND 6.9 ± 0.8 c 26.2 ± 2.1 a 20.7 ± 2.0

145

Section Section Melioides Section Fraxinus Dipetalae

Green Peak # Compound Name White Green ‘Patmore’ Blue European Manchurian Black Seedling Oleuropein Related 58 ND ND ND ND ND 1.3 ± 0.1 4.0 ± 0.5 Compound 2 59 Ligustroside 10.4 ± 1.7 b 14.8 ± 1.2 a 9.8 ± 1.1 b ND 8.1 ± 1.2 b 3.9 ± 0.4 c 1.4 ± 0.5 64 Unknown Secoiridoid 2 ND ND 0.8 ± 0.3 ND ND ND ND

Coumarin-Secoiridoid 49 Escuside ND ND ND 2.1 ± 0.2 ND ND ND

Flavonoids 146

60 Apigenin glucoside 0.2 ± 0.01 ab 0.2 ± 0.02 a 0.1 ± 0.02 b ND ND ND ND 63 Kaempherol galactoside ND ND ND ND 0.3 ± 0.1 ND ND 67 Luteolin 0.1 ± 8e-3 a ND 0.04 ± 6e-3 b ND ND ND ND 68 Apigenin 0.1 ± 6e-3 b 0.1 ± 0.01 a 0.1 ± 0.01 a ND ND ND ND

Phenolic Polymer

Lignin 18.5 ±1.0 b 14.6 ±1.2 c 13.6 ± 1.0 c 21.6 ± 0.8 a 5.8 ± 0.8 d 18.1 ± 1.2 b 20.5 ± 1.0

aND = Not detected

146

CHAPTER 5

EFFECTS OF METHYL JASMONATE ON INDUCED RESPONSES OF

MANCHURIAN AND WHITE ASH TO EMERALD ASH BORER ELICITORS

AND A NECROTROPHIC PATHOGEN

ABSTRACT

The emerald ash borer (EAB) (Agrilus planipennis), an invasive wood-boring pest, has the potential to eliminate ash (Fraxinus spp.) from North American forests and urban environments. An Asian ash species (F. mandshurica) is resistant to attack by

EAB while North American species, including white ash (F. americana), are susceptible.

Resistance of deciduous trees to wood-boring beetles is attributed, in part, to induction of phenolics in phloem tissues upon insect feeding.

We hypothesized that resistant Manchurian ash accumulates phenolics at higher levels than susceptible white ash when induced with MeJA and/or EAB larval elicitors.

We used these treatments as proxies for a bioassay with the insect, which is not available at present. Treatment with a necrotrophic fungal pathogen, Botryosphaeria spp., was used to determine if there were any negative biological effects on a common ash pathogen when trees were treated with and without MeJA.

147

Changes in phenolic profiles and lignin content in ash phloem were monitored within and between species. In total, 31 phenolics (seven of which were shared by the two species) were quantified in response to the treatments. MeJA induced accumulation of individual phenolics and lignin in both species. Furthermore, shared compounds showed significant variation in concentration between species. Treatment with EAB elicitors and the necrotrophic pathogen had no effect on phenolic chemistry or lignin.

148

INTRODUCTION

Manchurian ash (F. mandshurica), an Asian species of ash, shares a co-evolutionary history with EAB, and as a result is able to resist attack from the pest (Rebek et al. 2008).

All North American ash species tested to date show varying degrees of susceptibility, with black (F. nigra), green (F. pennsylvanica), and white (F. americana) ashes being highly susceptible to EAB, and blue ash (F. quadrangulata) showing some level of resistance (Smith et al. 2005, Anulewicz et al. 2008, Rebek et al. 2008, Herms et al. unpublished). Resistance mechanisms of deciduous trees against phloem feeding buprestids are hypothesized to be the result of a combination of constitutive and induced chemical and physical defenses (Eyles et al. 2007). Phenolic compounds are known to play critical roles in plant-insect interactions and are a major component of the phytochemical constituents in the genus Fraxinus (Kostova and Iossifova 2007). A few studies have focused on constitutive differences in phenolic chemistry in phloem tissues of North American and Asian ash species (Eyles et al. 2007, Cipollini et al. 2011,

Chapter 4), yet no information is currently available on induced responses of phenolics in phloem tissues of ash.

Jasmonates, such as methyl jasmonate (MeJA), play a central role in regulating induced defense responses against herbivores and necrotropic pathogens that cause various types of tissue damage in plants (Howe and Jander 2008). Wounds created by chewing insects, such as EAB, and necrosis of plant tissues caused by phytotoxins produced by necrotrophic pathogens (such as Botryosphaeria spp.), activate similar defense response pathways in plants and are regulated by jasmonate signaling cascades

149

(Glazebrook 2005, Spoel et al. 2007, Howe and Jander 2008, Kliebenstein and Rowe

2008).

MeJA is an airborne plant-signaling molecule responsible for interplant communication (Green 1972, Farmer and Ryan 1990). MeJA is converted to its non- volatile, in planta active form, jasmonic acid (JA), by the enzyme methyl jasmonate esterase (MJE), which is responsible for regulating changes in the production of defensive traits such as toxic secondary metabolites, i.e. phenolics and phenolic polymers such as lignin (Gundlach et al. 1992, Koo and Howe 2009). MeJA has the ability to prime plants through modifications of the transcriptome/metabolome. While it may not directly confer resistance to treated plants, treatment with MeJA can prepare the plant to respond more quickly and with a more intense response following attack by an insect or pathogen (Ballare 2011).

In other instances, exogenous application of MeJA to the outer bark of certain tree species has been observed to have direct impacts on insect performance and induction of defense related phytochemicals (Erbilgin et al. 2006). In ash, exogenous application of

MeJA to plant tissues has been observed to modulate volatile emission profiles

(Rodriguez-Saona et al. 2006) and provide the same level of protection against EAB colonization as a protective insecticide spray (Chapter 2).

In this study we hypothesized that resistant Manchurian ash accumulates soluble phenolics and lignin at higher levels than susceptible white ash when primed with MeJA and induced with insect elicitors (EAB larval homogenate), and/or inoculation with a necrotrophic fungal pathogen commonly associated with ash-dieback of branches and twigs (Botryosphaeria spp.) (Sinclair and Lyon 2005). We used a necrotrophic pathogen

150

that commonly affects ash in order to evaluate if MeJA induced defenses negatively impacted its growth/pathogenicity.

MATERIALS AND METHODS

Experimental Design

Eighteen ash trees (10 white and eight Manchurian ash trees) were acquired from

Bailey Nursery, Inc., St. Paul MN, USA in the spring of 2007 and immediately planted in seven-gallon pots in a medium containing pine bark and Comtil (6:1, v:v, pine bark and

Comtil). The trees were maintained at the Howlett Hall nursery (40.002° N/83.029° W),

The Ohio State University, until the beginning of the experiment on June 29th, 2008, when Manchurian and white ash trees had mean stem diameters 15 cm above the soil line of 2.9 ± 0.2 cm (s.e.m.) and 3.0 ± 0.2 cm, respectively. Because MeJA is an airborne plant-signaling molecule with the potential to induce biochemical changes in neighboring plants, the trees were physically separated into a MeJA treatment and a H2O control group (each with five white and four Manchurian ash trees) by placing the two groups on either side of a polyhouse. This placement separated the two groups by approx. 50 ft.

Light intensity, temperature, and relative humidity were taken three times a day (10 am,

12 pm, and 2 pm) within each experimental group throughout the course of the experiment (11 days total) to determine if there were differences in the microenvironments for each treatment group (Figure 5.3). The position of each tree in a treatment group was randomly assigned.

151

MeJA or water were applied to three randomly selected branches on any given tree, depending on treatment group. The first group received an exogenous application of a 1

M solution of MeJA in water with Tween 20, while the second treatment group received only water with Tween 20. Previous field studies had demonstrated that 1 M MeJA exogenously applied to ash outer bark is able to induce the accumulation of phenolics in ash phloem tissues (data not shown). Lower concentrations of MeJA (0.1 M) had no effect on phenolic chemistry, and higher concentrations (2 M) appeared to be phytotoxic, resulting in necrosis of treated tissues (personal observation, Don Cipollini personal communication).

A 20 cm segment of each branch was marked beginning at the branch-stem junction and

1 ml of the MeJA and H2O solutions were applied along the 20 cm length of branch using a cotton swab to evenly coat the whole branch segment. MeJA and water treatments were implemented on July 2nd, 2008 and applied every other day for a week prior to the application of subtreatments, corresponding to days one, three, five, and seven of the experiment. On day eight, a mechanical wound was created on the upper side of each treated branch, using a scalpel, to remove a 1 cm2 window of phloem tissue, 5 cm away from the stem-branch junction (excluding the unwounded control branch) (Figure 5.1).

Subtreatments were immediately applied and consisted of: 1) an unwounded control, 2) an EAB larval homogenate elicitor treatment, and 3) a necrotrophic pathogen isolated from infected ash material in the field (Botryosphaeria spp.). Subtreatments were randomly assigned to each branch within a tree. Unwounded control branches were parafilmed five cm away from the stem branch interface to be consistent with treated branches.

152

Figure 5.1. MeJA or water were applied to a 20 cm branch segment (between orange tape markers), while the sub-treatment was applied within a 1 cm2 bark window positioned on the upper side of the branch, 5 cm from the stem-branch junction. .

153

Figure 5.2. Diagrammatic representation of artificial wound created on ash branch where EAB larval homogenate was applied and phloem tissue collected around wound for chemical analyses.

154

Figure 5.3. Environmental data were collected daily throughout the course of the experiment in both the MeJA and H2O control experimental blocks. Data points represent the mean (± SEM) of three separate measurements taken at 10 am, 12 pm, and 2 pm. While on some days there might have been significant differences between the two blocks (e.g. on day 8 for temperature and days 3, 8, and 10 for humidity), the two blocks did not differ overall (ANOVA).

155

EAB larval homogenate (see below for preparation protocol) was applied by packing the wound with larval material, overlaying it with a plug of 2% water agar to rehydrate the larval material, and then covering with parafilm to seal in the subtreatment to minimize desiccation while rehydrating the larval components. The fungal subtreatment consisted of plugs of mycelium taken from the margins of cultures actively growing on

2% water agar (see below for details). The mycelial plugs were placed, mycelium-side down, on the wounds and the infection courts were parafilmed to minimize desiccation and contamination. Lesion lengths on fungal inoculated trees were measured on day 11 of the experiment. On day 11 (July 12th), treated branches were cut from trees at the stem junction, placed in Ziploc bags, put on ice, and transported back to the lab. In the lab, a one cm by one cm band of phloem tissue around the wound site and extending around the circumference of the branch was removed and used for phenolic and lignin analyses

(Figure 5.2). Phloem tissue was immediately frozen in liquid N2 and stored at -80° C until phenolic and lignin extraction.

Larval Homogenate Preparation

Currently there are no reliable protocols to artificially challenge ash trees with EAB.

Thus, a larval homogenate preparation was used in this experiment to treat trees, assuming such preparation contains elicitors of host defense responses.

Fourth instar EAB larvae were collected from infested green and white ash trees in an urban setting in Toledo, OH (41.67° N/83.55° W) in the summer of 2007. Instar was determined by measuring peristoma width, which correlates with instar in Agrilus spp.

156

larvae (Loerch and Cameron 1983). A total of 30 EAB larvae were prepared with an average peristomal width of 1.26 ± 0.03 mm and weight of 124.5 ± 9.52 mg. Larvae were frozen and ground to a fine powder in liquid N2. Each individual ground larva was placed in a 1.5 ml microcentrifuge tube that was covered with parafilm with holes and then lyophilized. A single larva was used for each subtreatment application.

Isolation of Botryosphaeria

A Botryosphaeria spp. was isolated from dead green ash branch material displaying characteristic necrotic cankers (Sinclair and Lyon 2005), covered with pycnidia, on the Ohio State University campus in the fall of 2007. A pure culture of the pathogen was obtained using the methods employed to isolate a fungal pathogen from pinecones as described in (Whitehill et al. 2007). Identification to the genus level of the fungal pathogen was obtained by microscopic examination of conidia isolated from pycnidia developing on pure cultures grown on 2% water agar for two weeks in the dark.

Analysis of Phenolics and Lignin

Phenolics and lignin were extracted and analyzed, and individual phenolics identified, as described in Chapter 4.

157

Statistical Analyses

Environmental data (Figure 5.3) were analyzed using univariate ANOVA. Peak areas of individual compounds and lignin did not meet assumptions of normality and homogeneity of variance, even after several transformations (squre-root, 1/square-root, log10, arcsin). Thus, non-parametric methods were used. Peak areas of individual compounds and lignin concentrations were rank-transformed and analyzed using the

Kruskal-Wallis test to identify significant differences among all treatment, species, and subtreatment combinations. The Kruskal-Wallis test was also used post-hoc to separate treatment, species, and subtreatment combinations (! = 0.05). Subtreatment had no effect on any of the response variables. Therefore, each response variable in each subtreatment was averaged within a tree. This average was then used for all post-hoc analyses to test for effects of treatment, species, and their interactions (Tables 5.2 - 5.4 and Figures 5.5 -

5.8).

RESULTS

Identification of Phenolics

Thirty-one compounds were identified from both species of ash (Table 5.1 and

Figure 5.4). Seven compounds were detected in both ash species (Table 5.2 and Figure

5.5), 16 only in Manchurian ash phloem (Table 5.3 and Figure 5.6) and eight only in white ash phloem (Table 5.4 and Figure 5.7).

158

Figure 5.4. Overlay of HPLC chromatograms showing effects of MeJA in the absence of a subtreatment: black trace - Manchurian ash treated with MeJA; magenta trace - Manchurian ash treated with H2O; blue trace - white ash treated with MeJA; orange trace - white ash treated with H2O. Peaks correspond to identities presented in Table 5.1

159

Figure 5.5. Concentrations of individual compounds shared between Manchurian and white ash presented as absorbance units µl-1 extract. Bars are means of individual peak areas (± SEM) for each compound (n = 4 for Manchurian ash and n = 5 for white ash where each n is composed of information from three separate subsamples). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes.

160

Figure 5.6. Concentrations of individual compounds specific to Manchurian ash presented as absorbance units µl-1 extract. Bars are means of individual peak areas (± SEM) for each compound (n = 4 for Manchurian ash where each n is composed of information from three separate subsamples). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes.

161

Figure 5.7. Concentrations of individual compounds specific to white ash. Bars are means of individual peak areas (± SEM) for each compound (n = 5 for white ash where each n is composed of information from three separate subsamples). EAD 1 = elenolic acid derivative 1 and ORC 1 = oleuropein related compound 1. Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes.

162

Figure 5.8. Lignin concentration (± SEM) in the phloem of Manchurian and white ash (n = 4 for Manchurian ash, and n = 5 for white ash). Letters indicate significantly different medians between treatment groups and were separated using the Kruskal-Wallis test. Means are used in place of medians for presentation purposes.

163

Effect of MeJA Treatment, Species, and Subtreatment on Individual Phenolics and Lignin

Of the 31 compounds investigated, 17 were significantly affected by treatment with

MeJA (Tables 5.2 - 5.4 and Figure 5.5 - 5.7). Two of the seven compounds shared between species, plus lignin, were significantly induced by the MeJA treatment compared to controls (Table 5.2 and Figures 5.5 and 5.8). In Manchurian ash, 10 compounds were significantly affected by the MeJA treatment (Table 5.3 and Figure 5.6). In white ash, five compounds were significantly affected by treatment with MeJA, with three compounds decreasing and the remaining two increasing in concentration (Table 5.4 and

Figure 5.7). Significant differences between species were observed for six of the seven shared compounds, while lignin concentrations were not significantly different (Table 5.2 and Figure 5.5 and 5.8). Subtreatment had no significant effect on any individual phenolic compound and lignin (Tables 5.2 - 5.4 and Figures 5.4 – 5.6).

DISCUSSION

In this study, we have demonstrated that exogenously applied, 1 M MeJA induces significant changes in stem phloem phenolic chemistry and lignin concentrations of ash species, both resistant and susceptible to EAB (Figure 5.5 – 5.8 and Table 5.2 – 5.4).

Application of EAB elicitors in the form of larval homogenate or inoculation with a necrotrophic pathogen had no significant effects on induced responses.

MeJA induction of defense responses in plants has been characterized previously in many plant species, and includes the induction of defense gene products and defensive

164

secondary metabolites such as phenolics (Farmer and Ryan 1990, Baldwin et al. 1994,

Baldwin 1998, Orozco-Cardenas and Ryan 1999, Black et al. 2003, Moore et al. 2003,

Engelberth et al. 2004, Chen et al. 2005, Erbilgin et al. 2006, Tamogami et al. 2008,

Akiyama et al. 2009, Shivaji et al. 2010). Previous studies on ash have demonstrated that foliar application of MeJA to Manchurian ash seedlings (15-30 cm tall) affected their volatile emission profiles, suggesting the jasmonate pathway is active in Fraxinus spp.

(Rodriguez-Saona et al. 2006). Also, application of MeJA to stem tissues of various ash species in the field provided protection against EAB equivalent to that of an insecticide formulated against wood-boring beetles (Chapter 2). This study represents the first evidence that MeJA has the ability to significantly induce the accumulation of phenolics in phloem tissues of Fraxinus spp. following direct application to the outer bark. The induction of phenolic chemistry by exogenous MeJA could explain, at least in part, the observed protection of susceptible ash in the field (Chapter 2).

In this study we did not observe any significant effect of larval homogenate on ash phloem. We hypothesize that this is a result of unintentional bias in our sampling method. Phloem tissues from larval homogenate-treated branches were collected by removing a one cm by one cm band of phloem tissue around the wound site that extended around the circumference of the branch, thus including a large proportion of tissue that was distal to the treatment site, which would result in a significant dilution of host responses. A similar dilution effect may have masked the local responses to inoculation with the necrotrophic pathogen, Botryosphaeria spp.

Concentrations of six of the seven shared phenolic compounds differed significantly between species. Such differences have been noted previously and their potential roles in

165

defense against EAB have been discussed (Eyles et al. 2007, Cipollini et al. 2011). In white ash, MeJA induced significant reductions for tyrosol hexoside (Figure 5.5). A reduction in compound concentration following application of MeJA could be attributed to shunting of the precursors for these compounds toward biosynthesis of more complex phenolic compounds (Strack 1997). On the other hand, MeJA induced a 275% increase in the concentration of verbascoside in the same tissues (Figures 5.4 and 5.5, and Table

5.2). Previous studies have found that verbascoside is present in constitutively higher concentrations in white ash phloem than in green ash cv. Patmore, green ash seedlings,

European, and Manchurian ash phloem (Eyles et al. 2007, Chapter 4). Verbascoside has no known defensive roles against insects, although it may play a role in resistance of

Plantago major against slugs (Molgaard 1986). The strong response of verbascoside to

MeJA treatment suggests it may play a critical role in the general defense responses of white ash.

In Manchurian ash, only one shared compound, oleuropein, increased significantly in the MeJA treated trees, suggesting a possible role in defense. This is also supported by the higher constitutive levels of oleuropein in Manchurian ash than in susceptible

European, black, white, and green ash.

Sixteen compounds were unique to Manchurian ash and 10 of these were significantly affected by treatment with MeJA. All 10 compounds increased significantly in concentration (Figure 5.4 and 5.6, and Table 5.3). Of the compounds that increased in concentration in Manchurian ash, one was a phenolic acid (vanillic acid hexoside), six were hydroxycoumarins (esculetin C, fraxidin A, fraxin, fraxidin B, mandshurin, and fraxetin), and three were phenylethanoid glycosides (calceolarioside A, B, and C). These

166

compounds were found to be present constitutively in the phloem tissues of Manchurian,

European, black, blue, green, and white ashes (Chapter 4). Manchurian ash phloem tissues contained higher levels of these compounds when compared to other ash species except for esculetin C (which was more concentrated in the constitutive phloem tissues of blue ash), mandshurin and calceolarioside A (which were more abundant in black ash phloem), and fraxetin, which was not detected in constitutive Manchurian ash phloem tissues (Chapter 4). Thus, Manchurian ash appears to be very plastic in its phenolic chemistry response to MeJA, which may be indicative of its ability to mount a more fine- tuned defense against EAB. While the modes of action of phenolic compounds in defense against herbivores are diverse and remain largely uncharacterized (Appel 1993,

Duffey and Stout 1996), induction of phenolics in Manchurian ash phloem tissues treated with MeJA suggests that phenolics may be responsible, at least in part, for its observed resistance to EAB in the field (Rebek et al. 2008, Chapter 2).

Eight compounds were unique to white ash phloem extracts and MeJA significantly affected five of these compounds. Two compounds, elenolic acid derivative and oleuropein related compound 1, decreased in concentration; oleuropein A fell below the limit of detection; while two compounds, unknown 2 and unknown 3, increased in concentration following treatment with MeJA (Figure 5.4 and 5.7, and Table 5.4).

Further work is needed to identify unknown compounds 2 and 3 before any hypotheses can be drawn regarding their potential defensive roles in white ash, but they may also be active in protection of white ash against EAB attack afforded by treatment with MeJA

(Chapter 2).

167

In conclusion, exogenous application of MeJA to Manchurian and white ash phloem tissues elicited strong responses in phenolic chemistry that are consistent with results of defense responses observed in other tree species treated with MeJA (Miller et al. 2005,

Erbilgin et al. 2006). Almost half (11 of 23 compounds) of the phenolic compounds in

Manchurian ash phloem increased significantly following treatment with MeJA, compared to only 20% (three of 15 compounds) of the compounds in white ash. The diversity of phenolic compounds induced in Manchurian ash phloem could be indicative of its ability to mount an effective defense against colonization by EAB in the field. This study highlights the efficacy of MeJA as an effective tool to study induced defenses of

Fraxinus spp, although further work is needed to identify responses that are induced in phloem tissues by actively feeding EAB larvae.

ACKNOWLEDGEMENTS

Thanks to Dr. Brian McSpadden-Gardener for help with statistical analyses and guidance on aspects dealing with the experimental design. We also thank Scott Williams for help in the implementation and execution of the experiment as well as processing and analyses of samples. Funding provided through the OARDC and a joint venture between the US

Forest Service and The Ohio State University.

168

Table 5.1. Characteristics of HPLC peaks analyzed with a photo diode array detector (PDA) and putative identifications based on retention time and UV spectrum based on identities presented in Chapter 4.

Peak Number Species Average RT " max (nm) Putative ID 1 Both 5.24 278 Hydroxytyrosol hexoside 2 Manchurian 6.28 276.5 Vanillic acid hexoside 3 Both 7.34 275.3 Tyrosol hexoside 4 White 9.76 267.7 Elenolic Acid Derivative 1 5 Manchurian 10.26 334.7, sh 290 Esculetin B 6 Manchurian 10.89 290.7, 338.3 Esculetin C 7 White 11.85 264.7 Syringin 8 Manchurian 12.58 290.7, 337.1 Fraxidin A 9 Manchurian 13.10 sh 300, 341.9 Fraxin 10 Manchurian 14.14 291.9, 337.1 Fraxidin B 11 Manchurian 15.46 328.7 Mandshurin 12 Manchurian 16.64 337.1 Fraxetin 13 Manchurian 17.92 326.4, sh 295 Chlorogenic Acid 14 Manchurian 23.57 276.5, 327.6 Pinoresinol Dihexoside 15 White 26.34 278.9 Oleuropein Related Compound 1 16 Manchurian 26.45 326.4, sh 295 Unknown #1 17 Manchurian 28.15 326.4, 288.3 Calceolarioside C 18 White 28.25 332.3, sh 290 Forsythoside A O-glucoside - A 19 Manchurian 29.19 326.4, sh 300 Caffeoylshikimic Acid 20 Both 29.87 277 10-Hydroxyoleuropein 21 Both 32.70 278.9 Pinoresinol 22 Manchurian 33.65 328.7, sh 290 Calceolarioside A 23 White 35.65 sh 278 Oleuropein A 24 Both 36.11 331.1, sh 290 Verbascoside 25 Manchurian 39.23 327.6, sh 285 Calceolarioside B 26 Both 39.79 sh 281 Oleuropein 27 White 40.12 331, sh 300 Unknown #2 28 White 40.58 315.7 Unknown #3 29 White 41.29 331.1, sh 286 Verbascoside B 30 Both 43.63 238.7 Ligustroside 31 Manchurian 44.39 282.4, 334.7 Unknown #4

169

2 Table 5.2. # values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of species, treatment, and subtreatment effects on HPLC peaks shared between F. americana and F. mandshurica.

Species (df=1) Treatment (df=1) Subtreatment (df=2) Compound 2 2 2 # # # Hydroxtyrosol 33.805*** 0.152 0.818 hexoside Tyrosol hexoside 2.786 22.567** 1.038

10-Hydroxyoleuropein 10.23*** 0.167 0.931

Pinoresinol 3.158*** 0.027 0.957

Verbascoside 39.005*** 7.237** 0.07

Oleuropein 33.603*** 0.006 0.881

Ligustroside 17.979*** 1.228 0.537

Lignin 0.007 3.840* 4.680 *P < 0.05 **P < 0.01 ***P < 0.001

170

2 Table 5.3. # values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of treatment and subtreatment effects on individual HPLC peaks in F. mandshurica.

Treatment (df=1) Subtreatment (df=2) Response Variable Chi-Square Chi-Square Vanillic Acid hexoside 9.363** 1.415 Esculetin B 0.403 1.280 Esculetin C 3.853* 0.140 Fraxidin A 12.403*** 0.020

Fraxin 10.830*** 0.155 Fraxidin B 14.520*** 0.155

Mandshurin 4.083* 1.635

Fraxetin 10.453*** 0.365

3-Caffeoyl-quinic Acid 2.803 2.895 Pinoresinol Dihexoside 1.203 3.065

Unknown #1 0.083 0.455

Calceolarioside C 5.880* 2.205

Caffeoylshikimic Acid 0.030 0.335

Calceolarioside A 8.670** 0.315

Calceolarioside B 6.750** 2.135

Unknown #4 0.053 4.515 *P < 0.05 **P < 0.01 ***P < 0.001

171

2 Table 5.4. # values, degrees of freedom (df), and associated significance from the Kruskal-Wallis test of treatment and subtreatment effects on individual HPLC peaks in F. americana.

Subtreatment Treatment (df=1) (df=2) Response Variable Chi-Square Chi-Square Elenolic Acid Derivative 1 11.710*** 0.157 Syringin 1.497 0.209 Oleuropein Related Compound 1 10.874*** 1.102 Forsythoside A O-glucoside - A 1.033 0.836 Oleuropein A 7.759** 1.140 Unknown #2 9.000** 1.680 Unknown #3 10.874*** 0.519 Verbascoside B 0.413 0.163 *P < 0.05 **P < 0.01 ***P < 0.001

172

CHAPTER 6

CHARACTERIZATION OF 4TH INSTAR EAB LARVAL ORAL SECRETIONS:

A MOLECULAR WINDOW INTO EAB – ASH INTERACTIONS

ABSTRACT

Previous chapters describe putative constitutive resistance traits to EAB in

Manchurian ash and induced responses to wounding and elicitation with a known defense related phytohormone (MeJA). However, treatment with a larval homogenate preparation did not elicit any host defense responses. Studies in other systems have demonstrated that insect feeding or application of insect oral secretions to plant wounds elicits a stronger response from plants than just mechanical damage by itself.

Characterization of insect oral secretions has provided insight into the molecular interactions between plants and insects. Therefore, in this study we characterized the oral secretions of 4th instar EAB larvae actively feeding on susceptible green ash in the field using shotgun proteomics and metabolomics. We identified 27 distinct proteins of plant and bacterial origins, and separated 15 low molecular weight compounds. Proteins from plants showed a diverse range of functions, from involvement in regulation of reactive

173

oxygen species to proteins that are known to be antinutritive or directly toxic to insects.

Bacterial proteins were most notably associated with detoxification of H2O2.

Metabolomic analyses revealed several plant phenolics found in ash phloem, while other compounds have yet to be identified. We hypothesize that EAB larval oral secretions may play a critical role in the success of EAB in susceptible ash in the field.

174

INTRODUCTION

EAB is devastating North American forests, and its progress does not appear to be slowing (Cappaert 2005, Kovacs et al. 2010). The Asian ash species, Manchurian ash, is resistant to attack by EAB, while all North American ash species are susceptible (Rebek et al. 2008). The interactions between ash and EAB at the molecular level are poorly understood at present and, so far, studies have focused on the characterization of constitutive defenses of ash trees and the characterization of EAB transcripts, but with no direct link between the two (Eyles et al. 2007, Mittapalli et al. 2010, Bai et al. 2011).

Therefore, information about the direct interactions between ash and EAB is needed so that mechanisms of resistance can be identified.

An increasing number of studies has demonstrated that insect feeding or application of insect oral secretions to plant wounds elicits a stronger response from plants than just mechanical damage by itself (Arimura et al. 2004a, Arimura et al. 2004b, Philippe et al.

2010). For instance, elicitors have been identified in the oral secretions of lepidopteran species and the oviposition fluid of weevils (Kessler and Baldwin 2002). Two main classes of insect elicitors have been identified and include fatty acid-amino acid conjugates (FACs) and lytic enzymes such as glucose oxidase and $-glucosidase (Felton and Eichenseer 1999, Howe and Jander 2008). Microbes present in the digestive organs of herbivores may also be involved in the production of elicitors present in insect oral secretions (Spiteller et al. 2000).

In this study we characterized the oral secretions of 4th instar EAB larvae feeding on green ash in the field using techniques that monitor global characteristics of the proteome

175

and metabolome (Fiehn 2002, Hu et al. 2007). Shotgun proteomics and metabolomics utilize the separation power of liquid chromatography (LC) coupled with the sensitivity of mass spectrometry (MS) to identify individual peptides resulting from digestion of total protein extracts with a proteolytic enzyme, as well as low molecular weight metabolites (Oliver et al. 1998, Washburn et al. 2001, Fiehn 2002). The ultimate goal of

EAB oral secretion characterization is to identify molecules that are responsible for the elicitation/suppression of defense responses in ash.

MATERIALS AND METHODS

Collection of EAB Oral Secretion

An EAB-infested green ash tree (approximately 50-70% canopy dieback d.b.h 31.3 cm) was selected on August 20th, 2009 in Bowling Green, OH in an open field next to the water treatment plant for the city (41.38° N/83.61° W). The outer bark and phloem tissue were removed from the tree using a chisel and hammer starting at breast height to slowly expose the inner sapwood. Forth instar EAB larvae were removed from the tree using bioquip tweezers. Upon removal from the tree, we observed that EAB larvae produce a droplet of oral secretion (Figure 6.1). The amount of oral secretion produced by an individual larva varied, but 4th instar EAB larvae typically yielded between < 1 – 5 µl of oral secretion. We chose to use only 4th instar EAB larvae oral secretions for our proteomic and metabolomic analyses in order to maintain consistency in our analyses. A

20 µl pipette was used to extract droplets of oral secretions from EAB larvae immediately

176

Figure 6.1. A 4th instar EAB larva that has just been removed from feeding on a green ash tree in the field. The liquid at the insect mouth is characteristic of actively feeding EAB larvae that are removed from trees in the field. The oral secretion droplet was collected and used in our proteomic and metabolomic analyses.

177

upon removal from the tree. Oral secretions were then transferred into a 1.5-ml microcentrifuge tube. In order to minimize any potential degradation of proteins in EAB oral secretions, extracts from 10 EAB larvae were pooled into a single tube that was immediately placed on ice until a final volume of ~50 µl was collected in total from all larvae. EAB oral secretions were placed on ice until they could be transported back to the lab (approximately 3.5 hrs). A 1-D protein gel was run following extraction of oral secretion proteins to test for any potential degradation of which no evidence was observed (data not shown). Once in the lab, all EAB oral secretion samples were pooled, flash frozen in liquid N2 and then stored at -80° C until processing.

Extraction of Proteins and Metabolites

Samples were removed from storage, placed on ice, and allowed to thaw. Proteins were precipitated from samples by adding 500 µl of an ice-cold precipitation solution

(10% TCA and 20 mM DTT in acetone) to 20 µl of EAB oral secretions, followed by vortexing. The precipitation solution was slightly modified from (Vâlcu and Schlink

2006). Samples were then stored overnight at -20° C. Samples were subsequently centrifuged (26,000 g) for 30 min at 4° C. Following centrifugation, the supernatant

(metabolite fraction) was moved to a new tube and stored at -80° C until LC-MS-PDA analysis. Protein pellets were washed by adding 1 ml of washing solution (20 mM DTT in chilled acetone) to the protein extract, vortexing samples briefly (~3 s), placing the sample at -20° C for 1 hr, centrifuging samples for 30 min at 4° C, and discarding the supernatant. Protein pellets were washed a total of three times to remove any residual

178

TCA from protein pellets. Immediately following the last washing step, pellets were allowed to dry for 10 min in a fume hood at RT. Pellets were then re-solubilized into a multidimensional protein identification technology (MuDPIT) compatible buffer.

Protein Extraction and Preparation for Shotgun Proteomics

Protein pellets were first re-constituted in 50 µl of 6.0 M urea with 2.5 µl of 200 mM

DTT in 25 mM ammonium bicarbonate. Pellets were placed on a shaker at room temperature for up to 4 hr to allow for complete re-solubilization of the pellet. Following resolubilization, samples were vortexed and proteins were alkylated with 10 µl of 200 mM iodoacetamide in 25 mM ammonium bicarbonate. Samples were then immediately placed in the dark at RT and allowed to react for 1 hr. Following alkylation, leftover iodoacetamide was consumed by adding 10 µl of the reducing agents (200 mM DTT in

25 mM ammonium bicarbonate) to prevent alkylation of trypsin. Protein samples were then diluted with 450 µl of 25 mM ammonium bicarbonate to reduce the concentration of urea which, unless diluted, can interfere with mass spectrometry analysis. Samples were stored at -80° C until ready for trypsin digestion. Immediately preceding shot-gun proteomic analyses, samples were digested by adding trypsin in a 1:30 ratio to the protein extract which was then incubated overnight at 37° C.

179

Shotgun Proteomic Analysis

Tryptic peptides were sequenced via capillary nano-LC-MS/MS on a Thermo

Finnigan LTQ mass spectrometer equipped with a nanospray source operated in positive ion mode. Capillary nano-LC-MS/MS was performed using similar methods to those described in (Reddish et al. 2008). Solvent A (50 mM acetic acid in water) and solvent B

(acetonitrile) were used for all chromatographic separations. Samples (5 %l from each sample) were prepared in solvent A, injected onto a µ-Precolumn Cartridge (Dionex,

Sunnyvale, CA), and washed with 50 mM acetic acid. The injector port was switched to inject and the peptides were eluted from the trap (Precolumn Cartridge ) onto the column. A 5 cm, 75 µm ID ProteoPep II C18 column (New Objective, Inc. Woburn,

MA) packed directly in the nanospray tip was used for all chromatographic separations. Peptides were eluted directly into the LTQ system using a gradient of 2-80% solvent B over 45 minutes, with a flow rate of 300 nl/min. The total run time was 65 minutes. The nanospray source was operated with a spray voltage of 3 KV and a capillary temperature of 200˚ C was used. The scan sequence of the mass spectrometer was based on the TopTen™ method; the analysis was programmed for full scan (recorded between 350 – 2,000 Da), and a MS/MS scan to generate product ion spectra to determine amino acid sequence in consecutive instrument scans of the ten most abundant peaks in the spectrum. The CID fragmentation energy was set at 35%. Dynamic exclusion was enabled with a repeat count of 2 within 10 seconds, a mass list size of 200, an exclusion duration of 350 seconds, the low mass width was 0.5, and the high mass width was 1.5. The raw data files collected using the mass spectrometer were converted

180

to mzXML and MGF files using MassMatrix data conversion tools version 1.3

Daemon by Matrix Science version 2.2.2 (Boston, MA). The .mgf files were searched against the NCBInr database using the following taxonomy and versions: 1) all NCBI version 20100817 limited to 11673899 sequences; 2) taxonomy as Viridaeplantae (green plants) version 20100817 limited to 826678 sequences; 3) taxonomy as Eubacteria

(bacteria) version 20100817 limited to 6609957 sequences; and 4) taxonomy as

Drosophila (fruit flies) version 20100817 limited to 218491 sequences. We also searched .mgf files against the Tribolium 20100809 database limited to 26850 sequences.

Files were searched against a FASTA version of the Tribolium genome (Richards et al.

2008). Trypsin was selected as the digest enzyme with up to two missed cleavages.

Carbamidomethyl and oxidation were set as the fixed and variable modifications.

Peptide and fragment mass tolerances were set to ± 1.2 Da and 0.8 Da respectively. The maximum number of missed cleavages was set to three. Sequence data were also automatically searched against a decoy database in order to avoid false positives.

MASCOT based probability scores were used to evaluate protein identities and were considered correct if the match had a score greater than 60, which indicates identity or significant (P < 0.01) similarity. All protein identities reported in this paper were checked manually to confirm –b and –y ion sequence tags in MS/MS spectra. Any protein that did not pass any of the above criteria was not used in further analyses.

181

Protein Gene Ontology Annotation

Gene Ontology (GO) annotations for biological processes were added to all proteins

(Tables 6.1 and 6.2). Gene ontologies were added by searching individual gi numbers

(obtained from the MASCOT search files) in the Protein Information Resource Database

(National Biomedical Research Foundation, pir.georgetown.edu, Washington, D.C.).

Proteins that were not found in the PIR database were subsequently searched in the NCBI databank to obtain a basic biological understanding of the protein. Most proteins were associated with multiple biological process GO terms and those not associated with a biological process GO term had a molecular function term that was recorded (Tables 6.1 and 6.2). Proteins with no associated GO terms for any category are referred to as miscellaneous proteins (Tables 6.1 and 6.2). A single biological process category was selected for proteins associated with multiple GO terms by choosing the most specific and biologically relevant term to plant defense against herbivores. Green plant and bacterial specific biological process GO categories were loaded onto the QuickGO (Binns et al. 2009) annotation page in order to visualize relationships between biological processes for plant and bacterial proteins independently (Figures 6.2 and 6.3).

Metabolomics

For information regarding the instrument, solvents, methods, and instrument conditions please refer to Chapter 4 (Analysis and Identification of Soluble Phenolics with HPLC-ESI-MS-PDA). The MS detector was optimized previously to obtain

182

maximum yields of [M-H]- ions of oleuropein and verbascoside standards, which are commonly found in phloem extracts of green ash. The optimized MS parameters were: capillary voltage, -80 V; needle voltage, -5 kV (for the negative ion mode; the parameters remained the same in the positive mode except the polarity was reversed). Survey scan was set to detect molecules between 50 – 2000 m/z. All metabolite fractions from oral secretionry extracts were run using both the TurboDDS and full scan mode in negative and positive ion modes. The TurboDDS trigger threshold was set to 2,000 counts (parent ion counts). In the TurboDDS scan mode, MS scan parameters were set on-the-fly by the instrument to detect the most abundant parent ions and obtain maximum yields of those compound fragments. If TurboDDS was triggered, the mode temporarily switched to

MSn mode to perform daughter scans of the putative parent ions. The TurboDDS trigger threshold for daughter ions was 2,000, 200, and 20 ions for MS2, MS3, and MS4 respectively. Full scan parameters were set following the conditions mentioned above.

Full scan chromatograms were overlaid with PDA chromatogram traces at 280 nm to

- match [M-H] parent ions to !max data. Metabolite fractions were run in both the negative and positive ion mode to maximize the number of compounds detected. In this dissertation, we only present compounds detected in the negative ion mode (Table 6.3).

PDA data of individual compounds run on the HPLC-ESI-MS-PDA were analyzed using

PolyView software (Varian). Identities of phenolic compounds present in oral secretion metabolite fractions were based on the congruence of parent and daughter ions, !max, and retention time with data presented in Chapter 4 (Tables 4.1 and 4.2). Data acquisition and processing were all performed using MS Workstation 6.9.2 (Varian).

183

RESULTS

Shotgun Proteomics

Sixteen distinct proteins matching to green plant proteins were identified, while 11 proteins matched to bacterial proteins (Table 6.1 and 6.2). None of the proteins we identified matched genes in the Drosophila or Tribolium genomes. Our annotation led to the identification of 11 separate biological process GO terms associated with proteins matching to green plants, and eight distinct biological process GO terms matching to bacteria (Tables 6.1 and 6.2). One green plant protein (gi 2213425) did not have any associated GO term, but was associated with a molecular function having lactoylglutathione lyase activity. Two protein identities (gi 225425278 and gi

147770315) were not associated with any GO annotation and were therefore classified as miscellaneous (Table 6.1). Two bacterial protein matches (gi 72382545 and gi

229155701) were only associated with the GO annotation for molecular function hydrolase activity and one protein (gi 222823797) was not associated with any GO annotation and was classified as miscellaneous (Table 6.2).

Metabolomics

Analysis of metabolites from EAB oral secretions revealed the presence of 15 compounds (Table 6.3). We were able to identify three compounds (oleuropein, verbascoside, and ligustroside) matching to phenolics commonly found in green ash

184

phloem tissues (Chapter 4). MS data of the [M-H]- and main fragments for compounds detected in the negative ion mode using LC-MS are also presented (Table 6.3).

DISCUSSION

The identification of proteins presumably of plant and bacterial origin, and metabolites matching to ash phloem phenolics as well as unknown metabolites is consistent with components found in the oral secretions of other insect species (Hartley and Lawton 1991, Cardoza et al. 2006, Maffei et al. 2006). In this study we use the term oral secretion to describe the liquid excreted by EAB larvae following removal from ash trees in the field. “Oral secretions” is synonymous with regurgitant, while saliva, on the other hand, refers to material that originates from the salivary glands and should not be confused with the oral secretions analyzed in this study (Felton and Eichenseer 1999).

Oral secretions may be collected by disturbing the insect through handling, squeezing with forceps, or even removing the head of the insect, and arise primarily from the alimentary canal of insects (Felton and Eichenseer 1999). The protein composition of saliva and regurgitant/oral secretions are quite different, but components of saliva can still be found in oral secretions (Felton and Eichenseer 1999). Insect saliva is involved in multiple activities in the plant-insect interaction, which include digestion, lubrication of mouthparts, detoxification, excretion, defense against predators, pH regulation, suppression of host responses, and elicitation of host responses (Ribeiro et al. 1995).

Therefore, interpretation of our data should include a basic understanding that while the

185

majority of components identified can be associated with the alimentary canal of the insect, some components may also be derived from the saliva of the insect.

Unexpectedly, a search against the Drosophila and Tribolium databases yielded poor protein matches, none of which were reliable based on Mascot scores (data not shown).

We would have expected at least some insect-derived salivary proteins in the oral secretions. However, EAB larvae appear to have reduced/vestigial salivary glands, which could explain the absence of insect derived proteins (Om Mittapalli personal communication).

On the other hand, the identification of plant proteins in EAB larval oral secretions is consistent with this fluid originating mostly in the alimentary canal of an insect actively feeding on its host. Most of the plant proteins we identified in the oral secretions of EAB larvae are well characterized in other plant-insect and plant-pathogen interactions, with known roles in signaling, ROS metabolism, cell wall modification, response to stress, and defense against insects. Furthermore, insect mouth parts are known to be contaminated with microbes and the digestive organs of herbivores are known to contain a wide variety of bacterial endosymbionts. EAB is no exception (Vasanthakumar et al. 2008). This diverse endosymbiotic bacterial flora likely accounts for the presence of prokaryotic proteins in our oral secretions (Hartley and Lawton 1991, Spiteller et al. 2000,

Vasanthakumar et al. 2008).

The majority of plant proteins identified in EAB oral secretions are associated with general plant defense responses against herbivores and pathogens (Felton and Eichenseer

1999, Kessler and Baldwin 2002, Maffei et al. 2007a, Howe and Jander 2008, Zhu-

Salzman et al. 2008). The presence of such proteins suggests that green ash responds to

186

EAB larval feeding, but since green ash is highly susceptible, these defense responses are clearly not sufficient to ward off an attack.

Two of the 16 plant proteins are of particular interest. The first protein of interest, a hypothetical protein with lactoylglutathione lyase activity (gi 2213425) found in this study, was constitutively expressed at a significantly (> 5-fold) higher level in resistant

Manchurian ash phloem than in the other three susceptible North American ash species

(Chapter 3, Table 3.5). The second protein, a phenylcoumaran benzylic ether reductase

(PCBER)-like protein identified in EAB oral secretions (gi 169639230), is similar to a protein constitutively expressed at a significantly (> 5-fold) higher level in resistant

Manchurian ash phloem. This suggests that these two enzyme functions are quantitatively linked to resistance, since they were detected at low levels in susceptible species. PCBER enzymes are involved in the synthesis of lignans, a class of secondary metabolites with known roles in plant defense against herbivores (Chapters 3 and 4). The constitutive expression of these proteins in Manchurian ash phloem tissues, and their presence in EAB larval oral secretions points to potentially important roles in the interaction between ash and EAB.

A calcium dependent protein kinase (CDPK) (gi 242054877) was also found. This and similar proteins, along with ROS, play important roles in early signal transduction events associated with plant responses to insect feeding and eventually lead to the production and release of defensive phytochemicals (Maffei et al. 2007b). CDPKs have been recently implicated in the upregulation of plant defenses via direct effects on the transcription factors of defense responses that are independent of phytohormone- signaling pathways (Kanchiswamy et al. 2010). The constitutive expression (based on

187

transcriptomic profiling) of CDPKs was also found to be significantly higher in resistant

Manchurian ash phloem tissues than in green and black ash phloem tissues (Bai et al.

2011), although we did not detect them in our ash phloem proteomics studies (Chapter

3).

Previous studies on tobacco (Nicotiana bentamiana) have shown that transcripts of pectin methyl esterase (PME), another protein found in EAB oral secretions, are induced in host tissues treated with the oral secretions of Manduca sexta (Schmidt et al. 2005).

PME are important cell wall enzymes that can strengthen or loosen and degrade cell walls by acting on pectin (a major component of the middle lamella of plant cell walls)(Micheli

2001). Therefore, PME can have opposing effects on the texture and mechanical properties of plant cell walls. Subsequent studies on PME following application of M. sexta oral secretions to WT and PME-silenced tobacco plants found that plants silenced for PME had decreased levels of jasmonic acid (JA) and JA-isoleucine (phytohormones regulating defense responses against chewing herbivores and necrotrophic pathogens) and increased levels of salicylic acid (a phytohormone associated with defense against piercing insects and biotrophic pathogens), as well as reduced levels of trypsin inhibitors

(the primary defense against M. sexta in tobacco). Therefore, the authors concluded that

PME contributes to the induction of antiherbivore defenses indirectly by affecting cell wall properties (Korner et al. 2009). Conversely, another protein detected in EAB oral secretions (gi 225425278) was identified as a glycosyltransferase belonging to the GT47 protein family (Zhong and Ye 2003). This family of proteins in plants is involved in the synthesis of pectin (Zhong and Ye 2003) and may thus be associated with the PME present in EAB oral secretions. Degradation of plant cells is hypothesized to enhance

188

herbivore performance by influencing the efficiency of food utilization (Korner et al.

2009).

!- Oxidative stress resulting from the generation of ROS such as superoxide anion (O2 ) and hydrogen peroxide (H2O2) is commonly associated with plant responses to abiotic and biotic stress (Foyer and Noctor 2005). In plants, the enzymes peroxidase (POD) and superoxide dismutase (SOD) are part of the first lines of antioxidant defense (Maffei et

!- !- al. 2007b). POD catalyzes the initial formation of O2 and SOD catalyzes O2 into hydrogen peroxide (H2O2). H2O2 can act as a signaling molecule in plants following insect attack or H2O2 levels can be elevated as long as an insect attack persists, which can have direct detrimental effects on the insect (Bi and Felton 1995, Gatehouse 2002, Maffei et al. 2006). PODs also catalyze lignification in plants (for a discussion of the role of lignin in plant defense against herbivores please refer to Chapter 1)(Ros Barcelo 1997,

Almagro et al. 2009). Both plant enzyme functions were identified in the oral secretions of EAB larvae, suggesting an important role for ROS in the interaction between ash and

EAB.

A heat shock protein (HSP70) matching to a plant protein was also identified in EAB oral secretions. HSP70s act as chaperones to protect cells from thermal and oxidative stress (Tkalec et al. 2007). HSP70s may be produced in green ash tissues as a means of protection against the toxic effects of ROS accumulation following insect attack.

Toxic plant proteins can affect insect pests by directly disrupting the midgut lining

[peritrophic membrane (PM)], which interferes with digestion and absorption of nutrients, ultimately leading to death of the pest (Zhu-Salzman et al. 2008). There are four types of toxic plant proteins to insects. Three of the four types (Fitches and

189

Gatehouse 1998) were identified in the oral secretions of EAB larvae: 1) a precursor of a cysteine protease, which is known to permeabilize the PM of insect pests (most likely by directly degrading PM proteins)(Mohan et al. 2006); 2) a leucine aminopeptidase, which is thought to function in a similar manner by directly damaging the insect midgut (Felton

2005); and 3) chitinases, of which two classes were detected in EAB oral secretions. The mechanisms of action of chitinases against insects is not well defined, yet in some cases, chitinases have been observed to be an effective antiherbivore defense (Lawrence and

Novak 2006, Zhu-Salzman et al. 2008). Lectins are the only type of proteins toxic to herbivores (Fitches and Gatehouse 1998) that were not detected in the oral secretions of

EAB larvae. The presence of a diverse array of toxic plant proteins in EAB oral secretions highlights a presumably limited role for known toxic plant proteins in green ash defense against EAB.

Other plant proteins identified in EAB oral secretions include an aspartate oxidase (gi

224058437), a triose phosphate isomer (gi 115440977), a putative mitochondrial NAD- dependent malate dehydrogenase (gi 21388550), and a predicted GTPase (gi 147770315).

While a literature search did not reveal any significant roles for these proteins in plant defense, their potential importance in the interaction between ash and EAB should not be overlooked. Identification of defense related proteins in oral secretions of EAB larvae feeding on susceptible green ash has the potential to give invaluable insights into the types of active defense mechanisms employed by green ash in its attempt to resist attack by EAB. The identification of resistance mechanisms in green ash, though ineffective against EAB, will help to inform our understanding of resistance mechanisms in

Manchurian ash. Also, knowledge about the types of defense mechanisms that green ash

190

deploys against EAB could help in understanding the nature of the presumed resistance of lingering green ash trees in the field (Koch et al. 2010).

The identification of bacterial proteins in EAB larval oral secretions may seem counterintuitive, but a growing body of literature supports the presence of microbes in insect mouthparts and evidence suggests they may even be a source of potential elicitors of plant defense responses (Hartley and Lawton 1991, Spiteller et al. 2000, Cardoza et al.

2006). Microbes contaminating insect mouthparts and colonizing insect digestive organs have been implicated in the production of compounds capable of inducing plant volatiles

(Spiteller et al. 2000). Also, bacteria may interfere with plant defense responses by activating defense pathways specific to bacterial pathogens that are antagonistic to defense responses against chewing insects (Biere et al. 2004, Glazebrook 2005, Beckers and Spoel 2006, Spoel et al. 2007). Investigations on EAB gut microbiota across all life stages identified 132 separate taxonomic units associated with EAB, some of which have the potential to function mutualistically with their host (Vasanthakumar et al. 2008). In this study we identified 11 distinct proteins matching bacterial proteins (Table 6.2). One protein (gi 87311876) was of particular interest, and was identified as a bacterial catalase.

Catalases catalyze the decomposition of H2O2 to water and oxygen (Willekens et al.

1997). The presence of a bacterial catalase in EAB oral secretions suggests the possibility of a mutualistic relationship with EAB larvae, which may enable them to overcome the ROS generated by green ash (see above). Bacteria may indeed play a critical role in the interaction between ash and EAB, enabling the insect to successfully overcome the defenses of susceptible hosts.

Three metabolites in EAB oral secretions were matched to plant phenolics known to

191

occur constitutively in green ash phloem tissues (Table 6.3). Twelve metabolites were not identified. The presence of the plant phenolics oleuropein, ligustroside, and verbascoside in EAB oral secretions suggests that their roles in defense against EAB is limited, even though the oleuropein aglycone is active in other plant systems in defense against herbivores (Konno et al. 1999, Eyles et al. 2007). Glycosylated plant phenolics

(of which oleuropein, ligustroside, and verbascoside all are examples) may not be active until the sugar moiety is removed.

Unidentified metabolites could represent EAB specific elicitors and therefore their identities should be elucidated. Low molecular weight elicitors (FACs) have been found in the oral secretions and oviposition fluids of several insect species. FACs such as volicitin (lepidopterans) and bruchins (weevils) play critical roles in the elicitation of plant defense responses (Alborn et al. 1997, Doss et al. 2000, Oliver et al. 2002). The identification of an EAB specific elicitor would allow for the identification of ash specific induced mechanisms of resistance against EAB.

In conclusion, in this study we analyzed EAB 4th instar larval oral secretions using proteomic and metabolomic approaches. The identification of plant and bacterial proteins with important and defined roles in other plant-insect interactions highlights the efficacy of our approach and its value in advancing our knowledge of the EAB-ash system, while identification of low molecular weight metabolites holds potential for the characterization of EAB specific elicitors responsible for inducing defense responses in ash. Identification of EAB elicitors will greatly expedite our understanding of the ash-

EAB interaction and could potentially lead to ash receptors involved in the defense response.

192

ACKNOWLEDGEMENTS

I would like to thank Ronald Batallas for his significant contribution to this project.

Ronald helped with the collection of EAB oral secretions. He also extracted and prepared oral secretions for both proteomic and metabolomics analyses. Preliminary organization of the data was also contributed by Ronald. Zhifeng Zheng should also be mentioned for his help with oral secretion collection and the contribution of excellent images of EAB larvae oral secretions in the field (Figure 6.1). Shotgun proteomic sample analyses and Mascot database searches were performed at the Central Campus Instrument

Center (CCIC) by Sasha Popova-Butler.

193

194

195

Figure 6.3. GO annotation term relationships obtained from the QuickGO resource (Binns et al. 2009) for biological functions of 9 terms associated with proteins identified as bacterial proteins in EAB oral secretions. Organization of proteins in Table 6.2 are based on these relationships, with the parent category serving as a point of organization for the daughter terms.

195

Table 6.1. Plant proteins identified by MS and MASCOT analysis in EAB oral secretions with a Mascot score of 60 or greater.

MASCOT Gi score Nominal Sequence accession Protein Identification Organism [Number of Mass (Da) coverage (%) Number novel Peptides] Gene Ontology Annotation – Biological Process Cellular Component Organization or Biogenesis

Cell wall modification 20455195 Pectinesterase Daucus carota 34404 113 [2] 7

Metabolic Process Phenylcoumaran benzylic ether reductase-like Linum strictum subsp. 196 169639230 33629 177 [2] 8 protein corymbulosum Oryza sativa - Japonica 115440977 Os01g0841600 (Triose phosphate isomerase) 27484 124 [1] 8 Group

Carbohydrate metabolic process Oryza sativa - Japonica 2425170 Basic class III chitinase OsChib3b 32582 67 [1] 10 Group

Chitin Catabolic Process 5880843 Chitinase precursor Petroselinum crispum 29430 121 [1] 7

Malate Metabolic Process Putative mitochondrial NAD-dependent malate 21388550 Solanum tuberosum 36429 93 [2] 20 dehydrogenase

Oxidation reduction 224058437 Predicted protein (Aspartate Oxidase) Populus trichocarpa 54253 61 [1] 2

Protein phosphorylation

196

Hypothetical protein SORBIDRAFT_03g038870 242054877 Sorghum bicolor 61203 76 [1] 4 (Calcium dependent protein kinase)

Proteolysis 2511693 Cysteine proteinase precursor Phaseolus vulgaris 51418 96 [1] 3 2980774 Leucyl aminopeptidase-like protein Arabidopsis thaliana 55366 81 [3] 6

Superoxide Metabolic Process

160347126 Allergen Ole e 5 Olea europaea 15354 72 [3] 66

Response to Stimulus

Response to Oxidative Stress 147820487 Hypothetical protein 197 Vitis vinifera 30242 173 [2] 17 (Secretory Peroxidase – Class III)

Response to Stress 30575576 HSP70 Citrus x paradisi 10699 111 [1] 29

Gene Ontology Annotation – Molecular Function

Lactoylglutathione lyase activity 2213425 Hypothetical protein Citrus x paradisi 32737 225 [3] 13

Gene Ontology Annotation – Miscellaneous 225425278 Hypothetical protein isoform 2 (Glycosyltransferase) Vitis vinifera 54495 201 [3] 8

147770315 Hypothetical protein (Predicted GTPase) Vitis vinifera 39976 62 [2] 5

197

Table 6.2. Bacterial proteins identified by MS and MASCOT analysis in EAB oral secretions with a Mascot score of 60 or greater.

MASCOT Gi accession Nominal score [Number Sequence Protein Identification Organism Number Mass (Da) of novel Coverage (%) Peptides] Gene Ontology Annotation – Biological Process Metabolic Process Glycine Metabolic Process Zunongwangia profunda 295133078 Serine hydroxymethyltransferase 46565 65 [1] 2 SM-A87

Oxidation-Reduction Process Herbaspirillum 300311100 NADH dehydrogenase I subunit F 47976 60 [1] 4 seropedicae SmR1

198 Regulation of Transcription (DNA-dependent) Achromobacter 293604654 LuxR family transcriptional regulator 23680 68 [1] 4 piechaudii ATCC 43553

Response to Stimulus Chemotaxis Bacillus anthracis str. 30261768 Flagellar Motor Switch Protein 13002 62 [1] 10 Ames

Defense response 282863551 Transcriptional regulator, winged helix family Streptomyces sp. ACTE 121513 65 [1] 1

DNA Repair Corynebacterium Putative uncharacterized protein (Endonuclease 237785928 kroppenstedtii DSM 13006 67 [1] 6 III) 44385

198

Regulation of DNA repair Edwardsiella tarda 294634756 Regulatory protein RecX 18285 62 [1] 11 ATCC 23685

Response to Oxidative Stress Probable catalase hydroperoxidase hpii Blastopirellula marina 87311876 65354 84 [1] 3 oxidoreductase protein DSM 3645

Gene Ontology Annotation – Molecular Function

Hydrolase activity Serine peptidase (Alpha/beta hydrolase superfamily) Prochlorococcus marinus 72382545 fused to N-terminal uncharacterized domain specific 54816 68 [1] 2 str. NATL2A to cyanobacteria Bacillus cereus ATCC 229155701 Putative uncharacterized protein 42253 65 [1] 2 4342 199

Gene Ontology Annotation – Miscellaneous Campylobacter lari 222823797 Putative uncharacterized protein 61819 65 [1] 3 RM2100

199

Table 6.3. Metabolites identified by LC-MS-PDA from EAB oral secretions of 4th instar larvae feeding on susceptible green ash in the field. Numbers in bold text represent the dominant ion on which subsequent fragmentation patterns are based.

MS4 Fragments MS2 Fragments m/z MS3 Fragments m/z RTa [M-H]-b m/z (decreasing Putative Identity (decreasing abundance) (decreasing abundance) abundance) 38.41 539 377, 275, 291, 359 275, 291, 359, 345, 179 111, 155, 137 Oleuropein 39.33 623 461, 462, 315, 463, 459, 477 315, 297, 315 Verbascoside 40.56 987 788, 805, 403, 877, 546 41.80 621 459, 460, 461, 469 204, 206 44.30 523 361, 291, 259 291, 359, 223 110, 138, 125, 136 Ligustroside 45.05 663 501, 297, 502, 503, 509 297, 315, 279 45.62 857 471, 695, 625, 593 299, 383, 413 46.89 661 509, 499, 459, 510, 501, 365 221, 203, 278, 227 200 54.91 369 193, 135, 293, 251, 311, 277 55.12 621 562, 427, 505, 387, 369 387, 444, 528, 329 55.71 919 725, 723, 727, 685, 726 491, 489, 668, 526 56.21 855 797, 739, 661, 621, 737 739, 620, 621, 529 57.42 1387 1193, 959, 725, 1153, 958, 1195 60.28 793 537, 538, 555 225, 463 61.58 563 325, 297, 298, 282 296, 265, 297, 306 aRetention time bMolecular ion

200

CHAPTER 7

CONCLUSIONS AND SUGGESTIONS FOR FUTURE WORK

The central goal of this dissertation research was to identify mechanisms of resistance in ash against EAB. Overall, the major findings, conclusions, and suggestions for future work resulting from my dissertation were:

1. Proteomic and phenolic analyses of constitutive ash phloem tissues support

the interpretation that species displaying different resistance phenotypes

should be examined in the context of their phylogenetic relationships

(Chapters 3 and 4). Future studies will lead more quickly to the

identification of resistance mechanisms in ash if a comparative species

approach is informed by the evolutionary relationships between ash species.

2. Specific constitutive proteins and phenolic compounds are strongly associated

with the resistant Manchurian ash (Chapters 3 and 4). The identification of

traits associated with Manchurian ash provides a starting point that can guide

future investigations into constitutive mechanisms of resistance. Future

investigations into both proteins and phenolics can be answered via functional

genomics approaches and bioassays that directly test the effect of putative

defense traits in Manchurian ash against living EAB larvae.

201

3. MeJA is a viable management option for EAB that can be developed into a

practical and effective tool for pesticide-free management of EAB in the field

(Chapters 2 and Chapter 5). Future studies will need to develop a cost

effective and biologically relevant treatment for ash trees that can be used

commercially.

4. Oral secretions of EAB contain a variety of proteins and metabolites.

Therefore, investigations into EAB oral secretions provide a starting point to

study the interactions between EAB – Fraxinus spp. at the molecular level

(Chapter 6).

Previous studies of phloem phenolic chemistry compared resistant Manchurian ash to susceptible green and white ashes and found obvious differences between the constitutive phloem phenolic chemistry of the three species under investigation (Eyles et al. 2007). In this dissertation I expanded upon these early studies and helped lay a foundation upon which future work can build. Overall, I approached my dissertation research using the conceptual framework of the three components of defense against phloem feeding buprestids: 1) constitutive (pre-formed) physical and chemical defenses; 2) induced chemical defenses such as secondary metabolites, phenolic polymers such as lignin, and defensive proteins that may have negative impacts on the invading insect pest that slow the growth of the insect; and 3) formation of a wound (necrophylactic) periderm (i.e., callus tissue) around the insect pest that isolates the larvae and re-establishes integrity of the damaged regions. I was not able to completely research each of these components but

202

my work provides a roadmap for future studies to further address the different components of tree resistance to phloem feeding buprestids.

In Chapter 3, phylogenetic distinctions between the ash species under investigation were confirmed in the differences and similarities of their total constitutive phloem proteomes (Figure 3.4 and Table 3.2). Furthermore, four constitutive proteins were identified that were significantly more highly expressed (> 5-fold; P < 0.05) in phloem tissues of resistant Manchurian ash than in susceptible ash species, indicating that they may play roles in defense (Table 3.3). These four proteins provide a starting point for functional genomics approaches to begin and address constitutive mechanisms of ash resistance to EAB. Before that can be done, however, studies should expand the number of genotypes and species under investigation to first confirm if the expression of these genes at the transcript (qRT-PCR) or protein (Western blot) levels is consistently higher in resistant accessions (Koch, personal communication). If the associations between the expression levels of these putative resistance genes and the resistant phenotype are confirmed, a concerted effort to clone and directly test these genes against EAB larvae should be undertaken. Transformation protocols are now available for black, green, and white ashes (Du and Pijut 2009, Pijut personal communication). Insertion of these genes into the susceptible ash species, or knockouts of Manchurian ash should be pursued so their definitive roles in defense can be identified. Alternate, supporting approaches may also include testing purified forms of these proteins in artificial diets.

203

The studies described in Chapter 4 indicated that differences and similarities of constitutive phloem phenolic chemistry of the different ash species under investigation were also reflective of their phylogenetic relationships (Figure 4.1 and 4.5 and Table 4.1 and 4.2). Sixty-eight constitutive phenolic compounds were characterized across the six ash species, but only two compounds, pinoresinol dihexoside and unknown coumarin 3, were found to be unique in the phloem tissues of Manchurian ash and may therefore be related to the resistant phenotype. Their effects against EAB larvae should be tested via an artificial diet. However, the modes by which phenolics affect herbivores are diverse, and the synergistic or antagonistic effects of certain phenolics of interest interacting with other molecular components in planta should be considered in artificial diet studies

(Felton et al. 1992, Appel 1993, Duffey and Stout 1996).

Early in my dissertation research, I attempted to address the role of induced resistance mechanisms. Since no viable protocol to directly challenge live trees with EAB larvae was available at the time, I used a larval homogenate as a proxy, but this proved to be inadequate, since the trees did not respond to the treatment (Chapter 5). It was not until very recently that the use of EAB eggs became a viable option as a means to investigate induced defenses.

A preliminary experiment was undertaken in the summer of 2010 in which five-sets of

EAB eggs (each set containing between 1-5 EAB eggs) were applied to the branches of water stressed and unstressed Manchurian and black ash trees. Only two tree replicates were used at the time for each species x treatment combination, due to a shortage of EAB

204

eggs, and thus no generalizations can be made at this point. However, EAB neonates that hatched and fed into water stressed Manchurian ash trees died as soon as they began to feed (Figures 7.1 and 7.2), suggesting that water stress was enhancing the defensive capabilities or toxicity of Manchurian ash, but not black ash. A similar experiment may be very useful to identify the mechanism responsible for EAB larval mortality and thus

Manchurian ash resistance.

Work on induced defenses continued as described in Chapter 6, in which I characterized the oral secretion proteome and metabolome of EAB 4th instar larvae feeding on green ash in the field. By studying EAB oral secretions, we have gained an initial understanding of the types of potential defenses originating from the plant, and the types of counter defenses potentially benefiting the insect. Future investigations should focus on using EAB oral secretions as a means to induce and elucidate defense mechanisms in ash, an approach that has been very successful in other plant-insect systems (Alborn et al. 1997, Spiteller et al. 2000, Cardoza et al. 2006, Howe and Jander

2008, Korner et al. 2009, Philippe et al. 2010).

Wound periderms have been implicated in several other tree-buprestid associations as being the ultimate mechanism of resistance of angiospermous trees to phloem feeding wood-borers (Anderson 1944, Dunn 1990, Miller 1991). It has been suggested that wound periderms have increased levels of defensive chemicals that aid in defense (Eyles et al. 2007). However, my personal observations of necrophylatic periderms against

EAB in the field occur at a much lower frequency than one would expect if this were

205

Figure 7.1. Black circles show the entrance holes of EAB neonates in an artificial inoculation experiment using eggs inserted under a flap of outer bark. Neonates immediately died upon initiation of feeding on drought stressed Manchurian ash trees.

206

Figure 7.2. Close-up of the posterior end of a dead EAB neonate that fed on a drought stressed Manchurian ash tree. The arrow is pointing directly at the dead EAB neonate.

207

Figure 7.3. The gall-like structure on the main stem of this Manchurian ash tree is the result of a wound periderm forming over an EAB larval gallery.

208

such a critical mechanism of resistance against EAB. For instance, over the course of the three years that I worked in the ash plot in Bowling Green, Ohio, I only observed two out of 32 Manchurian ash trees with obvious signs of wound periderms associated with larval activity (Figure 7.3). Dissection of one of the two wound periderms revealed an EAB gallery that had been completely overgrown with callus tissue, phloem tissues apparently reconnected around the gallery, with no obvious, overall adverse effect on the tree

(Figure 7.4). These observations suggest that while wound periderms are probably not the primary mechanism of resistance in Manchurian ash (as they are too infrequent), they do occur and therefore should still be investigated further to determine the regulatory mechanisms governing their initiation.

Based on these limited observations, and on existing hypotheses (Anderson 1944,

Dunn 1990, Miller 1991), I propose that wound periderms are indeed the final line of defense for Manchurian ash in the field and that its resistance is due to a combination of constitutive and induced physical and chemical defenses that work together at different points in the life cycle of EAB larvae to stop their advance.

In its native host range, EAB persists at low levels by colonizing stressed or dying

Manchurian ash trees in the field (Liu et al. 2007, Wei et al. 2007). Therefore, the success or failure of the three components of defense against phloem feeding buprestids are contingent upon the physiological state of the host. For instance, in Chapter 2 we tested the effects of girdling and MeJA priming on four species of ash. We found that: 1) girdling enhanced susceptiblity of tissues below the girdle to colonization by EAB,

209

Figure 7.4. Anatomy of a Manchurian ash wound periderm. The numbers 1 – 3 are used to denote the innermost layer (1) expanding out to the outermost layer of the wound periderm (3). (1) Is the end of an EAB larval gallery. Notice the thick callus tissue that formed at the end of the gallery and surrounding the gallery. The EAB gallery itself was initially created at the cambial-phloem interface, but callus tissue completely grew over and covered the initial EAB gallery (2). On top of the original gallery and callus tissue the phloem layer (3) continued to grow seemingly uninterrupted by the damage caused by the EAB larvae.

210

except in Manchurian ash; and 2) trees treated with MeJA were able to resist colonization at levels comparable to trees treated with a protective insecticide. In birch, tissues below a girdle have reduced concentrations of total phenolics and a slower rate of wound periderm formation, resulting from increased carbon stress due to isolation from the flow of photosynthate produced in the foliage (Muilenburg 2010). Therefore, differences in phloem chemistry above versus below girdled tissues can give insight into potential resistance mechanisms against EAB. Although Manchurian ash was not made more susceptible following girdling in our experiment, more time (greater than the 2-3 months used in this study) may be required before significant changes can be observed. Future investigations should focus on characterizing the chemical differences between above versus below girdled tissues, with the ultimate goal of identifying putative resistance related traits in resistant ash species.

Other work described in this dissertation (Chapter 5) has demonstrated that MeJA application to outer bark of ash elicits significant changes in phenolic chemistry. Thus,

MeJA has the potential to be developed into a viable treatment for EAB that is ecologically friendly (Chapter 2). Future work should expand upon this investigation using trees of varying ages and sizes to determine if the effects of MeJA have an ontogenetic or size dependent limit. One approach may be to develop the dosage of

MeJA and method of application to ash trees that provide consistent and effective protection against EAB and, concurrently, elucidate the mechanism by which MeJA provides protection. My work has already begun to address this latter point. Of the 31 compounds investigated in white and Manchurian ash in Chapter 5, 17 were significantly

211

affected by treatment with MeJA. Thus, future work could focus on testing the effects of these 17 compounds in an artificial diet and/or investigate the responses of specific ash phloem genes upon exogenous application of MeJA.

212

BIBLIOGRAPHY

Agius, A. C., D. M. McCullough, D. A. Cappaert. 2005. Published. Host range and preference of the emerald ash borer in North America: Preliminary results., pp. 28-29. In V. Mastro, R. Reardon (ed.), Emerald ash borer research and technology development meeting, 2005, Morgantown, WV. USDA Forest Services. Agriculture, O. D. of. 2010. Remaining Counties Added to Ohio Emerald Ash Borer Quarantine. In O. D. o. Agriculture [ed.]. Office of Communication, Reynoldsburg, OH. Agriculture, U. S. D. of. 2006. Questions and Answers: Emerald Ash Borer Quarantine. Akai, S., and M. Fukutomi. 1980. Preforemed internal physical defences., pp. 139-159. In J. Horsfall and C. EB (eds.), Plant Disease. An advanced treatise. , vol. Volume V. How Plants defend themselves. Academic Press Inc., New York. Akiyama, T., S. Jin, M. Yoshida, T. Hoshino, R. Opassiri, and J. R. K. Cairns. 2009. Expression of an endo-(1,3;1,4)-beta-glucanase in response to wounding, methyl jasmonate, abscisic acid and ethephon in rice seedlings. J Plant Physiol 166: 1814-1825. Alborn, H. T., T. C. J. Turlings, T. H. Jones, G. Stenhagen, J. H. Loughrin, and J. H. Tumlinson. 1997. An elicitor of plant volatiles from beet armyworm oral secretion. Science 276: 945-949. Almagro, L., L. V. G. Ros, S. Belchi-Navarro, R. Bru, A. R. Barcelo, and M. A. Pedreno. 2009. Class III peroxidases in plant defence reactions. Journal of Experimental Botany 60: 377-390. Alvarez, M. E., R. I. Pennell, P. J. Meijer, A. Ishikawa, R. A. Dixon, and C. Lamb. 1998. Reactive oxygen intermediates mediate a systemic signal network in the establishment of plant immunity. Cell 92: 773-784. Anderson, R. F. 1944. The Relation Between host condition and Attacks by the bronzed birch borer. Journal of Economic Entomology 37: 588-596. Anulewicz, A. C., D. G. McCullough, and D. L. Miller. 2006. Oviposition and Development of Emerald Ash Borer (Agrilus planipennis) (Coleoptera: Buprestidae) on hosts and potential hosts in no-choice bioassays. The Great Lakes Entomologist 39: 99-112. Anulewicz, A. C., D. G. McCullough, and D. L. Cappaert. 2007. Emerald ash borer (Agrilus planipennis) density and canopy dieback in three North American ash species. Arboriculture and Urban Forestry 33: 338-338. Anulewicz, A. C., D. G. McCullough, D. L. Cappaert, and T. M. Poland. 2008. Host Range of the Emerald Ash Borer (Agrilus planipennis Fairmaire) (Coleoptera: Buprestidae) in North America: Results of Multiple-Choice Field Experiments. Environmental Entomology 37: 230-241. Appel, H. M. 1993. Phenolics in ecological interactions – importance of oxidation. J. Chem. Ecol. 19: 1521-1552. Arimura, G., D. P. W. Huber, and J. Bohlmann. 2004a. Forest tent caterpillars (Malacosoma disstria) induce local and systemic diurnal emissions of terpenoid volatiles in hybrid poplar (Populus trichocarpa x deltoides): cDNA cloning, 213

functional characterization, and patterns of gene expression of (-)-germacrene D synthase, PtdTPS1. Plant J. 37: 603-616. Arimura, G., R. Ozawa, S. Kugimiya, J. Takabayashi, and J. Bohlmann. 2004b. Herbivore-induced defense response in a model legume. Two-spotted spider mites induce emission of (E)-beta-ocimene and transcript accumulation of (E)-beta- ocimene synthase in Lotus japonicus. Plant Physiology 135: 1976-1983. Asada, K. 1999. The water-water cycle in chloroplasts: Scavenging of active oxygens and dissipation of excess photons. Annu. Rev. Plant Physiol. Plant Molec. Biol. 50: 601-639. Aukema, J. E., D. G. McCullough, B. Von Holle, A. M. Liebhold, K. Britton, and S. J. Frankel. 2010. Historical Accumulation of Nonindigenous Forest Pests in the Continental United States. Bioscience 60: 886-897. Bai, X. D., L. Rivera-Vega, P. Mamidala, P. Bonello, D. A. Herms, and O. Mittapalli. 2011. Transcriptomic Signatures of Ash (Fraxinus spp.) Phloem. PLoS One 6: -. Baldwin, I. T. 1998. Jasmonate-induced responses are costly but benefit plants under attack in native populations. Proc. Natl. Acad. Sci. U. S. A. 95: 8113-8118. Baldwin, I. T., E. A. Schmelz, and T. E. Ohnmeiss. 1994. Wound-Induced Changes in Root and Shoot Jasmonic Acid Pools Correlate with Induced Nicotine Synthesis in Nicotiana-Sylvestris Spegazzini and Comes. J. Chem. Ecol. 20: 2139-2157. Ballare, C. L. 2011. Jasmonate-induced defenses: a tale of intelligence, collaborators and rascals. Trends in Plant Science In Press, Corrected Proof. Baranchikov, Y., E. Mozolevskaya, G. Yurchenko, and M. Kenis. 2008. Occurrence of the emerald ash borer, Agrilus planipennis in Russia and its potential impact on European forestry. EPPO Bulletin 38: 233-238. Barrell, D., E. Dimmer, R. P. Huntley, D. Binns, C. O'Donovan, and R. Apweiler. 2009. The GOA database in 2009--an integrated Gene Ontology Annotation resource. Nucl. Acids Res. 37: D396-403. Barter, G. W. 1957. Studies of the Bronze Birch Borer, Agrilus anxius Gory, in New Brunswick. The Canadian Entomologist: 12-36. Beckers, G. J. M., and S. H. Spoel. 2006. Fine-tuning plant defence signalling: Salicylate versus jasmonate. Plant Biol. 8: 1-10. Bennett, R. N., and R. M. Wallsgrove. 1994. Secondary metabolites in plant defence mechanisms. New Phytol. 127: 617-633. Bentham, G. 1876. Oleaceae, vol. 2. Bhadauria, V., S. Banniza, L.-X. Wang, Y.-D. Wei, and Y.-L. Peng. 2009. Proteomic studies of phytopathogenic fungi, oomycetes and their interactions with hosts. European Journal of Plant Pathology 126: 81-95. Bi, J. L., and G. W. Felton. 1995. Foliar oxidative stress and insect herbivory - primary compounds, secondary metabolites, and reactive oxygen species as components of induced resistance. J. Chem. Ecol. 21: 1511-1530. Biere, A., H. B. Marak, and J. M. M. van Damme. 2004. Plant chemical defense against herbivores and pathogens: generalized defense or trade-offs? Oecologia 140: 430-441.

214

Binns, D., E. Dimmer, R. Huntley, D. Barrell, C. O'Donovan, and R. Apweiler. 2009. QuickGO: a web-based tool for Gene Ontology searching. Bioinformatics 25: 3045-3046. Black, C. A., R. Karban, L. D. Godfrey, J. Granett, and W. E. Chaney. 2003. Jasmonic acid: A vaccine against leafminers (Diptera : Agromyzidae) in celery. Environmental Entomology 32: 1196-1202. Blake, P. S., J. M. Taylor, and W. E. Finch-Savage. 2002. Identification of abscisic acid, indole-3-acetic acid, jasmonic acid, indole-3-acetonitrile, methyl jasmonate and gibberellins in developing, dormant and stratified seeds of ash (Fraxinus excelsior). Plant Growth Regul 37: 119-125. Bonello, P., and J. T. Blodgett. 2003. Pinus nigra-Sphaeropsis sapinea as a model pathosystem to investigate local and systemic effects of fungal infection of pines. Physiological and Molecular Plant Pathology 63: 249-261. Bonello, P., T. R. Gordon, D. A. Herms, D. L. Wood, and N. Erbilgin. 2006. Nature and ecological implications of pathogen-induced systemic resistance in conifers: A novel hypothesis. Physiological and Molecular Plant Pathology 68: 95-104. Borg-Karlson, A. K., G. Nordlander, A. Mudalige, H. Nordenhem, and C. R. Unelius. 2006. Antifeedants in the feces of the pine weevil Hylobius abietis: Identification and biological activity. J. Chem. Ecol. 32: 943-957. Bourgeois, J., and L. Malek. 1991. Purification and Characterization of an Aspartyl Proteinase from Dry Jack Pine Seeds. Seed Science Research 1: 139-147. Breiteneder, H., K. Pettenburger, A. Bito, R. Valenta, D. Kraft, H. Rumpold, O. Scheiner, and M. Breitenbach. 1989. The gene coding for the major birch pollen allergen Betv1, is highly homologous to a pea disease resistance response gene. The EMBO Journal 8: 1935-1938. Brobey, R. K. B., F. C. Mei, X. Cheng, and L. Soong. 2006. Comparative two- dimensional gel electrophoresis maps for promastigotes of Leishmania amazonensis and Leishmania major. Braz J Infect Dis 10: 1-6. Brodeur-Campbell, S. E., J. A. Vucetich, D. L. Richter, T. A. Waite, J. N. Rosemier, and C. J. Tsai. 2006. Insect herbivory on low-lignin transgenic aspen. Environmental Entomology 35: 1696-1701. Bryant, J. P., R. K. Swihart, P. B. Reichardt, L. Newton. 1994. Biogeography of woody plant chemical defense against snowshoe hare browsing: Comparison of Alaska and eastern North America. Oikos 70: 385-395. Burlat, V., M. Kwon, L. B. Davin, and N. G. Lewis. 2001. Dirigent proteins and dirigent sites in lignifying tissues. Phytochemistry 57: 883-897. Cabral, M. M. O., P. Azambuja, O. R. Gottlieb, and E. S. Garcia. 2000. Effects of some lignans and neolignans on the development and excretion of Rhodnius prolixus. Fitoterapia 71: 1-9. Call, V. B., and D. L. Dilcher. 1992. Investigations of Angiosperms from the Eocene of Southeastern North-America - Samaras of Fraxinus-Wilcoxiana Berry. Rev Palaeobot Palyno 74: 249-266. Cappaert, D., Deborah G. McCullough, Theresa M. Poland, N. Siegert 2005. Emerald ash borer in North America: a research and regulatory challenge. American Entomologist 51: 152-165.

215

Cardoso, S. M., S. Guyot, N. Marnet, J. A. Lopes-da-Silva, C. M. G. C. Renard, and M. A. Coimbra. 2005. Characterisation of phenolic extracts from olive pulp and olive pomace by electrospray mass spectrometry. J Sci Food Agr 85: 21-32. Cardoza, Y. J., K. D. Klepzig, and K. F. Raffa. 2006. Bacteria in oral secretions of an endophytic insect inhibit antagonistic fungi. Ecological Entomology 31: 636-645. Carpentier, S. C., B. Panis, A. Vertommen, R. Swennen, K. Sergeant, J. Renaut, K. Laukens, E. Witters, B. Samyn, and B. Devreese. 2008. Proteome analysis of non-model plants: A challenging but powerful approach. Mass Spectrometry Reviews 27: 354-377. Centre, W. C. M. 1996. Ectodemia castaneae (American chestnut moth), pp. http://www.iucnredlist.org. In I. 2010 [ed.], IUCN Red List of Threatened Species. Chen, H., B. C. McCaig, M. Melotto, S. Y. He, and G. A. Howe. 2004. Regulation of plant arginase by wounding, jasmonate, and the phytotoxin coronatine. Journal of Biological Chemistry 279: 45998-46007. Chen, H., C. G. Wilkerson, J. A. Kuchar, B. S. Phinney, and G. A. Howe. 2005. Jasmonate-inducible plant enzymes degrade essential amino acids in the herbivore midgut. Proc. Natl. Acad. Sci. U. S. A. 102: 19237-19237. Chen, M. S. 2008. Inducible direct plant defense against insect herbivores: A review. Insect Sci. 15: 101-114. Chinese Academy of Sciences, I. o. Z. 1986. Agrilus marcopli Obenberger, China Agricultural Press, Beijing, China. Cipollini, D., Q. Wang, J. Whitehill, J. Powell, P. Bonello, and D. Herms. 2011. Distinguishing Defensive Characteristics in the Phloem of Ash Species Resistant and Susceptible to Emerald Ash Borer. J. Chem. Ecol.: 37: 450-459. Coley, P. D. 1983. Herbivory and Defensive Characteristics of Tree Species in a Lowland Tropical Forest. Ecol Monogr 53: 209-233. Cordeiro, M. C., M. S. Pais, and P. E. Brodelius. 1994. Tissue-specific expression of multible forms of cyprosin (aspartic proteinase) in flowers of Cynara- cardunculus. Physiol. Plant. 92: 645-653. Davies, D. R. 1990. The structure and function of the aspartic proteinases. Annual Review of Biophysics and Biophysical Chemistry 19: 189-215. Davis, P. H., M. Chen, X. Zhang, C. G. Clark, R. R. Townsend, and S. L. Stanley. 2009. Proteomic comparison of Entamoeba histolytica and Entamoeba dispar and the role of E. histolytica alcohol dehydrogenase 3 in virulence. PLoS Neglected Tropical Diseases 3: e415. DeGomez, T. E., C. J. Hayes, J. A. Anhold, J. D. McMillin, K. M. Clancy, and P. P. Bosu. 2006. Evaluation of insecticides for protecting southwestern ponderosa pines from attack by engraver beetles (Coleoptera : Curculionidae : Scolytinae). Journal of Economic Entomology 99: 393-400. Dodds, H. D., and G. W. Otis. 2009. Breeding Birds of Mature Woodlands of Point Pelee National Park Prior to Infestation by Emerald Ash Borer, Agrilus planipennis. Can. Field-Nat. 123: 99-106. Doss, R. P., J. E. Oliver, W. M. Proebsting, S. W. Potter, S. R. Kuy, S. L. Clement, R. T. Williamson, J. R. Carney, and E. D. DeVilbiss. 2000. Bruchins: insect-

216

derived plant regulators that stimulate neoplasm formation. Proc. Natl. Acad. Sci. U. S. A. 97: 6218-6218. Dowd, P. F. 1994. Enhanced maize (Zea mays L.) pericarp browning - associations with insect resistance and involvement of oxidizing enzymes. J. Chem. Ecol. 20: 2777- 2803. Dowd, P. F., C. Michael Smith, and T. C. Sparks. 1983. Detoxification of plant toxins by insects. Insect Biochemistry 13: 453-468. Driesche, J. V., and R. G. V. Driesche. 2000. Nature out of Place - Biological Invasions in the Global Age, Washington DC. Du, N. X., and P. M. Pijut. 2009. Agrobacterium-mediated transformation of Fraxinus pennsylvanica hypocotyls and plant regeneration. Plant Cell Reports 28: 915-923. Duffey, S. S., and M. J. Stout. 1996. Antinutritive and toxic components of plant defense against insects. Arch Insect Biochem 32: 3-37. Dunn, J. P., Daniel A. Potter, and Thomas W. Kimmerer. 1990. Carbohydrate reserves, radial growth, and mechanisms of resistance of oak trees to phloem- boring insects. Oecologia 83: 458-468. Dunn, J. P., T. W. Kimmerer, and D. A. Potter. 1987. Winter starch reserves of white oak as a predictor of attack by the twolined chestnut borer, Agrilus bilineatus (Weber) (Coleoptera: Buprestidae). Oecologia 74: 352-355. Dunn, J. P., Thomas W. Kimmerer, and Gerald L. Nordin. 1986. The role of host tree condition in attack of whthe oaks by the twolined chestnut borer, Agrilus bilineatus (Weber) (Coleoptera: Buprestidae). Oecologia 70: 596-600. Efron, B., E. Halloran, and S. Holmes. 1996a. Bootstrap confidence levels for phylogenetic trees (vol 93, pg 7085, 1996). Proc. Natl. Acad. Sci. U. S. A. 93: 13429-13434. Efron, B., E. Halloran, and S. Holmes. 1996b. Bootstrap confidence levels for phylogenetic trees. Proc. Natl. Acad. Sci. U. S. A. 93: 7085-7090. Ehrlich, P. R., and P. H. Raven. 1964. Butterflies and plants: a study in coevolution. EVOLUTION 18: 586-608. Engelberth, J., H. T. Alborn, E. A. Schmelz, and J. H. Tumlinson. 2004. Airborne signals prime plants against insect herbivore attack. Proc. Natl. Acad. Sci. U. S. A. 101: 1781-1785. Erbilgin, N., P. Krokene, E. Christiansen, G. Zeneli, and J. Gershenzon. 2006. Exogenous Application of Methyl Jasmonate Elicits Defenses in Norway Spruce (Picea abies) and Reduces Host Colonization by the Bark Beetle Ips typographus. Oecologia 148: 426-436. Eyles, A., P. Bonello, R. Ganley, and C. Mohammed. 2010. Induced resistance to pests and pathogens in trees. New Phytol. 185: 893-908. Eyles, A., W. Jones, K. Riedl, D. A. Herms, D. Cipollini, S. Schwartz, K. Chan, and P. Bonello. 2007. Comparative phloem chemistry of Manchurian (F. mandshurica) and two North American Ash Species (F. americana and F. pennsylvanica). J. Chem. Ecol. 33: 1430-1448. Farmer, E. E., and C. A. Ryan. 1990. Interplant Communication - Airborne Methyl Jasmonate Induces Synthesis of Proteinase-Inhibitors in Plant-Leaves. Proc. Natl. Acad. Sci. U. S. A. 87: 7713-7716.

217

Feeny, P. 1970. Seasonal Changes in Oak Leaf Tannins and Nutrients as a Cause of Spring Feeding by Winter Moth Caterpillars. Ecology 51: 565-&. Feeny, P. P. 1969. Inhibitory effect of oak leaf tannins on the hydrolysis of proteins by trypsin. Phytochemistry 8: 2119-2126. Felton, G. W. 2005. Indigestion is a plant's best defense. Proc. Natl. Acad. Sci. U. S. A. 102: 18771-18772. Felton, G. W., and H. Eichenseer. 1999. Herbivore saliva and its effects on plant defense against herbivores and pathogens. Induced plant defenses against pathogens and herbivores‚ÄìBiochemistry, ecology, and agriculture, eds. Agrawal, Tuzun and Bent: 19-36. Felton, G. W., K. K. Donato, R. M. Broadway, and S. S. Duffey. 1992. Impact of oxidized plant phenolics on the nutritional quality of dietary-protein to a noctuid herbivore, Spodoptera exigua. J. Insect Physiol. 38: 277-285. Fiehn, O. 2002. Metabolomics – the link between genotypes and phenotypes. Plant Molecular Biology 48: 155-171. Fitches, E., and J. A. Gatehouse. 1998. A comparison of the short and long term effects of insecticidal lectins on the activities of soluble and brush border enzymes of tomato moth larvae (Lacanobia oleracea). J. Insect Physiol. 44: 1213-1224. Flores, T., A. Alape-Giron, M. Flores-Diaz, and H. E. Flores. 2002. Ocatin. A Novel Tuber Storage Protein from the Andean Tuber Crop Oca with Antibacterial and Antifungal Activities. Plant Physiology 128: 1291-1302. Foyer, C. H., and G. Noctor. 2005. Oxidant and antioxidant signalling in plants: a re- evaluation of the concept of oxidative stress in a physiological context. Plant, Cell & Environment 28: 1056-1071. Fraenkel, G. S. 1959. The raison d'etre of secondary plant substances. Science 129: 1466-1466. FRAXIGEN. 2005. Ash species in Europe: biological characteristics and practical guidelines for sutainable use, Oxford Forestry Institute, University of Oxford, UK. Gandhi, K., and D. Herms. 2010a. North American arthropods at risk due to widespread Fraxinus mortality caused by the Alien Emerald ash borer. Biological Invasions 12: 1839-1846. Gandhi, K. J. K., and D. A. Herms. 2010b. Direct and indirect effects of alien insect herbivores on ecological processes and interactions in forests of eastern North America. Biological Invasions 12: 389-405. Gang, D. R., M. A. Costa, M. Fujita, A. T. Dinkova-Kostova, H. B. Wang, V. Burlat, W. Martin, S. Sarkanen, L. B. Davin, and N. G. Lewis. 1999. Regiochemical control of monolignol radical coupling: a now paradigm for lignin and lignan biosynthesis. Chem. Biol. 6: 143-151. Garcia, E. S., M. M. O. Cabral, G. A. Schaub, O. R. Gottlieb, and P. Azambuja. 2000. Effects of lignoids on a hematophagous bug, Rhodnius prolixus: feeding, ecdysis and diuresis. Phytochemistry 55: 611-616. Gatehouse, J. A. 2002. Plant resistance towards insect herbivores: a dynamic interaction. New Phytol. 156: 145-169.

218

Ge, X., C. Dietrich, M. Matsuno, G. Li, H. Berg, and Y. Xia. 2005. An Arabidopsis aspartic protease functions as an anti-cell-death component in reproduction and embryogenesis. EMBO reports 6: 282-288. Giese, P. 1996. Basketry: Philosophy, Plants, Environment. Gion, J.-M., C. Lalanne, G. Le Provost, H. Ferry-Dumazet, J. Paiva, P. Chaumeil, J.-M. Frigerio, J. Brach, A. Barré, A. de Daruvar, S. Claverol, M. Bonneu, N. Sommerer, L. Negroni, and C. Plomion. 2005. The proteome of maritime pine wood forming tissue. Proteomics 5: 3731-3751. Glazebrook, J. 2005. Contrasting mechanisms of defense against biotrophic and necrotrophic pathogens. Annual Review of Phytopathology 43: 205-227. Gould, J., J. Tanner, D. Winograd, and S. Lane. Year. Published. Initial studies on the laboratory rearing of emerald ash borer and foreign exploration for natural enemies, pp. 73–74. In V. C. Mastro and R. Reardon (eds.), Proceedings of Emerald Ash Borer Research and Technology Development Meeting, 2005, Morgantown, W.Va. USDA Forest Service. Green, T. R. a. C. A. R. 1972. Wound-induced proteinase inhibitor in plant leaves: a possible defense mechanism against insects. Science 175: 776-777. Guevara, M. G., G. R. Daleo, and C. R. Oliva. 2001. Purification and characterization of an aspartic protease from potato leaves. Physiol. Plant. 112: 321-326. Guevara, M. G., C. R. Oliva, M. Machinandiarena, and G. R. Daleo. 1999. Purification and properties of an aspartic protease from potato tuber that is inhibited by a basic chitinase. Physiol. Plant. 106: 164-169. Guevara, M. G., C. R. Oliva, M. Huarte, and G. R. Daleo. 2002. An aspartic protease with antimicrobial activity is induced after infection and wounding in intercellular fluids of potato tubers. European Journal of Plant Pathology 108: 131-137. Guevara, M. G., P. Verissimo, E. Pires, C. Faro, and G. R. Daleo. 2004. Potato aspartic proteases: Induction, antimicrobial activity and substrate specificity. Journal of Plant Pathology 86: 233-238. Guinier, E. 1886. Effets de l'annelation de l'ecorce. Rev. Eaux For 25: 287-293. Gundlach, H., M. J. Muller, T. M. Kutchan, and M. H. Zenk. 1992. Jasmonic Acid Is a Signal Transducer in Elicitor-Induced Plant-Cell Cultures. Proc. Natl. Acad. Sci. U. S. A. 89: 2389-2393. Guo, H., A. H. Liu, M. Ye, M. Yang, and D. A. Guo. 2007. Characterization of phenolic compounds in the fruits of Forsythia suspensa by high-performance liquid chromatography coupled with electrospray ionization tandem mass spectrometry. Rapid Commun Mass Sp 21: 715-729. Haack, R. A., and Daniel M. Benjamin. 1982. The Biology and Ecology of the Twolined Chestnut Borer, Agrilus bilineatus (Coleoptera: Buprestidae), on oaks, Quercus spp., in Wisconsin. The Canadian Entomologist 114: 385-396. Haack, R. A., Eduard Jendek, Houping Liu, Kenneth R. Marchant, Toby R. Petrice, Therese M. Poland, and Hui Ye. 2002. The Emerald Ash Borer: A New Exotic Pest in North America. Newsletter of the Michigan Entomological Society 47: 1- 5. Hallenbeck, W. H., and K. M. Cunningham-Burns. 1985. Pesticides and human health, Springer Verlag.

219

Hanks, L. M., T. D. Paine, and J. G. Millar. 1991. Mechanisms of Resistance in Eucalyptus Against Larvae of the Eucalyptus Longhorned Borer (Coleoptera: Cerambycidae). Environmental Entomology 20: 1583-1588. Hanover, J. W. 1975. Physiology of Tree Resistance to Insects. Annual Review of Entomology 20: 75-95. Hansbrough, T., and R. G. Merrifield. 1963. The influence of partial girdling on cone and seed production of loblolly pine, Agricultural Experiment Station, Louisiana State University & A & M College. Harborne, J. B. 1993. Introduction to ecological biochemistry, Academic Pr. Harris, M. K. 1975. Allopatric resistance – searching for sources of Insect Resistance for use in Agriculture. Environmental Entomology 4: 661-669. Hartley, S. E., and J. H. Lawton. 1991. Biochemical aspects and significance of the rapidly induced accumulation of phenolics in birch foliage. Phytochemical induction by herbivores. Wiley, New York: 105-132. Herms, D. A. 2002. Effects of Fertilization on Insect Resistance of Woody Ornamental Plants: Reassessing an Entrenched Paradigm. Environmental Entomology 31: 923-933. Herms, D. A., and W. J. Mattson. 1992. The Dilemma of Plants: To Grow or Defend. The Quarterly Review of Biology 67: 283-283. Herms, D. A., A. K. Stone, and J. A. Chatfield. 2004. Emerald Ash Borer: The Beginning of the End of Ash in North America?, pp. 62–71. In J. A. Chatfield, E. A. Draper, H. M. Mathers, D. E. Dyke, P. J. Bennett and J. F. Boggs [eds.], Ornamental plants: annual reports and research reviews 2003 - 2004. Ohio Agricultural Research and Development Center, Ohio State University, Wooster, OH. Herms, D. A., A. Smith, K. Gandhi. Year. Published. Patterns of EAB-induced ash decline and mortality in the forests of Southeast Michigan. In, Emerald Ash Borer and Asian Longhorned Beetle Research and Development Review, 2006, Cincinnati, OH. Herms, D. A., D. G. McCullough, D. R. Smitley. 2004. Under Attack, pp. 20-26, American Nurseryman. Hillis, W. E. 1987. Heartwood and Tree Exudates, Springer Verlag, Berlin, Germany. Hines, J. W., and H. J. Heikkenen. 1977. Beetles (Coleoptera-Scolytidae, Cerambycidae, Curculionidae, and Buprestidae) attracted to severed Virginia Pine (Pinus virginiae Mill.). Environmental Entomology 6: 123-127. Howe, G. A., and G. Jander. 2008. Plant immunity to insect herbivores. Annual Review of Plant Biology 59: 41-66. Hu, L., M. Ye, X. Jiang, S. Feng, and H. Zou. 2007. Advances in hyphenated analytical techniques for shotgun proteome and peptidome analysis--A review. Analytica Chimica Acta 598: 193-204. Iossifova, T., B. Vogler, and I. Kostova. 2002. Escuside, a new coumarin-secoiridoid from Fraxinus ornus bark. Fitoterapia 73: 386-389. Iverson, L., A. Prasad, J. Bossenbroek, M. Schwartz, D. Syndor. Year. 2006. Modeling EAB spread in Ohio. In, Emerald Ash Borer and Asian Longhorned Beetle Research and Development Reveiw, 2006, Cincinnati, OH.

220

James, P. 1997. Protein identification in the post-genome era: the rapid rise of proteomics. Quarterly Reviews of Biophysics 30: 279-331. Janzen, D. H. 1966. Coevolution of mutualism between ants and acacias in Central America. EVOLUTION 20: 249-&-249-&. Jeandroz, S., A. Roy, and J. Bousquet. 1997. Phylogeny and phylogeography of the circumpolar genus Fraxinus (Oleaceae) based on internal transcribed spacer sequences of nuclear ribosomal DNA. Mol. Phylogenet. Evol. 7: 241-251. Jensen, S. R., H. Franzyk, and E. Wallander. 2002. Chemotaxonomy of the Oleaceae: iridoids as taxonomic markers. Phytochemistry 60: 213-231. Jiang, B. H., U. Siregar, K. O. Willeford, D. S. Luthe, and W. P. Williams. 1995. Association of a 33-kilodalton cysteine proteinase found in corn callus with the inhibition of fall armyworm larval growth. Plant Physiology 108: 1631-1640. Johnson, L. A. S. 1957. A review of the family Oleaceae. Contributions from the New South Wales National Herbarium 2: 395-418. Johnson, R., and D. A. Wichern. 1998. Applied Multivariate Statistical Analysis. Kanchiswamy, C. N., H. Takahashi, S. Quadro, M. E. Maffei, S. Bossi, C. Bertea, S. A. Zebelo, A. Muroi, N. Ishihama, H. Yoshioka, W. Boland, J. Takabayashi, Y. Endo, T. Sawasaki, and G. Arimura. 2010. Regulation of Arabidopsis defense responses against Spodoptera littoralis by CPK-mediated calcium signaling. BMC Plant Biol. 10. Kang, J. H., and I. T. Baldwin. 2006. Isolation and characterization of the threonine deaminase promoter in Nicotiana attenuata. Plant Sci. 171: 435-440. Karamloo, F., A. Wangorsch, H. Kasahara, L. B. Davin, D. Haustein, N. G. Lewis, and S. Vieths. 2001. Phenylcoumaran benzylic ether and isoflavonoid reductases are a new class of cross-reactive allergens in birch pollen, fruits and vegetables. European Journal of Biochemistry 268: 5310-5320. Karban, R., A. A. Agrawal, and M. Mangel. 1997. The benefits of induced defenses against herbivores. Ecology 78: 1351-1355. Keena, M. A., J. Gould, and L. S. Bauer. Year. Published. Factors that influence emerald ash borer (Agrilus planipennis) adult longevity and oviposition under laboratory conditions. In K. A. McManus and Gottschalk (eds.), Proceedings 20th U.S. Department of Agriculture interagency research forum on invasive species 2009, 2009. USDA Forest Service. Kessler, A., and I. T. Baldwin. 2002. Plant responses to insect herbivory: The emerging molecular analysis. Annual Review of Plant Biology 53: 299-328. Kieffer, P., J. Dommes, L. Hoffmann, J. F. Hausman, and J. Renaut. 2008. Quantitative changes in protein expression of cadmium‐exposed poplar plants. Proteomics 8: 2514-2530. Kliebenstein, D. J., and H. C. Rowe. 2008. Ecological costs of biotrophic versus necrotrophic pathogen resistance, the hypersensitive response and signal transduction. Plant Sci. 174: 551-556. Knight, K. S., and Long. 2008. How fast will trees die? A transition matrix model of ash decline in forest stands infested by emerald ash borer. Knoblauch, E. 1895. Oleaceae, pp. 1-16. In A. Engler (ed.), Die Naturlichen Pflanzenfamilien, vol. 2.

221

Koch, J. L., M. E. Mason, D. W. Carey, K. S. Knight, P. T. M., and D. A. Herms. Year. Published. Survey for Tolerance to Emerald ash borer within North American Ash Species. In C. H. Michler and M. D. Ginzel (eds.), Proceedings of Symposium on Ash in North America, 2010, West Lafayette, IN Purdue University. U.S. Forest Service, Northern Research Station Newtown Square, PA. Koistinen, K., P. Soininen, T. Venalainen, J. Hayrinen, R. Laatikainen, M. Perakyla, A. Tervahauta, and S. Karenlampi. 2005. Birch PR-10c interacts with several biologically important ligands. Phytochemistry 66: 2524-2533. Koistinen, K. M., V. H. Hassinen, P. A. M. Gynther, S. J. Lehesranta, S. I. Keinanen, H. I. Kokko, E. J. Oksanen, A. I. Tervahauta, S. Auriola, and S. O. Karenlampi. 2002. Birch PR-10c is induced by factors causing oxidative stress but appears not to confer tolerance to these agents. New Phytol. 155: 381- 391. Konno, K., C. Hirayama, H. Yasui, and M. Nakamura. 1999. Enzymatic activation of oleuropein: A protein crosslinker used as a chemical defense in the privet tree. Proc. Natl. Acad. Sci. U. S. A. 96: 9159-9164. Koo, A. J. K., and G. A. Howe. 2009. The wound hormone jasmonate. Phytochemistry 70: 1571-1580. Koricheva, J., S. Larsson, and E. Haukioja. 1998. Insect performance on experimentally stressed woody plants: A meta-analysis. Annual Review of Entomology 43: 195-216. Korner, E., C. C. von Dahl, G. Bonaventure, and I. T. Baldwin. 2009. Pectin methylesterase NaPME1 contributes to the emission of methanol during insect herbivory and to the elicitation of defence responses in Nicotiana attenuata. Journal of Experimental Botany 60: 2631-2640. Kostova, I. 2001. Fraxinus ornus L. Fitoterapia 72: 471-480. Kostova, I., and T. Iossifova. 2007. Chemical components of Fraxinus species. Fitoterapia 78: 85-106. Koul, O. 2008. Phytochemicals and insect control: An antifeedant approach. Crit Rev Plant Sci 27: 1-24. Kovacs, K. F., R. G. Haight, D. G. McCullough, R. J. Mercader, N. W. Siegert, and A. M. Liebhold. 2010. Cost of potential emerald ash borer damage in US communities, 2009-2019. Ecol. Econ. 69: 569-578. Lawrence, S. D., and N. G. Novak. 2006. Expression of poplar chitinase in tomato leads to inhibition of development in colorado potato beetle. Biotechnol Lett 28: 593- 599. Levin, D. A. 1976. Chemical defense of plants to pathogens and herbivores. Annu. Rev. Ecol. Syst. 7: 121-159. Lingelsheim, A. 1920. Oleaceae–Oleoideae–Fraxineae, pp. 1-61. In A. Engler (ed.), Die Naturlichen Pflanzenfamilien, vol. 243. Liska, A. J., and A. Shevchenko. 2003. Expanding the organismal scope of proteomics: Cross-species protein identification by mass spectrometry and its implications. Proteomics 3: 19-28. Liu, H. P., L. S. Bauer, D. L. Miller, T. H. Zhao, R. T. Gao, L. W. Song, Q. S. Luan, R. Z. Jin, and C. Q. Gao. 2007. Seasonal abundance of Agrilus planipennis (Coleoptera : Buprestidae) and its natural enemies Oobius agrili (Hymenoptera : 222

Encyrtidae) and Tetrastichus planipennisi (Hymenoptera : Eulophidae) in China. Biological Control 42: 61-71. Liu, J.-J., and A. K. M. Ekramoddoullah. 2006. The family 10 of plant pathogenesis- related proteins: Their structure, regulation, and function in response to biotic and abiotic stresses. Physiological and Molecular Plant Pathology 68: 3-13. Loerch, C. R., and E. A. Cameron. 1983. Determination of Larval Instars of the Bronze Birch Borer, Agrilus anxius (Coleoptera: Buprestidae). Ann. Entomol. Soc. Am. 76: 948-952. Maffei, M. E., A. Mithofer, and W. Boland. 2007a. Before gene expression: early events in plant-insect interaction. Trends in Plant Science 12: 310-316. Maffei, M. E., A. Mithofer, and W. Boland. 2007b. Insects feeding on plants: Rapid signals and responses preceding the induction of phytochemical release. Phytochemistry 68: 2946-2959. Maffei, M. E., A. Mithofer, G. I. Arimura, H. Uchtenhagen, S. Bossi, C. M. Bertea, L. S. Cucuzza, M. Novero, V. Volpe, S. Quadro, and W. Boland. 2006. Effects of Feeding Spodoptera littoralis on Lima Bean Leaves. III. Membrane Depolarization and Involvement of Hydrogen Peroxide. Plant Physiology 140: 1022-1035. Margni, M., D. Rossier, P. Crettaz, and O. Jolliet. 2002. Life cycle impact assessment of pesticides on human health and ecosystems. Agriculture, ecosystems & environment 93: 379-392. Markwalder, H. U., and H. Neukom. 1976. Diferulic acid as a possible crosslink in hemicelluloses from wheat germ. Phytochemistry 15: 836-837. Mayer, A. M. 2004. Resistance to herbivores and fungal pathogens: Variations on a common theme? A review comparing the effect of secondary metabolites, induced and constitutive, on herbivores and fungal pathogens. Isr. J. Plant Sci. 52: 279-292. McCullough, D., S. Katovich. 2004. Pest Alert: Emerald Ash Borer. U.S. Dept. of Agriculture, Forest Service, Northeastern Area, State & Private Forestry. McCullough, D. G., D. L. Cappaert, and T. M. Poland. 2005. Evaluation of insecticides for control of emerald ash borer: Summary of 2004 trials. McCullough, D. G., T. M. Poland, and D. Cappaert. 2009a. Attraction of the emerald ash borer to ash trees stressed by girdling, herbicide treatment, or wounding. Can. J. For. Res.-Rev. Can. Rech. For. 39: 1331-1345. McCullough, D. G., T. M. Poland, A. C. Anulewicz, and D. Cappaert. 2009b. Emerald Ash Borer (Coleoptera: Buprestidae) Attraction to Stressed or Baited Ash Trees. Environmental Entomology 38: 1668-1679. Merkle, S. A., G. M. Andrade, C. J. Nairn, W. A. Powell, and C. A. Maynard. 2007. Restoration of threatened species: a noble cause for transgenic trees. Tree Genet. Genomes 3: 111-118. Micheli, F. 2001. Pectin methylesterases: cell wall enzymes with important roles in plant physiology. Trends in Plant Science 6: 414-419. Miller, B., L. L. Madilao, S. Ralph, and J. Bohlmann. 2005. Insect-induced conifer defense. White pine weevil and methyl jasmonate induce traumatic resinosis, de novo formed volatile emissions, and accumulation of terpenoid synthase and

223

putative octadecanoid pathway transcripts in Sitka spruce. Plant Physiology 137: 369-382. Miller, R. O., P. D. Bloese, J. W. Hanover, R. A. Haack. 1991. Paper birch and European white birch vary in growth and resistance the bronze birch borer. Journal of the American Society of Horticulture Science 116: 580-584. Mittapalli, O., X. D. Bai, P. Mamidala, S. P. Rajarapu, P. Bonello, and D. A. Herms. 2010. Tissue-Specific Transcriptomics of the Exotic Invasive Insect Pest Emerald Ash Borer (Agrilus planipennis). PLoS One 5: -. Miyazawa, M., Y. Ishikawa, H. Kasahara, J. Yamanaka, and H. Kameoka. 1994. An insect growth-inhibitory lignan from flower buds of Magnolia-fargesii. Phytochemistry 35: 611-613. Mohan, S., P. W. K. Ma, T. Pechan, E. R. Bassford, W. P. Williams, and D. S. Luthe. 2006. Degradation of the S. frugiperda peritrophic matrix by an inducible maize cysteine protease. J. Insect Physiol. 52: 21-28. Molgaard, P. 1986. Population-genetics and geographical-distribution of caffeic acid- esters in leaves of Plantago major in Denmark. J Ecol 74: 1127-1137. Moore, J. P., N. D. Paul, J. B. Whittaker, and J. E. Taylor. 2003. Exogenous jasmonic acid mimics herbivore-induced systemic increase in cell wall bound peroxidase activity and reduction in leaf expansion. Functional Ecology 17: 549-554. Muilenburg, V. L. 2010. Biochemistry, physiology, and ecology of paper birch defenses to bronze birch borer and their responses to anthropogenic greenhouse gases. Ph.D., Ohio Agricultural Research and Development Center The Ohio State University Wooster, OH. Mullick, D. B., G. D. Jensen. 1973. New concepts and terminology of coniferous periderms: necrophylactic and exophylactic periderms. Canadian Journal of Botany 51: 1459-1470. Mutlu, A., and S. Gal. 1999. Plant aspartic proteinases: enzymes on the way to a function. Physiol. Plant. 105: 569-576. Nesom, G. 2000. White ash: Fraxinus americana L. In N. USDA [ed.]. USDA. Niemela, P., and W. J. Mattson. 1996. Invasion of North American forests by European phytophagous insects - Legacy of the European crucible? Bioscience 46: 741-753. Nilo, R., C. Saffie, K. Lilley, R. Baeza-Yates, V. Cambiazo, R. Campos-Vargas, M. González, L. A. Meisel, J. Retamales, H. Silva, and A. Orellana. 2009. Proteomic analysis of peach fruit mesocarp softening and chilling injury using difference gel electrophoresis (DIGE). BMC Genomics 11: 43-43. Noel, A. R. A. 1970. GIRDLED TREE. Bot. Rev. 36: 162-&. ODA. 2005. Impact of EAB. http://www.ohioagriculture.gov/plant/ppc/eab/plnt-eab- impact.stm ODNR. 2005. Potential economic effect of emerald ash borer activity on Ohio's Economy and to homeowners could reach $3 billion http://www.dnr.state.oh.us/forestry/health/pdf/Potential Economic Impact of EAB.pdf Oliver, J. E., R. P. Doss, B. Marquez, and E. D. Devilbiss. 2002. Bruchins, plant mitogens from weevils: structural requirements for activity. J. Chem. Ecol. 28: 2503-2513.

224

Oliver, S. G., M. K. Winson, D. B. Kell, and F. Baganz. 1998. Systematic functional analysis of the yeast genome. Trends Biotechnol. 16: 373-378. Orozco-Cardenas, M., and C. A. Ryan. 1999. Hydrogen peroxide is generated systemically in plant leaves by wounding and systemin via the octadecanoid pathway. Proc. Natl. Acad. Sci. U. S. A. 96: 6553-6557. Ossipov, V., S. Ossipova, V. Bykov, E. Oksanen, J. Koricheva, and E. Haukioja. 2008. Application of metabolomics to genotype and phenotype discrimination of birch trees grown in a long-term open-field experiment. Metabolomics 4: 39-51. Painter, R. H. 1951. Insect Resistance in Crop Plants. Soil Science 72: 481. Park, C.-J., K.-J. Kim, R. Shin, J. M. Park, Y.-C. Shin, and K.-H. Paek. 2004. Pathogenesis-related protein 10 isolated from hot pepper functions as a ribonuclease in an antiviral pathway. The Plant Journal: For Cell and Molecular Biology 37: 186-198. Pauwels, L., D. Inze, and A. Goossens. 2009. Jasmonate-inducible gene: what does it mean? Trends in Plant Science 14: 87-91. Pechan, T., A. Cohen, W. P. Williams, and D. S. Luthe. 2002. Insect feeding mobilizes a unique plant defense protease that disrupts the peritrophic matrix of caterpillars. Proc. Natl. Acad. Sci. U. S. A. 99: 13319-13323. Perez-Castorena, A. L., S. Escalona, and A. R. De Vivar. 1997. Constituents of Fruits, Leaves and Bark of Fresno (Fraxinus uhdei). REVISTA LATINOAMERICANA DE QUIMICA 25: 86-90. Petrice, T., R. Haack. 2006. Effects of Cutting Date, outdoor storage conditions, and splitting on survival of Agrilus planipennis (Coleoptera: Buprestidae) in Firewood Logs. Journal of Economic Entomology 99: 790-796. Philippe, R. N., S. G. Ralph, S. D. Mansfield, and J. Bohlmann. 2010. Transcriptome profiles of hybrid poplar (Populus trichocarpa x deltoides) reveal rapid changes in undamaged, systemic sink leaves after simulated feeding by forest tent caterpillar (Malacosoma disstria). New Phytol. 188: 787-802. Pichersky, E., and D. R. Gang. 2000. Genetics and biochemistry of secondary metabolites in plants: an evolutionary perspective. Trends in Plant Science 5: 439- 445. Poland, T. M., and D. G. McCullough. 2006. Emerald Ash Borer: Invasion of the Urban Forest and the Threat to North Americas Ash Resource. Journal of Forestry 104: 118-124. Prasad, A. M., L. R. Iverson, M. P. Peters, J. M. Bossenbroek, S. N. Matthews, T. Davis Sydnor, and M. W. Schwartz. 2010. Modeling the invasive emerald ash borer risk of spread using a spatially explicit cellular model. Landscape ecology 25: 353-369. Puehringer, H. 2003. MdAP, a novel protein in apple, is associated with the major allergen Mal d 1. Gene 321: 173-183. Puhringer, H., D. Moll, K. Hoffmann-Sommergruber, B. Watillon, H. Katinger, and M. L. D. Machado. 2000. The promoter of an apple Ypr10 gene, encoding the major allergen Mal d 1, is stress- and pathogen-inducible. Plant Sci. 152: 35-50. Pureswaran, D. S., and T. M. Poland. 2009. Host Selection and Feeding Preference of (Agrilus planipennis) (Coleoptera: Buprestidae) on Ash (Fraxinus spp.). Environmental Entomology 38: 757-765. 225

Rakwal, R., G. K. Agrawal, and M. Yonekura. 2001. Light-dependent induction of OsPR10 in rice (Oryza sativa L.) seedlings by the global stress signaling molecule jasmonic acid and protein phosphatase 2A inhibitors. Plant Sci. 161: 469-479. Rasmann, S., and A. A. Agrawal. 2009. Plant defense against herbivory: progress in identifying synergism, redundancy, and antagonism between resistance traits. Current Opinion in Plant Biology 12: 473-478. Rawlings, N. D., and A. J. Barrett. 1995. Families of aspartic peptidases, and those of unknown catalytic mechanism, pp. 105-120, Proteolytic Enzymes: Aspartic and Metallo Peptidases, vol. 248. Academic Press Inc, San Diego. Rebek, E. J., D. A. Herms, and D. R. Smitley. 2008. Interspecific Variation in Resistance to Emerald Ash Borer (Coleoptera: Buprestidae) among North American and Asian Ash (Fraxinus spp.). Environmental Entomology 37: 242- 246. Reddish, J. M., N. St-Pierre, A. Nichols, K. Green-Church, and M. Wick. 2008. Proteomic analysis of proteins associated with body mass and length in yellow perch,Perca flavescens. Proteomics 8: 2333-2343. Reiners, W. A., and N. M. Reiners. 1970. Energy and Nutrient Dynamics of Forest Floors in 3 Minnesota Forests. J Ecol 58: 497-&. Ribeiro, J. M. C., R. F. Chapman, and G. Boer. 1995. Insect saliva: function, biochemistry, and physiology. Regulatory mechanisms in insect feeding.: 74-97. Richards, S., R. A. Gibbs, G. M. Weinstock, S. J. Brown, R. Denell, R. W. Beeman, R. Gibbs, G. Bucher, M. Friedrich, C. J. P. Grimmelikhuijzen, M. Klingler, M. D. Lorenzen, S. Roth, R. Schroder, D. Tautz, E. M. Zdobnov, D. Muzny, T. Attaway, S. Bell, C. J. Buhay, M. N. Chandrabose, D. Chavez, K. P. Clerk- Blankenburg, A. Cree, M. Dao, C. Davis, J. Chacko, H. Dinh, S. Dugan- Rocha, G. Fowler, T. T. Garner, J. Garnes, A. Gnirke, A. Hawes, J. Hernandez, S. Hines, M. Holder, J. Hume, S. N. Jhangiani, V. Joshi, Z. M. Khan, L. Jackson, C. Kovar, A. Kowis, S. Lee, L. R. Lewis, J. Margolis, M. Morgan, L. V. Nazareth, N. Nguyen, G. Okwuonu, D. Parker, S. J. Ruiz, J. Santibanez, J. Savard, S. E. Scherer, B. Schneider, E. Sodergren, S. Vattahil, D. Villasana, C. S. White, R. Wright, Y. Park, J. Lord, B. Oppert, S. Brown, L. J. Wang, J. Savard, Y. Liu, K. Worley, C. G. Elsik, J. T. Reese, E. Elhaik, G. Landan, D. Graur, P. Arensburger, P. Atkinson, J. Beidler, J. P. Demuth, D. W. Drury, Y. Z. Du, H. Fujiwara, V. Maselli, M. Osanai, H. M. Robertson, Z. Tu, J. J. Wang, S. Z. Wang, H. Song, L. Zhang, E. Sodergren, D. Werner, M. Stanke, B. Morgenstern, V. Solovyev, P. Kosarev, G. Brown, H. C. Chen, O. Ermolaeva, W. Hlavina, et al. 2008. The genome of the model beetle and pest Tribolium castaneum. Nature 452: 949-955. Robert, N., J. Ferran, C. Breda, P. Coutos-Thevenot, M. Boulay, D. Buffard, and R. Esnault. 2001. Molecular characterization of the incompatible interaction of vitis vinifera leaves with Pseudomonas syringae pv. pisi: Expression of genes coding for stilbene synthase and class 10 PR protein. European Journal of Plant Pathology 107: 249-261. Rodriguez-Saona, C., T. M. Poland, J. R. Miller, L. L. Stelinski, G. G. Grant, P. de Groot, L. Buchan, and L. MacDonald. 2006. Behavioral and electrophysiological responses of the emerald ash borer, Agrilus planipennis, to 226

induced volatiles of Manchurian ash, Fraxinus mandshurica. Chemoecology 16: 75-86. Ros Barcelo, A. 1997. Lignification in plant cell walls. Int Rev Cytol 176. Roush, R. T., and B. E. Tabashnik. 1990. Pesticide resistance in arthropods, Chapman and Hall Ltd, London. Ruuhola, T., J. P. Salminen, S. Haviola, S. Y. Yang, and M. J. Rantala. 2007. Immunological memory of mountain birches: Effects of phenolics on performance of the autumnal moth depend on herbivory history of trees. J. Chem. Ecol. 33: 1160-1176. Ryan, C. A. 1990. Protease inhibitors in plants – genes for improving defense against insects and pathogens. Annual Review of Phytopathology 28: 425-449. Sarkkinen, P., N. Kalkkinen, C. Tilgmann, J. Siuro, J. Kervinen, and L. Mikola. 1992. Aspartic proteinase from Barley grains is related to mammalian lysosomal cathepsin-D. Planta 186: 317-323. Savarese, M., E. De Marco, and R. Sacchi. 2007. Characterization of phenolic extracts from olives (Olea europaea cv. Pisciottana) by electrospray ionization mass spectrometry. Food Chem 105: 761-770. Schmidt, D. D., C. Voelckel, M. Hartl, S. Schmidt, and I. T. Baldwin. 2005. Specificity in ecological interactions. Attack from the same lepidopteran herbivore results in species-specific transcriptional responses in two solanaceous host plants. Plant Physiology 138: 1763-1763. Shivaji, R., A. Camas, A. Ankala, J. Engelberth, J. H. Tumlinson, W. P. Williams, J. R. Wilkinson, and D. S. Luthe. 2010. Plants on Constant Alert: Elevated Levels of Jasmonic Acid and Jasmonate-Induced Transcripts in Caterpillar-Resistant Maize. J. Chem. Ecol. 36: 179-191. Siegart, N. W., D. G. McCullough, A. Tluczek. Year. Published. Two years under the bark: understanding multiple-year development of emerald ash borer. In, Emerald Ash Borer and Asian Longhorned Beetle Research and Development Reveiw, 2006, Cincinnati, OH. Simoes, I., and C. Faro. 2004. Structure and function of plant aspartic proteinases. European Journal of Biochemistry 271: 2067-2075. Sinclair, W. A., and H. H. Lyon. 2005. Diseases of trees and shrubs, Cornell University Press. Smith, A., D. A. Herms, and R. P. Long. Year. Published. The impact of emerald ash borer on forests within the Huron River watershed. In V. Mastro and R. Reardon (eds.), Emerald Ash Borer Research and Technology Development Meeting, 2005, Pittsburgh, Pennsylvania. USDA Forest Service publication FHTET-2004- 15. Smith, E. L., A. J. Storer, and B. K. Roosien. Year. Published. Emerald ash borer infestation rates in Michigan, Ohio, and Indiana. In K. A. McManus and K. W. Gottschalk (eds.), 20th U.S. Department of Agriculture interagency research forum on invasive species 2009, 2009 January 13-16 2009, Annapolis, MD. U.S. Department of Agriculture, Forest Service, Northern Research Station. Smitley, D., D.A. Herms, and E.J. Rebek. Year. Published. Timing of imidacloprid soil drenches for emerald ash borer control, pp. 26–30. In V. Mastro, R. Reardon and G. Parra (eds.), Proceedings of Emerald Ash Borer Research and Technology 227

Development Meeting 2006, Pittsburgh, PA. USDA Forest Service Forest Health Technology Enterprise Team. Spiteller, D., K. Dettner, and W. Boland. 2000. Gut Bacteria May Be Involved in Interactions between Plants, Herbivores and Their Predators: Microbial Biosynthesis of N-Acylglutamine Surfactants as Elicitors of Plant Volatiles. Biological Chemistry 381: 755-762. Spoel, S. H., J. S. Johnson, and X. Dong. 2007. Regulation of tradeoffs between plant defenses against pathogens with different lifestyles. Proc. Natl. Acad. Sci. U. S. A. 104: 18842-18847. Strack, D. 1997. Phenolic Metabolism, pp. 387-416. In D. PM and H. JB (eds.), Plant Biochemistry. Academic Press, New York, NY. Suzuki, R., and H. Shimodaira. 2006. Pvclust: an R package for assessing the uncertainty in hierarchical clustering. Bioinformatics 22: 1540-1542. Swain, T. 1979. Tannins and lignins, pp. 657-682. In G. Rosenthal and D. H. Janzen (eds.), Herbivores: Their interaction with secondary plant metabolites. Academic Press, New York. Swihart, R. K., J. P. Bryant, L. Newton. 1994. Lagitudinal Patterns in consumption of woody plants by snowshoe hares in the eastern United States. Oikos 70: 427-434. Sydnor, T. D., M. Bumgardner, and A. Todd. 2007. The potential economic impacts of emerald ash borer (Agrilus planipennis) on Ohio, U.S., communities. Arboriculture & Urban Forestry. 33: 48-54. Takenaka, Y., T. Tanahashi, M. Shintaku, T. Sakai, N. Nagakura, and Parida. 2000. Secoiridoid glucosides from Fraxinus americana. Phytochemistry 55: 275-284. Tamogami, S., R. Ralkwal, and G. K. Agrawal. 2008. Interplant communication: Airborne methyl jasmonate is essentially converted into JA and JA-Ile activating jasmonate signaling pathway and VOCs emission. Biochem. Biophys. Res. Commun. 376: 723-727. Taylor, H. 1945. Cyto-taxonomy and phylogeny of the Oleaceae. Brittonia 5: 337-367. Thines, B., L. Katsir, M. Melotto, Y. Niu, A. Mandaokar, G. H. Liu, K. Nomura, S. Y. He, G. A. Howe, and J. Browse. 2007. JAZ repressor proteins are targets of the SCFCO11 complex during jasmonate signalling. Nature 448: 661-U662. Timms, J. F., and R. Cramer. 2008. Difference gel electrophoresis. Proteomics 8: 4886- 4897. Tkalec, M., K. Malaric, and B. Pevalek-Kozlina. 2007. Exposure to radiofrequency radiation induces oxidative stress in duckweed Lemna minor L. Sci. Total Environ. 388: 78-89. Ünlü, M., M. E. Morgan, and J. S. Minden. 1997. Difference gel electrophoresis. A single gel method for detecting changes in protein extracts. Electrophoresis 18: 2071-2077. Urs, N. V. R., and J. M. Dunleavy. 1975. Enhancement of bactericidal activity of a peroxidase system by phenolic compounds. Phytopathology 65: 686-690. USDA, N. 2011. Emerald Ash Borer Quarantine. USDA. Vâlcu, C.-M., and K. Schlink. 2006. Efficient extraction of proteins from woody plant samples for two-dimensional electrophoresis. Proteomics 6: 4166-4175.

228

van Loon, L. C., M. Rep, and C. M. J. Pieterse. 2006. Significance of Inducible Defense-related Proteins in Infected Plants. Annual Review of Phytopathology 44: 135-162. Vance, C. P., T. K. Kirk, and R. T. Sherwood. 1980. Lignification as a Mechanism of Disease Resistance. Annual Review of Phytopathology 18: 259-288. Vasanthakumar, A., J. Handelsman, P. D. Schloss, L. S. Bauer, and K. F. Raffa. 2008. Gut Microbiota of an Invasive Subcortical Beetle, Agrilus planipennis Fairmarine, Across Various Life Stages. Environmental Entomology 37: 1344- 1353. Wainhouse, D., D. J. Cross, and R. S. Howell. 1990. The Role of Lignin as a Defense against the Spruce Bark Beetle Dendroctonus micans - Effect on Larvae and Adults. Oecologia 85: 257-265. Wallander, E. 2008. Systematic of Fraxinus (Oleaceae) and evolution of dioecy. Plant Systematics and Evolution 273: 25-49. Wallander, E., and V. A. Albert. 2000. Phylogeny and classification of Oleaceae based on rps16 and trnL-F sequence data. Am. J. Bot. 87: 1827-1841. Walling, L. L. 2000. The myriad plant responses to herbivores. Journal of Plant Growth Regulation 19: 195-216. Wang, C. S., J. C. Huang, and J. H. Hu. 1999. Characterization of two subclasses of PR-10 transcripts in lily anthers and induction of their genes through separate signal transduction pathways. Plant Molecular Biology 40: 807-814. Washburn, M. P., D. Wolters, and J. R. Yates. 2001. Large-scale analysis of the yeast proteome by multidimensional protein identification technology. Nat. Biotechnol. 19: 242-247. Wei, X., D. Reardon, Y. Wu, and J.-H. Sun. 2004. Emerald ash borer, Agrilus planipennis Fairmaire (Coleoptera: Buprestidae), in China: a review and distribution survey. Acta Entomology Sinica 47: 679-685. Wei, X., Y. Wu, R. Reardon, T. H. Sun, M. Lu, and J. H. Sun. 2007. Biology and damage traits of emerald ash borer (Agrilus planipennis Fairmaire) in China. Insect Sci. 14: 367-373. Whitehill, J. G. A., J. S. Lehman, and P. Bonello. 2007. Ips pini (Curculionidae: Scolytinae) is a Vector of the Fungal Pathogen, Sphaeropsis sapinea (Coelomycetes), to Austrian Pines, Pinus nigra (Pinaceae). Environmental Entomology 36: 114-120. Willekens, H., S. Chamnongpol, M. Davey, M. Schraudner, C. Langebartels, M. Van Montagu, D. Inz!©, and W. Van Camp. 1997. Catalase is a sink for H2O2 and is indispensable for stress defence in C3 plants. The EMBO Journal 16: 4806- 4816. Wittstock, U., and J. Gershenzon. 2002. Constitutive plant toxins and their role in defense against herbivores and pathogens. Current Opinion in Plant Biology 5: 300-307. Witzell, J., and J. A. Martin. 2008. Phenolic metabolites in the resistance of northern forest trees to pathogens - past experiences and future prospects. Can. J. For. Res.- Rev. Can. Rech. For. 38: 2711-2727.

229

Yan, Y. X., S. Stolz, A. Chetelat, P. Reymond, M. Pagni, L. Dubugnon, and E. E. Farmer. 2007. A downstream mediator in the growth repression limb of the jasmonate pathway. Plant Cell 19: 2470-2483. Yang, W., M. Ye, M. Liu, D. Z. Kong, R. Shi, X. W. Shi, K. R. Zhang, Q. A. Wang, and Z. Lantong. 2010. A practical strategy for the characterization of coumarins in Radix glehniae by liquid chromatography coupled with triple quadrupole-linear ion trap mass spectrometry. J. Chromatogr. A 1217: 4587-4600. Ye, M., Y. N. Yan, and D. A. Guo. 2005. Characterization of phenolic compounds in the Chinese herbal drug Tu-Si-Zi by liquid chromatography coupled to electrospray ionization mass spectrometry. Rapid Commun Mass Sp 19: 1469-1484. Yoshida, Y., R. Sano, T. Wada, J. Takabayashi, and K. Okada. 2009. Jasmonic acid control of GLABRA3 links inducible defense and trichome patterning in Arabidopsis. Development 136: 1039-1048. Yu, C. 1992. Agrilus marcopoli Obenberger, 2nd Edition ed. China Forestry Publishing House, Beijing, China. Zhong, R. Q., and Z. H. Ye. 2003. Unraveling the functions of glycosyltransferase family 47 in plants. Trends in Plant Science 8: 565-568. Zhu-Salzman, K., D. S. Luthe, and G. W. Felton. 2008. Arthropod-inducible proteins: Broad spectrum defenses against multiple herbivores. Plant Physiology 146: 852- 858. Zimmerman, M. H. 1960. Longitudinal and tangential movement in the phloem of White Ash (Fraxinus americana L.). Plant Physiology 35.

230