UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Terpenoid profiling and biosynthesis in Sitka spruce (Picea sitchensis) genotypes that are susceptible… Robert, Jeanne A. 2010

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata

Download

Media
24-ubc_2010_spring_robert_jeanne.pdf [ 1.88MB ]
Metadata
JSON: 24-1.0069660.json
JSON-LD: 24-1.0069660-ld.json
RDF/XML (Pretty): 24-1.0069660-rdf.xml
RDF/JSON: 24-1.0069660-rdf.json
Turtle: 24-1.0069660-turtle.txt
N-Triples: 24-1.0069660-rdf-ntriples.txt
Original Record: 24-1.0069660-source.json
Full Text
24-1.0069660-fulltext.txt
Citation
24-1.0069660.ris

Full Text

 TERPENOID PROFILING AND BIOSYNTHESIS IN SITKA SPRUCE (Picea sitchensis) GENOTYPES THAT ARE SUSCEPTIBLE OR RESISTANT TO ATTACK BY THE WHITE PINE WEEVIL (Pissodes strobi)   by   Jeanne A. Robert   A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF   DOCTOR OF PHILOSOPHY   in   The Faculty of Graduate Studies  (Botany)    THE UNIVERSITY OF BRITISH COLUMBIA  (Vancouver)       April 2010      © Jeanne A. Robert, 2010  ii  ABSTRACT White pine weevil, Pissodes strobi, is an insect that occurs throughout Canada that attacks a number of conifers including Sitka spruce, Picea sitchensis, a commercially and ecologically important tree for coastal B.C.. Because of attack by weevils, Sitka spruce is no longer replanted as a commercial species. The re-introduction of this species would be a valuable asset for sustainable coastal forestry. My research addresses the terpene composition and the molecular-genetic underpinning of Sitka spruce resin defenses against attack by white pine weevil.  In this thesis, I report that terpene profiles can be used to classify resistant tree genotypes. I analysed 111 different genotypes in order to determine the relationship of mono- and diterpenoid oleoresin compounds with the resistance rating. Dehydroabietic acid, a diterpene, was identified as a strong indicator of resistance. Two monoterpenes, (+)-3-carene and terpinolene were also associated with resistance in genotypes originating from the Haney region, an area which may have been subject to higher weevil pressure. In addition, I characterized weevil behavior and physiology (feeding patterns, host choice, ovary development, egg laying behavior, and larval development) in response to an extremely resistant Sitka spruce genotype (H898) in comparison to a highly susceptible genotype (Q903). My results suggest that the highly resistant genotype H898 has defense mechanisms that deter both male and female weevils during host selection and mating, that cause delayed ovary development in females, and prevent successful reproduction of weevils on H898 trees. Finally, I have identified the first (+)-3-carene and (+)-sabinene synthase genes in Sitka spruce. These terpene synthase (TPS) genes have very similar sequences, yet the encoded enzymes have different product profiles; this shows a new level of genetic diversity in the spruce TPS  iii gene family. In addition, different (+)-3-carene synthase genes are expressed in the resistant H898 tree genotype producing large amounts of (+)-3-carene, versus  the susceptible Q903 tree genotype that produces trace amounts of (+)-3-carene. This information will support the identification and breeding of resistant Sitka spruce in order to re-introduce it as a viable, native commercial species.  iv  TABLE OF CONTENTS  ABSTRACT .................................................................................................................... ii TABLE OF CONTENTS ................................................................................................. iv LIST OF TABLES ......................................................................................................... vii LIST OF FIGURES......................................................................................................... ix LIST OF ABBREVIATIONS ......................................................................................... xiii ACKNOWLEDGEMENTS ............................................................................................. xv DEDICATION .............................................................................................................. xvii CO-AUTHORSHIP STATEMENT .............................................................................. xviii 1 INTRODUCTION ................................................................................................. 1  1.1 White pine weevil biology ................................................................................. 2 1.2 Natural regulation of weevil outbreak ............................................................... 4 1.3 Current weevil-management strategies ............................................................ 6 1.4 Characteristics of weevil resistant trees ........................................................... 7 1.5 Geographic variation of Sitka spruce resistance to attack by weevil ................. 8 1.6 Resin canals and oleoresin .............................................................................. 9 1.7 Terpene compounds ........................................................................................ 9 1.8 The molecular-genetic basis of conifer terpenoid defenses ............................ 11 1.9 Purpose of this thesis ..................................................................................... 12 1.10 References .................................................................................................... 14  2 TERPENOID METABOLITE PROFILING IN SITKA SPRUCE IDENTIFIES ASSOCIATION OF DEHYDROABIETIC ACID, (+)-3-CARENE AND TERPINOLENE WITH RESISTANCE AGAINST WHITE PINE WEEVIL ................................................ 21  2.1 Introduction .................................................................................................... 21 2.2 Materials and methods ................................................................................... 23 2.2.1 Clonebank and resistance rating ................................................................ 23 2.2.2 Sample collection for metabolite profiling ................................................... 25 2.2.3 Assessment of branch age and within-tree sample location effects on terpene profiles .......................................................................................... 26 2.2.4 Terpene extraction and compound identification ........................................ 27 2.2.5 Statistical analysis ...................................................................................... 29 2.3 Results........................................................................................................... 31 2.3.1 Terpene profiles from lateral branches of the top whorl accurately reflect terpene profiles of the apical leader ........................................................... 31 2.3.2 Metabolite profiling of the Cowichan Lake clonebank identifies the diterpene resin acid dehydroabietic acid as an indicator for resistance ...................... 31 2.3.3 Accuracy of prediction of resistance groups based on terpenoid metabolites .   .................................................................................................................. 36 2.3.4 Identification of the monoterpenes (+)-3-carene and terpinolene, along with dehydroabietic acid, as indicator for resistance in genotypes from the high weevil hazard Haney area .......................................................................... 38     v 2.4 Discussion ..................................................................................................... 41 2.4.1 Dehydroabietic acid ................................................................................... 42 2.4.2 (+)-3-Carene and terpinolene ..................................................................... 42 2.4.3 Genomic and molecular underpinnings of terpene profiles ......................... 43 2.4.4 Applications for forest management and tree breeding .............................. 44 2.5 Conclusions ................................................................................................... 45 2.6 References .................................................................................................... 47  3 BEHAVIORAL, PHYSIOLOGICAL AND REPRODUCTIVE RESPONSE OF WHITE PINE WEEVIL TO RESISTANT AND SUSCEPTIBLE GENOTYPES OF SITKA SPRUCE ....................................................................................................................... 53  3.1 Introduction .................................................................................................... 53 3.2 Materials and methods ................................................................................... 55 3.2.1 Plant materials and maintenance ............................................................... 55 3.2.2 Insect rearing and maintenance ................................................................. 55 3.2.3 Assessment of constitutive and weevil-induced tree defense response ..... 56 3.2.3.1 Assessment of constitutive and traumatic resin ducts ......................... 56 3.2.3.2 Analysis of monoterpene profiles ........................................................ 58 3.2.4 Assessment of weevil responses to resistant and susceptible tree genotypes and to (+)-3-carene volatiles ...................................................................... 59 3.2.4.1 Assessment of insect ovary development ........................................... 60 3.2.4.2 Assessment of feeding and oviposition in no-choice assays ............... 61 3.2.4.3 Assessment of ovary development in no-choice assays ..................... 61 3.2.4.4 Assessment of feeding and oviposition in choice assays .................... 62 3.2.4.5 Assessment of weevil responses to (+)-3-carene volatiles using Y-tube assays ................................................................................................ 63 3.2.5 Statistical analyses .................................................................................... 65 3.3 Results........................................................................................................... 65 3.3.1 Assessment of host defense responses in weevil-resistant H898 and - susceptible Q903 Sitka spruce genotypes ............................................................. 65 3.3.2 Weevil behavior and ovary development affected by resistant and susceptible host trees in no-choice scenarios ........................................................ 69 3.3.3 Weevil behavior and reproductive success affected by resistant and susceptible host trees in choice scenarios ............................................................. 72 3.3.4 Weevil response to volatile (+)-3-carene .................................................... 76 3.4 Discussion ..................................................................................................... 79 3.5 Conclusions ................................................................................................... 82 3.6 References .................................................................................................... 84   4 RESISTANT AND SUSCEPTIBLE SITKA SPRUCE CHEMOTYPES SHOW DIFFERENTIAL EXPRESSION OF MULTIPLE (+)-3-CARENE SYNTHASE-LIKE GENES. ........................................................................................................................ 88  4.1 Introduction .................................................................................................... 88 4.2 Materials and methods ................................................................................... 91 4.2.1 cDNA isolation and functional characterization of (+)-3-carene synthase ... 92 4.2.2 Sequence confirmation with RACE ............................................................ 92 4.2.3 Functional characterization of (+)-3-carene synthase-like genes ................ 94 4.2.4 Amino acid alignment and enzyme tertiary structure .................................. 97  vi 4.2.5 Comparison of (+)-3-carene synthases in H898 and Q903 Sitka spruce genotypes .................................................................................................. 98 4.2.5.1 (+)-3-Carene synthase transcript expression patterns ........................ 98 4.2.5.2 RNA extraction protocol and cDNA synthesis ..................................... 99 4.2.5.3 Primer design and q RT-PCR protocol ............................................. 100 4.2.5.4 qRT-PCR calculations ...................................................................... 101 4.2.5.5 Statistical analysis ............................................................................ 102 4.2.5.6 Protein extraction ............................................................................. 102 4.2.5.7 Monoterpene synthase assays and enzyme activity ......................... 103 4.2.5.8 Monoterpene product profile ............................................................. 105 4.3 Results......................................................................................................... 106 4.3.1 Functional characterization identifies seven full-length (+)-3-carene synthase-like cDNAs ................................................................................ 106 4.3.2 Amino acid sequence comparisons show three types of (+)-3-carene synthase-like enzymes: CAR_1, CAR_2, and SAB .................................. 109 4.3.3 UTR groupings identify a fourth (+)-3-carene synthase-like sequence: CAR_3 ..................................................................................................... 115 4.3.4 Comparison of (+)-3-carene synthases in H898 and Q903 Sitka spruce genotypes ................................................................................................ 116 4.3.5 Comparison of monoterpene synthase enzyme activities in H898 and Q903 .   ................................................................................................................ 119 4.3.6 Comparison of monoterpene metabolite profiles in H898 and Q903 ......... 119 4.4 Discussion ................................................................................................... 122 4.4.1 Enzyme activity and terpene profile: an area for further research. ............ 124 4.5 Conclusions ................................................................................................. 125 4.6 References .................................................................................................. 127  5 GENERAL CONCLUSIONS ............................................................................ 134  5.1 References .................................................................................................. 139  APPENDIX: Chapter 2 Supplementary Mixed Model ANOVA output ..................... 141   vii LIST OF TABLES  Table 2-1 The average extracted compound (µg/g dry weight ± 1SE) for the resistant and susceptible groups in combination with the output of mixed model ANOVAs testing the difference between resistant and susceptible trees for each terpene compound.   .... 33  Table 2-2 Coefficients of the linear discriminants (LD1 and LD2) for the 27 compounds used in the discriminant function analysis.   ..................................................................... 35  Table 2-3 The observed resistance group (obs.) compared to the predicted resistance group (pred.) generated by the discriminant function model created from the entire data set as compared to the “leave-one-out” cross validation. The percentage of trees correctly identified (% corr.) using each model is also shown.   ....................................... 37  Table 2-4 Output of mixed model ANOVAs testing the difference between resistant and susceptible trees for each terpene compound for the Haney region only.   ...................... 39  Table 2-5 The average amount ± 1SE (µg /g dry weight) of terpenoid compounds associated with resistance for trees originating in Haney, British Columbia.   .................. 40  Table 4-1 3’-RACE gene-specific primer combinations.   ................................................. 93  Table 4-2 Gene-specific qRT-PCR primers (oriented 5’ to 3’). Stop codons are highlighted in bold.   ...................................................................................................... 100  Table 4-3 Percentage of each monoterpene product for the functionally characterized 3- carene synthase-like enzymes from Sitka spruce genotypes H898, Q903, and FB3_425 as they compare to the gene cloned from Interior spruce (F08) and the previously identified (+)-3-carene synthase from Norway spruce (JF67).   ..................................... 108  Table 4-4 Pairwise percent amino acid sequence identities of the functionally characterized cDNAs in H898 and Q903 in comparison with the Norway spruce (+)-3- carene synthase gene.   ................................................................................................ 110  Table S 1 – Output from ANOVA for each compounds identified for differences among cardinal directions.   ...................................................................................................... 141  Table S 2 – Output from ANOVA for each compound identified for differences between leader and the top branches.   ....................................................................................... 141  Table S 3 – Mixed model ANOVA output testing the difference between trees from the intermediate group and the susceptible group. Significant values are in bold. ND = not detected.   ..................................................................................................................... 142  Table S 4 – Mixed model ANOVA output testing the difference between trees from the intermediate group and the resistant group. Significant values are in bold. ND = not detected.   ..................................................................................................................... 143  Table S 5 – Mixed model ANOVA output testing the difference between trees from the Haney (H) region versus Big Qualicum (BQ). Significant values are in bold. ND = not detected.   ..................................................................................................................... 144  viii Table S 6 – Coefficients of linear discriminants (LD1 and LD2) for the Haney region only. Magnitude of the LD value represents the weight of that compound in the creation of the discriminant function.   ................................................................................................... 145  Table S 7 – Coefficients of the linear discriminant (LD1) for the Big Qualicum region. Magnitude of the LD value represents the weight of that compound in the creation of the discriminant function.   ................................................................................................... 146  Table S 8 – Mixed model ANOVA output testing the difference between trees from the resistant group and the susceptible group for the Big Qualicum region only.   ............... 147  ix LIST OF FIGURES  Figure 1-1 White pine weevil life stages and damage. Panel A shows an adult white pine weevil. Panel B shows a weevil-attacked leader curling over and forming the characteristic shepherd’s crook. Panel C shows developing larvae develop inside the dying leader. Images taken from: http://www.nrcan-rncan.gc.ca/cfs- scf/science/prodserv/pests/white_pine_weevil_e.html, http://www.pfc.forestry.ca/entomology/weevil/index_e.html.   ............................................ 3  Figure 2-1 Locations of origin for tree genotypes planted at Cowichan Lake clonebank. The filled circles indicate locations of origin including the Queen Charlotte Islands (OCI), Sitka spruce and interior spruce hybrid zone (HYB), west Vancouver Island (WVI), Big Qualicum (BQ), Squamish (SQ), Haney (H), and Hoquiam (HOQ). The filled triangle shows the sampling location at the Cowichan Lake clonebank. The grey hatched area indicates the high weevil hazard area. This map was modified from King et al. (2004).   ...................................................................................................................... 24  Figure 2-2 Discriminant function plot. Plot of the discriminant functions (LD1 and LD2) for actual data designated by the filled symbols and using blue for susceptible, red for intermediate, and green for resistant tree genotypes. The category predicted by the model is shown by the colour of the outer circle around each data point. The discriminant functions separated the three resistance categories with LD1 showing the best separation between resistant and susceptible genotypes, and LD2 showing the best separation of the intermediate genotypes.   ..................................................................... 36  Figure 3-1 The number and area of cortical and induced resin ducts for weevil- attacked and control trees resistant versus susceptible tree genotypes. The average number of constitutive cortical resin ducts (A) and induced traumatic resin ducts (B) per millimeter stem circumference (± 1SE) and the average area of cortical resin ducts (C) and induced traumatic resin ducts (D) per duct (± 1SE) for the resistant Sitka spruce genotype (H898) and the susceptible genotype (Q903). Although there are no apparent differences between the resistant and susceptible trees constitutive resin canals, the resistant genotype showed a significantly larger number and area of induced traumatic resin canals after weevil attack.   ..................................................................... 66  Figure 3-2 Leader diameter and bark thickness for resistant and susceptible tree genotypes. Panel A shows the leader diameter (mm) and panel B shows the average bark thickness (mm) for Sitka spruce genotypes the resistant genotype (H898) and the susceptible genotype (Q903). Different letters above the bars indicate a significant difference at (p<0.05). Although the susceptible leaders had a larger diameter, there was no significant difference in bark thickness between the resistant and susceptible genotypes.   .................................................................................................................... 67  Figure 3-3 The resistant and susceptible monoterpene profiles in response to weevil feeding. The average amount of monoterpene measured (µg/g dry weight ± 1SE) on each sampling day for H898 (A-D) and Q903 (E-H). Asterisk indicates a significant difference between treatment and control (p<0.05). nd = not detected. None of the measured monoterpenes changed significantly with weevil feeding over time except for (+)-3-carene which was significantly reduced on day 15 and day 22 in the weevil-feeding treatment. There was no detectable (+)-3-carene in the susceptible genotype Q903.   ............................................................................................................. 68  x  Figure 3-4 The cumulative number of weevil feeding holes on the leader and interwhorl of the resistant and susceptible tree genotypes. The number of feeding holes observed on H898 and Q903 trees (± 1SE) on the leader (A) and on the interwhorl directly below the leader (B). An asterisk indicates a significant difference between genotypes (p<0.05). The number of feeding holes on the susceptible leader is significantly higher than the number on the resistant leader on day 2, and the number of feeding holes on the resistant internode is significantly higher on day 2.   ....................... 70  Figure 3-5 Weevil ovary development on resistant and susceptible tree genotypes. The percentage weevils with low (gray bars), moderate (white bars), or mature (black bars) ovary development for each sampling day. Weevils were restricted to either H898 host trees (top) or to Q903 host trees (bottom). Weevils feeding on the resistant genotype had fewer mature ovaries by day 7, and fewer mature ovaries on day 21 after peak ovary maturation subsided.   ................................................................................... 71  Figure 3-6 Female and male choice of resistant versus susceptible host tree genotypes. The percentage of female weevils when mixed with male weevils (A), male weevils when mixed with female weevils (B), and male weevils only (C), observed on the susceptible genotype Q903 and on the resistant genotype H898 at each sampling day. The majority of female and male weevils were observed on susceptible trees although males appear to lose this preference at approximately 14 days. The male choice of susceptible trees is independent of the presence of females.   ........................................ 74  Figure 3-7 Cumulative feeding and oviposition punctures for weevils given a choice between resistant and susceptible host tree genotypes. The average number of feeding and oviposition holes (± 1SE) counted on the leader, interwhorl, and over the entire tree for the resistant H898 genotype and the susceptible Q903 genotype over 23 days of continuous weevil exposure when the weevils were given a choice between tree genotypes. When given a choice, weevils feed more on the susceptible trees and on the susceptible leader tissue.   .................................................................... 75  Figure 3-8 The number of surviving larvae observed on resistant and susceptible tree hosts for weevils given a choice between resistant and susceptible tree hosts. The number of larvae found in the leader and interwhorl of H898 versus Q903 when the weevils were given a choice between tree genotypes. There were no larvae recovered from the resistant H898 genotype whereas the susceptible leaders averaged nine live surviving larvae on the leader and two surviving larvae in the interwhorl tissue.   ...................................................................................................................................... 76  Figure 3-9 Dilutions of (+)-3-carene causing deterrence in Y-tube bioassays. The number of female weevils (A) and male weevils (B) either moving towards increasing dilutions of (+)-3-carene in pentane versus those that moved away or showed no response. A significant difference (p<0.05) between insect movement toward the treatment versus movement away is designated by an asterisk. Significant numbers of both males and females moved away from a 5% dilution of (+)-3-carene in pentane.   ... 77      xi Figure 3-10 Comparison of three monoterpenes, the resistant tree genotype and the susceptible tree genotype in Y-tube bioassays. The number of female weevils (A) and male weevil (B) either moving towards or away from a 5.0% dilution of (+)-3- carene compared with 5.0% dilutions of (+)-α-pinene and (-)-limonene, and also compared with tree genotypes H898 and Q903. A significant difference (p<0.05) between insect movement toward the treatment versus movement away is designated by an asterisk. All three monoterpenes showed similar levels of deterrence at the 5% dilution. The weevils did not distinguish between either unwounded or wounded resistant versus susceptible trees in the Y-tube bioassays.   ......................................................... 78  Figure 3-11 The amount of volatile (+)-3-carene reaching the insect from each treatment in the Y-tube bioassays. The average amount of (+)-3-carene measured in the Y-tube apparatus (µg/L air flow) from the treatment dilutions (A) and from unwounded and wounded H898 trees compared to known dilutions of (+)-3-carene (B). The volatile (+)-3-carene reaching the insect from an unwounded tree is undetectable by this method, and the amount of volatile (+)-3-carene from a wounded tree is equivalent to the amount of (+)-3-carene volatized from a 0.1% dilution of pure compound.   .......... 79  Figure 4-1 The conifer TPS-d subfamily. The conifer TPS-d subfamily is divided into three groups (TPS-d1, TPS-d2, and TPS-d3). The (+)-3-carene synthase identified in Norway spruce is designated by the red rectangle. Figure is modified from Martin et al. 2004.   ............................................................................................................................. 89  Figure 4-2 The major products of the Sitka spruce (+)-3-carene synthases and (+)- sabinene synthase. Total ion chromatogram of the products for each of the functionally characterized TPS enzymes encoded by candidate cDNAs (A): the (+)-3-carene synthases from H898 (H08 and H02), the (+)-3-carene synthase from Q903 (Q09), and the sabinene synthases (H05 and Q05). The mass spectra are shown for the major enzyme products including 3-carene (B), sabinene (D), and α-terpinolene (F) and this compared to the mass spectra derived from authentic standards (C, E, G).   ................ 107  Figure 4-3 Phylogenic tree of the sequenced (+)-3-carene synthases and the (+)- sabinene synthases. This tree is based on amino acid sequences of the (+)-3-carene and sabinene synthase genes in H898 and Q903 that have been functionally characterized. The tree also includes the FB3_425 Sitka spruce genotype (+)-3-carene synthase, the Interior spruce (+)-3-carene synthase, and the Norway spruce (+)-3- carene synthase (Fäldt et al. 2003). These are compared to a Sitka spruce (−)-pinene synthase (C. Keeling unpublished data) as an outgroup. The scale bar is equivalent to 0.1 amino acid substitutions per site. The (+)-3-carene synthase sequenced from the susceptible genotype (Q903) is more similar to (+)-3-carene synthases sequenced in other genotypes to the (+)-sabinene synthases than to the (+)-3-carene synthases sequenced from the resistant tree genotype (H898) at the amino acid level.   ............... 111  Figure 4-4 The amino acid alignment of the functionally characterized 3-carene synthase and sabinene synthase enzymes. The amino acids that differ between the 3-carene synthases and the sabinene synthases are marked with either a ‘•’ indicating a conservative amino acid change or a ‘▼’ indicating a non-similar amino acid change. Those amino acids that are within 20Å of the substrate in the modeled three-dimensional structure are indicated by the pink bar above the alignment. The white arrows in the pink bar represent those amino acids that appear to be directly involved with the substrate in the active site.   ............................................................................................................. 115  xii  Figure 4-5 Gene expression patterns for the (+)-3-carene synthases, the (+)- sabinene synthase and (−)-α-pinene synthase in resistant and susceptible tree genotypes. The average gene expression fold change (± 1SE) over the translation initiation factor (TIF) reference gene expression for each (+)-3-carene synthase (CAR_1, CAR_2, CAR_3) and the (+)-sabinene synthase (SAB) gene sequences using the α- pinene synthase (PIN) gene expression for comparison. An asterisk denotes a significant difference (p < 0.05) between methyl jasmonate treatment and the tween control. Transcripts of CAR_2 were not detectable (nd) in Q903, and transcripts of CAR_3 were not detectable (nd) in H898.   ................................................................... 117  Figure 4-6 Enzyme product and metabolite extract for (−)-α-pinene, (+)-sabinene, and (+)-3-carene for resistant and susceptible tree genotypes. The average amount of product (ng ± 1SE) per 100 µg of total protein extract and the average amount (µg/g dry weight ± 1SE) of extracted compound for the susceptible genotype Q903 and the resistant genotype H898. An asterisk denotes a significant difference (p < 0.05) between methyl jasmonate treatment and the tween control. Although metabolite extract levels for (+)-3-carene and (+)-sabinene are present in H898 and present in trace amounts in Q903, enzyme products are essentially not detectable in enzyme assays. Enzyme assays and metabolite extractions showed the presence of (−)-α-pinene.   ................... 121   xiii LIST OF ABBREVIATIONS A – adenine bp – base pairs BQ – Big Qualicum BCMFR – British Columbia Ministry of Forests and Range C – cytosine cDNA – copy DNA constructed from RNA template CFS – Canadian Forest Service Ct – number of PCR cycles to a threshold fluorescence value CYP – cytochrome P450 DMAPP – dimethylallyl diphosphate EST – expressed sequence tag FID – flame ionization detector FPP – farnesyl diphosphate G – guanine GCMS – gas chromatography coupled with mass spectrometry GPP – geranyl diphosphate GGPP – geranylgeranyl diphosphate H898 – resistant tree genotype 898 from Haney, British Columbia HOQ – Hoquiam, Washington HPLC – high-performance liquid chromatography HYB – Sitka spruce and interior spruce hybrid zone located in the Nass/Skeena area IPP – isopentyl diphosphate JH – juvenile hormone (insects) LB – Luria-Bertani broth, a liquid medium used to grow bacteria MeJa – methyl jasmonate  xiv MEP – 2-C-methylerythritol-4-phosphate mRNA – messenger RNA MTBE – tert-butyl methyl ether m/z – mass-to-charge ratio NPP – neryl diphosphate NRQ – normalized relative quantity (for use in quantitative real-time RT-PCR) ORF – open reading frame PCR – polymerase chain reaction Q903 – susceptible tree genotype 903 from the Queen Charlotte Islands, British Columbia QCI – Queen Charlotte Islands qRT-PCR – Quantitative real time RT-PCR RACE – rapid amplification of cDNA ends RNA – ribonucleic acid RT-PCR – reverse transcription polymerase chain reaction SIM – single ion monitoring (in GCMS) SRM – selected reaction monitoring SQ – Squamish, British Columbia T – thymine TB – terrific broth TIF – translation initiation factor EIF-5A TPS – terpene synthase tRNA – transfer RNA UTR – untranslated region (of mRNA) WVI – west Vancouver Island  xv  ACKNOWLEDGEMENTS I must first thank my husband, Terry Robert; I will always appreciate your unfailing love and support through my many endeavors. I would like to thank my daughter Genevieve, who cheers me at every thought.  I am also grateful to my brother, Andrew Horning, my sister, Susan Horning, and my mother, Trish Muller, for their interest and encouragement, and to my Grannie, Molly Horning, for her bi-weekly phone calls filled with love and support. I am thankful too, for my Dad, Patrick Horning; I will always miss his enthusiasm for my projects and his steady support. I am also extremely grateful to my PhD. supervisor, Jörg Bohlmann for scientific expertise, encouragement, and extraordinary support through my attempts to balance motherhood with the demands of a PhD. I also appreciate the efforts of my committee members, Reinhard Jetter and Kermit Ritland, to guide my research and for their valuable feedback throughout my project. I would like to thank our collaborators at the Ministry of Forests and Range, Alvin Yanchuk, Research Leader, Forest Genetics, B.C. Ministry of Forests and Range (BCMFR), John King, Research Scientist, BCMFR, and John Ogg, propagation technician, BCMFR, Rene Alfaro, Research Scientist in Forest Entomology, Canadian Forest Service, for time and consultation, feedback, and invaluable research materials. I also wish to thank Regina Gries from Simon Fraser University for information and equipment used in the insect bioassays. I must thank Chris Keeling and Dawn Hall, the best of office-mates, for tremendous scientific expertise, technical advice, understanding, emotional support and, most astoundingly, for laughing at my jokes. I am also indebted to Lina Madilao for plying me with wine, invaluable information, and so much encouragement. I have to thank Sharon Jancsik for patience, fantastic lab work, good snacks, and general wonderfulness. I am  xvi grateful to ‘team NCE’, Hannah Henderson, Maria Li, and Harpreet Dullat, for tremendous support, fashion advice, and for patiently fielding my many questions. For help with fieldwork, weevils, and trees, I am grateful to the wizards of the greenhouse, Tristan Gillan and Alfonso Lara Quesada. I would also like to thank Rick White for statistical advice and those mysterious R commands, and especially Karen Reid and Carol Ritland for fantastic lab management, laughs, and rational dialogue at crucial junctures in my PhD. I also appreciate Dona Sharma for her work as a summer undergraduate assistant and all other Bohlmann laboratory members who helped along the way.  xvii  DEDICATION    Dedicated to the memory of my Dad.   xviii CO-AUTHORSHIP STATEMENT  Chapter 2: Terpenoid metabolite profiling in Sitka spruce identifies association of dehydroabietic acid, (+)-3-carene and terpinolene with resistance against white pine weevil. This chapter was written by the author (J. Robert). The manuscript was reviewed and revised by Dr. Jörg Bohlmann (Professor, Ph.D. supervisor). This chapter was also reviewed by Dr. Alvin Yanchuk (Researcher leader, BCMFR), Dr. John King (Research scientist, BCMFR), Mr. Rick White (Managing Director, Statistical Consulting and Research Laboratory), Dr. Chris Keeling (Research Associate, Bohlmann Laboratory), Dr. Dawn Hall (Post-doctoral fellow, Bohlmann Laboratory), Dr. Kermit Ritland (Professor, Ph.D. committee member), and Dr. Reinhard Jetter (Associate Professor, Ph.D. committee member).  A number of people helped with the field collection at Cowichan Lake but the logistics, experimental design, and collection methods were all developed and implemented by the author. Input for GCMS programs and parameters as well as instruction on the proper use of the equipment was provided by Lina Madilao (Facilities Manager, Core Analytical Facility), but the GCMS data collection, analysis, and calculations were completed by the author. Statistical analysis was done in collaboration with Rick White. Statistical output was analysed and interpreted by the author.  Chapter 3: Behavioral, physiological and reproductive response of white pine weevil to resistant and susceptible genotypes of Sitka spruce. The experimental design, implementation, analysis, and writing for this chapter were entirely the work of the author. The manuscript was reviewed and revised by Dr. Jörg Bohlmann (Professor, PhD supervisor). This chapter was also reviewed by Dr. Chris Keeling (Research Associate, Bohlmann Laboratory), Dr. Dawn Hall (Post-doctoral fellow, Bohlmann  xix Laboratory), Dr. Kermit Ritland (Professor, Ph.D. committee member), and Dr. Reinhard Jetter (Associate Professor, Ph.D. committee member).  Chapter 4: Resistant and susceptible Sitka spruce chemotypes show differential expression of multiple (+)-3-carene synthase-like genes. This chapter was written the author. The manuscript was reviewed and revised by Dr. Jörg Bohlmann (Professor, Ph.D. supervisor). This chapter was also reviewed by Dr. Chris Keeling (Research Associate, Bohlmann Laboratory), Dr. Dawn Hall (Post-doctoral fellow, Bohlmann Laboratory), Dr. Kermit Ritland (Professor, Ph.D. committee member), and Dr. Reinhard Jetter (Associate Professor, Ph.D. committee member). Much of the early cloning work (section 4.2.1) and trouble-shooting was completed in close collaboration with Dr. Christopher I. Keeling and Britta Hamberger (Research assistant, Bohlmann Laboratory). Dr. Keeling and Ms. Hamberger cloned a number of 3-carene synthase-like gene candidates during my maternity leave in 2006. All of the final functional characterization and data analysis for the recombinant enzymes was done by the author. In the methyl-jasmonate treatment case study (section 4.2.5), Sharon Jancsik (Research assistant, Bohlmann Laboratory) extracted the RNA and performed the quantitative real- time RT-PCR plate set-up and reading. The author created the plate design, the primers, conducted quality control analysis, and the final data analysis for the quantitative real- time PCR. The enzyme assays were refined in close collaboration with Dr. Dawn Hall. All other aspects of the project were completed by the author.     1  1 INTRODUCTION Sitka spruce bark is soaked with apparently toxic terpene compounds, yet on most trees, white pine weevils thrive. Sitka spruce is a native, and commercially viable, species that is no longer replanted after harvest on British Columbia’s west coast because of its susceptibility to attack by weevils. In spite of this, however, Sitka spruce shows a remarkable range of resistance to attack by weevils. For the purpose of this thesis, resistance is defined as the ability of a plant to reduce damage by insect herbivores. In this thesis, I focused on the chemical compounds produced by Sitka spruce that result in reduced damage by white pine weevils. While most trees are repeatedly attacked year after year, there are some Sitka spruce trees that show almost complete resistance (i.e. sustain almost no damage) to attack by white pine weevils.  This thesis was designed to investigate the role of terpene compounds in the resistant trees. The introductory chapter (Chapter 1) details the insect-tree interaction that forms the basis of this research and briefly summarizes the many years of research that sets the foundation for the questions in my thesis. The introduction describes the fundamentals of weevil biology and damage, followed by information on biology of Sitka spruce and the details of tree defense against attack by the white pine weevils, specifically discussing the definition and role of terpene compounds contained in tree oleoresin. The subsequent research chapters move from a broad-scale analysis of the relationship between terpene compounds and white pine weevil resistance in Sitka spruce (Chapter 2), to the impact of tree defenses on the behavior and physiology of adult weevils exposed to resistant and susceptible tree genotypes (Chapter 3), and finally to a case study designed to investigate the molecular-genetic underpinnings for the production of one particular monoterpene, (+)-3- carene, that may play a role in tree resistance (Chapter 4). The concluding chapter (Chapter   2 5) synthesizes the new information from my thesis with findings from previous research, with broader implications for management of forest pests, and with the application of tree breeding for resistance.  1.1 White pine weevil biology White pine weevil (Pissodes strobi Peck.) occurs throughout Canada and attacks several commercially important conifer hosts. The most important hosts in British Columbia are Sitka Spruce, Picea sitchensis (Bong.) Carr., Engelmann spruce, Picea engelmannii Parry, and White spruce Picea glauca (Moench) Voss. Other hosts in North America include Eastern White Pine, Pinus strobus L., Jack Pine, Pinus banksiana, and Norway Spruce, Picea abies (L.) Karst. (Alfaro 1995).  Spruce is a tree species of major importance in British Columbia. Because of attack by white pine weevils, however, Sitka spruce is virtually non-existent as a commercial species (Tomlin and Borden 1994). This is a significant economic and ecological loss as Sitka spruce is a native species to coastal British Columbia with both high wood and fibre quality (Mitchell and Grout 1995). Interior spruce, Picea glauca x engelmannii, is still widely planted in British Columbia, but productivity losses in these plantations from weevils are substantial (Mitchell and Grout 1995); attack by weevils can reduce tree growth by up to 40% (Alfaro 1998).  Adult weevils make short flights early in the spring to facilitate dispersal although there is evidence to suggest that weevils often crawl, and rarely fly, later in the season (Harman and Kulman 1967). Adults lay their eggs at the tip of the previous year’s leader in May and early June. The eggs are laid inside excavations in the bark and are held in place with fecal matter. The larvae hatch in approximately ten days and feed downwards along the leader.   3 The larvae consume the cortex and phloem killing the previous year’s leader and causing the current year’s extending leader to die and curl into the characteristic “shepherd’s crook” form (Figure 1-1).  A B C  Figure 1-1 White pine weevil life stages and damage. Panel A shows an adult white pine weevil. Panel B shows a weevil-attacked leader curling over and forming the characteristic shepherd’s crook. Panel C shows developing larvae develop inside the dying leader. Images taken from: http://www.nrcan-rncan.gc.ca/cfs- scf/science/prodserv/pests/white_pine_weevil_e.html, http://www.pfc.forestry.ca/entomology/weevil/index_e.html.   The larvae pupate in chip cocoons at the base of the leader and new adults emerge in August or September of the same year. If weather conditions are unfavorable, then larvae or pupae may overwinter and emerge in the following spring. The life cycle is completed within one year.  Stem deformation occurs when lateral branches below the attacked leader compete for apical dominance. This can cause a forked or dwarfed tree form. Repeated attacks on new dominant laterals cause severe stem deformation over time. Thus, weevil attack causes   4 growth losses, a reduction in wood quality, a decrease in lumber recovery, and potentially tree death (Alfaro 1989).   1.2 Natural regulation of weevil outbreak Weevil populations are regulated by several major factors including host selection behavior, larval resource supply, climate, predators and parasitoids (Alfaro 1996), and physiological effects of host toxicity.  Host selection by adult white pine weevils has been studied for over 40 years. VanderSar et al. (1977a) hypothesized that the movement of weevils from eastern Canada to the west coast occurred as a result of modified host selection that allowed weevils to exploit relatively susceptible spruce species in British Columbia. Lewis et al. (2000) identified three sub-populations of white pine weevil in British Columbia: North Central Coast BC, Interior BC, and South Coast BC. Lewis et al. (2000) showed low levels of inbreeding and significant gene flow between populations but suggested that different populations may be using slightly different host selection behaviors. Tanaka (1999) found that coastal populations responded differently to host and male-produced volatiles than interior populations. VanderSar and Borden (1977b; 1977c) suggested that weevils visually orient to host trees, and then may use a combination of geotaxis and phototaxis, host volatiles, and gustatory sampling to determine appropriate host trees and oviposition sites. In contrast, Sahota et al. (1994) suggested that, when an adult female lands on a susceptible host tree, ovary maturation begins and the weevil’s ability to fly decreases. Sahota et al. (1994) suggested that host selection is an insect physiological response to a suitable food and oviposition source.    5 Once a suitable host tree is identified, successful attack depends both on weevil numbers and host suitability. Weevil larvae must be present in sufficient numbers to girdle the tree and overcome the host tree’s defense effort. Alfaro et al. (1996) found that adult emergence was very low in trees with fewer than 60 egg punctures per leader. Successful attacks had more egg punctures and a lower traumatic resin duct response (Alfaro et al. 1996). Thus, Alfaro et al. (1996) suggested that white pine weevils incorporate a mass attack strategy similar to that used by some bark beetles as larger numbers of oviposition sites on the leader result in a decreased resistance response by the host tree.  Climate, predators and parasites are also regulators of weevil attack. Weevils require a threshold accumulation of degree days (888 degree days above 7.2 ˚C for Sitka spruce) to complete larvae development and pupation (McMullen 1976; Overhulser and Gara 1981). In addition, adults require adequate levels of temperature and humidity for both feeding and oviposition. Overwintering may also be a significant cause of weevil death in natural conditions (Bellocq and Smith 1996). Some death also occurs as a result of predation by Lonchaea cortices (Diptera: Lonchaeidae), a species of fly (Hulme 1989; Hulme 1990).  Host suitability is also determined by the presence or absence of toxins. Toxins in host plants may affect a number of targets in the herbivorous insect including the insect nervous system, endocrine system, gonads, heart, or epidermis (Roitberg and Isman 1992). Leal et al. (1997) showed that adult weevils feeding on resistant trees can experience reduced egg production. They found that vitellogenin, an egg yolk protein precursor, showed low expression in weevils feeding on one resistant tree genotype. This lack of vitellogenin was associated with incomplete ovary development whereas adult female weevils fed on susceptible genotypes showed high expression of vitellogenin and full ovary development containing four to six fully developed eggs. The authors suggest a link between the juvenile   6 hormone (JH) production pathway and the expression of vitellogenin as the addition of JH to females fed on resistant trees caused a resurgence in the expression of vitellogenin. JH formation is intimately linked to terpenoid metabolism in insects (Seybold and Tittiger 2003) and may be altered significantly by the presence of pathway components (JH analogues or other terpenoid compounds) in host trees. There is also a suggestion that feeding deterrents or toxins alter weevil behavior. Alfaro et al. (1980) suggested that monoterpene compounds [e.g. (+)-camphor and limonene] stimulated feeding at low concentrations but became deterrent as the concentration increased.  1.3 Current weevil-management strategies Currently, the efforts to control weevil attack have focused on risk assessment and reduction. Risk assessment is based on a number of factors:  • levels of weevil populations in the local area, • stand age, a stand is most susceptible between age 10 to 30, • site index and growth rate, long leaders and thick bark are beneficial for weevil growth and development, • the accumulation of degree days required for the weevil to complete its life cycle, • the availability of host tree species, • stand density, dense stands have lower defect rates, • landscape heterogeneity, large monocultures of host increase risk of outbreaks (Alfaro 1998).  Weevil outbreaks are facilitated by monocultures of host trees on warm, highly productive sites. Management recommendations for controlling weevil outbreak include   7 dense replanting of host trees, planting non-host species, and planting a species mixture that will provide deciduous overstory (shading) for the developing host trees. Although it is unlikely that these methods will completely eliminate the weevil from commercially important plantations, the impact of weevil damage may be reduced. These control measures are not perfect however; Alfaro and Omule (1990) show that even densely planted Sitka spruce trees sustain an estimated 22% growth loss. Tree breeding programs initiated by the British Columbia Ministry of Forests and Range (BCMFR) have thus focused on the identification of resistance traits in Sitka spruce. Fill-planting (supplemental re-planting of areas with patches of dead or dying trees) resistant genotypes in current stands of susceptible trees and future plantations of more resistant trees in high weevil hazard areas are major goals in future weevil management strategies.  1.4 Characteristics of weevil resistant trees Even though silvicultural treatments such as shading and planting of non-host species can reduce productivity losses associated with weevils, these interventions are not sufficient to enable commercial-scale regeneration of Sitka spruce. The focus of recent research, therefore, has moved from silvicultural approaches to conducting tree breeding for resistance (King and Alfaro 2009). A number of large-scale studies have been conducted over the last 30 years in order to identify families of resistant trees (Ying 1991; Alfaro and Ying 1990). Although proximity to a weevil population increases the probability of attack (Alfaro et al. 1993), there are also a number of tree-inherent characteristics that have been studied as possible breeding targets for weevil resistance (Alfaro et al. 2004; NRC 2004):  • feeding stimulants and deterrents in bark (e.g. chemical attractants, toxins, or physical barriers to weevils such as sclerids or stone cells) • inhibition of weevil reproductive physiology   8 • resin canal density • physical and chemical properties of oleoresin • traumatic resin composition and flow • other morphological and anatomical characteristics  The large number of characteristics that can potentially increase tree resistance to weevils, the large variation in resistance, and the gradation of resistance among individuals suggest multi-allelic or multigenic resistance in Sitka spruce (Alfaro et al. 1993). Resistance characteristics can occur in combination, and the importance of a particular resistance trait can vary among individuals (Alfaro 1995). The genetic basis of resistance would help to more effectively target tree breeding programs and to ensure that a multitude of resistance mechanisms are present in deployed genotypes. Targeted selection of genes that are responsible for Sitka spruce resistance to weevils may ultimately enable the re-introduction of conifers like Sitka spruce, a commercially valuable, and now largely absent, native tree species into coastal British Columbia.  1.5 Geographic variation of Sitka spruce resistance to attack by weevils A number of studies conducted by the British Columbia Ministry of Forests show that variation in tree resistance also occurs at the landscape level. Patches of weevil resistant trees occur throughout British Columbia. Alfaro and Ying (1990) tested 34 Sitka spruce provenances (represented by several trees from 5-14 families) from areas on Vancouver Island, coastal British Columbia and along the southernmost arm of Alaska. Provenance accounted for 9% of the variation in the number of attacks per tree, although the effects of differential growth rates in different provenances were not taken into account for this analysis. Two areas on the southern mainland, Haney and Squamish, were noted in   9 particular to have fewer than average attacks per tree even though these areas are rated as high weevil hazard areas. Several provenances in the Skeena river area of northern British Columbia also showed lower than average number of attacks per tree. Because Sitka spruce often hybridizes with interior spruce in this area, Alfaro and Ying (1990) suggest that the hybrid trees may be more resistant to attack. O’Neill et al. (2001) compared the properties of traumatic resin canal formation in Sitka versus Interior spruce and suggest that the traumatic resin response of Interior spruce may be stronger than that of Sitka spruce.  1.6 Resin canals and oleoresin Resin canals have been long identified as constitutive defense mechanisms in conifers that must be avoided by an attacking weevil (Tomlin and Borden 1994; Alfaro 1995). In several species of spruce, including Sitka spruce, resin canals can be induced shortly after attack by weevils (Alfaro 1995). The extent of the induced traumatic resin canal response can vary depending on the severity of weevil attack. Alfaro et al. (1995) suggest that lighter, discontinuous rings of induced resin canals result from feeding damage alone, whereas oviposition and larval feeding induces a stronger, continuous response around the entire circumference of the stem. Induced traumatic resin canals contain high concentrations of terpenes excreted by cells in a localized necrotic area (Shrimpton 1973). Thus, the investigation of the terpene composition of oleoresin may further elucidate the resistance mechanisms employed by resistant trees.  1.7 Terpene compounds Terpenoids, or isoprenoids, are a very large group of compounds originating from a common five-carbon precursor molecule. More than 25,000 terpenoid compounds have been identified (Croteau et al. 2000).  Terpenes have been implicated in tree resistance in a number of conifer species including Sitka spruce. Tomlin and Borden (1994) suggest that   10 the odour of terpene compounds may be attractive to weevils. Tomlin and Borden (1994) also found that resistant Sitka spruce trees contained on average significantly more total diterpene resin acids (the total concentration of pimaric acid, isopimaric acid, levopimaric acid, palustric acid, dehydroabietic acid, abietic acid, and neoabietic acid per gram of tissue dry weight) than susceptible trees.  Terpenoid compounds are defined by five-carbon precursor molecules that link to form the precursors to monoterpenes (ten-carbon molecules), sesquiterpenes (fifteen-carbon molecules), and diterpenes (twenty-carbon molecules). There are four stages in terpenoid biosynthesis: 1) the formation of C5 units, 2) the formation of C10, C15, and C20 carbon chains by prenyltransferases, 2) the cyclization of these chains by terpene synthases, and 4) in some cases, further modification of terpene compounds by cytochrome P450s, for example.  The C5 unit occurs in two forms, dimethylallyl diphosphate (DMAPP), and its isomer, isopentyl diphosphate (IPP). DMAPP is converted from IPP through the action of IPP isomerase. The IPP precursor is made via the mevalonate pathway in the cytosol (Newman and Chappell 1999) and via the 2-C-methylerithritol-4-phosphate (MEP) pathway from pyruvate and glyceraldehyde-3-phosphate in the plastid (Eisenreich et al. 1998; Lichtenthaler 1999; Rohmer 1999). Some of the IPP synthesized in the plastid may be exported into the cytosol for use in the synthesis of sesquiterpene compounds (Croteau et al. 2000).  The C5 units remain single or are joined by prenyltransferases (Davis and Croteau 2000). Geranyl diphosphate (GPP) is the ten-carbon precursor for monoterpene production. More recently, neryl diphosphate (NPP), an isomer of GPP, was described as an alternative   11 monoterpene synthase substrate in tomato glandular trichomes (Schilmiller et al. 2009). Farnesyl diphosphate (FPP) is the fifteen-carbon precursor for sesquiterpene production, and geranylgeranyl diphosphate (GGPP) is the twenty-carbon precursor for diterpene production (Croteau et al. 2000).  Terpene synthase enzymes convert prenyldiphosphates to hemiterpenes, monoterpenes, sesquiterpenes and diterpenes. Synthesis of hemiterpenes (e.g. isoprene), monoterpenes, and diterpenes takes place in the plastid whereas sesquiterpenes are synthesized in the cytosol (Croteau et al. 2000). Although some terpenes are acyclic (e.g. myrcene and ocimene), most are cyclized by terpene synthases to produce a diverse array of terpene compounds in the plant.  Several processes can modify terpenoid compounds including oxidation, reduction, isomerization, and conjugation reactions (Wise and Croteau 1999). Cytochrome P450 enzymes are often involved in the post-cyclization modification of terpenoid compounds (Ro et al. 2005; Wise and Croteau 1999). Evidence presented by Wise and Croteau (1999) suggested that the cytochrome P450 enzymes, dehydrogenases, and reductases responsible for the modification of terpenoid compounds in plants are highly substrate- specific, although multi-substrate and multi-function cytochrome P450s for terpenoid oxidation are known (Ro et al. 2005). The focus of this project is on the monoterpene and diterpenoid products in Sitka spruce.   1.8 The molecular-genetic basis of conifer terpenoid defenses Terpene synthases have already been cloned and characterized for a number of coniferous species, and thus the method for sequence identification, cloning, and functional   12 characterization of terpene synthases has been previously established in conifers including Sitka spruce (Keeling and Bohlmann 2006).  Several sequence characteristics occur in terpene synthases that link them together as a common group. In general, terpene synthases contain a conserved DDxxD (D=aspartic acid) amino acid sequence involved in substrate binding (Croteau et al. 2000) and a conserved RR-motif (RRxxxxxxxxW) of unknown function. Terpene synthases are often cyclases, creating ring structures that ultimately form a variety of different terpene compounds recognizable in conifer tissues. In addition, terpene synthases are often not limited to a single product; many of the terpene synthases produce several different products (Steele et al. 1998; Davis and Croteau 2000).  Monoterpene and diterpene synthases contain N-terminal plastid-targeting peptide sequences (Bohlmann et al. 1998; Keeling and Bohlmann 2006). In monoterpene synthases, the N-terminal plastid-targeting sequence is generally found in the region upstream of the conserved RR-motif; it contains high numbers of serine and threonine residues with low numbers of acidic residues (Bohlmann et al. 1997). Functions of individual terpene synthases must be experimentally tested as gene sequences are highly similar, and therefore gene function cannot be predicted from sequence information alone (Bohlmann et al. 1998; Bohlmann et al. 2004; Keeling et al. 2008).  1.9 Purpose of this thesis The purpose of this project was threefold. First, because terpene profiles could be important in the resistance of Sitka spruce to white pine weevils, I designed a study to identify mono- and diterpene compounds that are important for resistance over a wide range of tree genotypes (Chapter 2). Secondly, I characterized the behavioral, physiological, and   13 reproductive response of weevils to a highly resistant tree genotype in comparison to a highly susceptible genotype (Chapter 3). The final project served as a case study to investigate the factors behind the differential production of (+)-3-carene, a monoterpene, in trees from two extremes of resistance: a highly resistant tree genotype (H898) versus a highly susceptible tree genotype (Q903) (Chapter 4). The results of this project address the mechanisms that control the (+)-3-carene phenotype and begin the characterization of the molecular-genetic basis of host resistance. This thesis was designed to provide information that may ultimately allow for the sustainable re-introduction of a commercially valuable, native conifer species while maintaining a functional equilibrium with the endemic insect herbivore.    14 1.10 References  Alfaro RI. 1989. Stem defects in Sitka spruce induced by Sitka spruce weevil, Pissodes strobi (Peck). Proc. IUFRO Working Group on Insects Affecting Reforestation, Vancouver, B.C. Edited by Alfaro RI and Glover SG. pp. 177–185.  Alfaro RI. 1995. An induced defense reaction in white spruce to attack by the white pine weevil, Pissodes strobi. Can. J. For. Res. 25(10):1725-1730.  Alfaro RI. 1996. Role of genetic resistance in managing ecosystems susceptible to white pine weevil. For. Chron. 72(4):374-380.  Alfaro RI. 1998. White pine weevil, Pissodes strobi: Risk factors, monitoring and management. Technology transfer note, Forestry Research Applications, Pacific Forestry Centre, Number 4, September 1998, Natural Resources Canada, Canadian Forest Service.  Alfaro RI, Borden JH, Fraser RG, and Yanchuk A. 1995. The white pine weevil in British Columbia: Basis for an integrated pest management system. For. Chron. 71(1):66-73.  Alfaro RI, Fangliang H, Kiss G, King J, and Yanchuk A. 1996. Resistance of white spruce to white pine weevil: development of a resistance index. For. Ecol. and Manage. 81:51-62.  Alfaro RI, Hulme M, and Ying C. 1993. Variation in attack by Sitka spruce weevil, Pissodes strobi (Peck), within a resistant provenance of Sitka spruce. J. Entomol. Soc. Brit. Columbia. 90:24-30.   15  Alfaro RI and Omule SAY. 1990. The effect of spacing on Sitka spruce weevil damage to Sitka spruce. Can. J. For. Res. 20(2):179-184.  Alfaro RI, Pierce HD, Borden JH, and Oehlschlager AC. 1980. Role of volatile and nonvolatile components of Sitka spruce bark as feeding stimulants for Pissodes strobi Peck (Coleoptera: Curculionidae). Can. J. Zool. 58:626-632.  Alfaro RI, vanAkker L, Jaquish B, and King J. 2004. Weevil resistance of progeny derived from putatively resistant and susceptible interior spruce parents. For. Ecol. and Manage. 202:369-377.  Alfaro RI and Ying CC. 1990. Levels of Sitka spruce weevil, Pissodes strobi (Peck), damage among Sitka spruce provenances and families near Sayward, British Columbia. Can. Ent. 122:607-615.  Bellocq MI and Smith SM. 1996. Mortality of the white pine weevil associated with silvicultural practices in jack pine plantations. For. Chron. 72:338-392.  Bohlmann J, Meyer-Gauen G, and Croteau R. 1998. Plant terpenoid synthases: Molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. USA. 95:4126-4133.  Bohlmann J, Martin DM, Miller B, and Huber DPW. (2004). Terpenoid synthases in conifers and poplars. In: Plantation Forest Biotechnology for the 21st Century. Edited by Walter C and Carson M.  Kerala, India: Research Signpost. pp.181-201    16 Bohlmann J, Steele CL, and Croteau R. 1997. Monoterpene synthases from Grand Fir (Abies grandis). J. Biol Chem. 272:21784-21792.  Croteau R, Kutchan TM, and Lewis NG. 2000. Natural products (secondary metabolites). In: Biochemistry and Molecular Biology of Plants. Edited by Buchanan BB, Gruissem W, Jones RL. American Society of Plant Physiologists: Rockville MD USA. pp. 1250-1318.  Davis EM and Croteau R. 2000. Cyclization enzymes in the biosynthesis of monoterpenes, sesquiterpenes, and diterpenes. Topics in Curr. Chem. 209:53-95.  Eisenreich W, Schwarz M, Cartayrade A, Arigoni D, Zenk MH, Bacher A. 1998. The deoxyxylulose phosphate pathway of terpenoid biosynthesis in plants and microorganisms. Chem. Biol. 5:R221-R233.  Harman DM and Kulman M. 1967. Flight and dispersal of the white-pine weevil. J. Econ. Entomol. 60(6):1682-1687.  Hulme MA. 1989. Laboratory assessment of predation by Lonchaea cortices (Diptera: Lonchaeidae) on Pissodes strobi (Coleoptera: Curculionidae). Envir. Entomol. 18:1011- 1014.  Hulme MA. 1990. Field assessment of predation by Lonchaea cortices (Diptera: Lonchaeidae) on Pissodes strobi (Coleoptera: Curculionidae) in Picea sitchensis. Environ. Entomol. 19:54-58.    17 Keeling CI and Bohlmann J. 2006. Genes, enzymes, and chemicals of terpenoid diversity in the constitutive and induced defense of conifers against insects and pathogens. New Phytol. 170:657-675.  Keeling CI, Weisshaar S, Lin RPC, and Bohlmann J. 2008. Functional plasticity of paralogous diterpene synthases involved in conifer defense. Proc. Natl. Acad. Sci. USA 105:1085-1090.  King JN and Alfaro RI. 2009. Developing Sitka spruce populations for resistance to the white pine weevil: summary of research and breeding program. B.C. Min. For. Range, For. Sci. Prog., Victoria, B.C. Tech. Rep. 050. www.for.gov.bc.ca/hfd/pubs/Docs/Tr/Tr050.htm.  Leal I, White EE, Sahota TS, and Manville JF. 1997. Differential expression of vitellogenin gene in the spruce terminal weevil feeding on resistant versus susceptible host trees. Insect Biochem. Mol. Biol. 27(6):569-575.  Lewis KG, El-Kassaby YA, Alfaro RI, and Barnes S. 2000. Population genetic structure of Pissodes strobi (Coleoptera: Curculionidae) in British Columbia, Canada. Ann. Entomol. Soc. Am. 93(4):807-818.  Lichtenthaler HK. 1999. The 1-deoxy-D-xylulose 5-phosphate pathway of isoprenoid biosynthesis in plants. Annu. Rev. Plant. Physiol. Mol. Biol. 50:47-65.  McMullen LH. 1976. Effect of temperature on oviposition and brood development of Pissodes strobi (Coleoptera: Curculionidae). Can. Entomol. 108:1167-1172.    18  Mitchell KJ and Grout SE. 1995. User’s guide for producing managed stand yield tables with WinTIPSY, version 1.3 under Microsoft Windows. Research Branch, BC Ministry of Forests, Victoria BC.  NRC Natural Resources Canada 2004. Weevil Management. Available from: http://www.pfc.forestry.ca/entomology/weevil/manage_e.html. Accessed August 23, 2004.  Newman JD and Chappell J. 1999. Isoprenoid biosynthesis in plants: Carbon partitioning within the cytoplasmic pathway. Crit. Rev. Biochem. Mol. Biol. 34:95-106.  O’Neill GA, Aitken SN, King JN, and Alfaro RI. 2001. Geographic variation in resin canal defenses in seedlings from the Sitka spruce x white spruce introgression zone. Can. J. For. Res. 32:390-400.  Overhulser DL and Gara RI. 1981. Site and host factors affecting the Sitka spruce weevil, Pissodes strobi, in western Washington. Envir. Entomol. 10:611-614.  Ro D-K, Arimura G-I, Lau SYW, Piers E, and Bohlmann J. 2005. Loblolly pine abietadienol/abietadienal oxidase PtAO is a multifunctional, multi-substrate cytochrome P450 monooxygenase. Proc. Natl. Acad. Sci. USA 102:8060-8065.  Rohmer M. 1999. The discovery of a mevalonate-independent pathway for isoprenoid biosynthesis in bacteria, algae and higher plants. Nat. Prod. Rep. 16:565-574.    19 Roitberg BD and Isman MB (eds.). 1992. Insect Chemical Ecology: An Evolutionary Approach. Chapman & Hall, New York, 359 pp.  Sahota TS, Manville JF, and White E. 1994. Interaction between Sitka spruce weevil and its host, Picea sitchensis (Bong) Carr.: A new mechanism for resistance. Can. Entomol. 126: 1067-1074.  Schilmiller AL, Schauvinhold I, Larson M, Xu R, Charbonneau AL, Schmidt A, Wilkerson C, Last RL, and Pichersky E. 2009. Monoterpenes in the glandular trichomes of tomato are synthesized from a neryl diphosphate precursor rather than geranyl diphosphate. Proc. Natl. Acad. Sci. 106(26):10865-10870.  Seybold SJ and Tittiger C. 2003. Biochemistry of molecular biology of de novo isoprenoid pheromone production in the Scolytidae. Annu. Rev. Entomol. 48:425-453.  Shrimpton DM. 1973. Extractives associated with wound response of lodgepole pine attacked by the mountain pine beetle and associated microorganisms. Can. J. Bot. 51:527-534.  Steele CL, Crock J, Bohlmann J, and Croteau R. 1998. Sesquiterpene synthases from grand fir (Abies grandis). J. Biol. Chem. 273(4):2078-2089.  Tanaka JA. 1999. Chemical ecology of the white pine weevil in British Columbia [Master's Thesis] Simon Fraser University Dept. of Biological Sciences.    20 Tomlin E and Borden JH. 1994. Development of a multi-component resistance index for Sitka spruce resistant to the white pine weevil. In: The White Pine Weevil: Biology, Damage and Management, Proceedings of a Symposium, January 19-21, Richmond, BC. Edited by R.I. Alfaro, G. Kiss, and R.G. Fraser. Can. For. Serv. Pac. For. Cent. FRDA Rep. 226. pp. 117-133.  VanderSar TJD and Borden JH. 1977a. Aspects of host selection behavior of Pissodes strobe (Coleoptera: Curculionidae) as revealed in laboratory feeding bioassays. Can. J. Zool. 55:405-414.  VanderSar TJD and Borden JH. 1977b. Role of geotaxis and phototaxis in the feeding and oviposition behavior of overwintered Pissodes strobi. Environ. Entomol. 6(5):743-749.  VanderSar TJD and Borden JH, 1977c. Visual orientation of Pissodes strobi Peck (Coleoptera: Curculionidae) in relation to host selection behavior. Can. J. Zool. 55:2042- 2049.  Wise ML and Croteau R. 1999. Monoterpene biosynthesis. In: Comprehensive natural products chemistry: Volume 2, Isoprenoids including carotenoids and steroids. Edited by Barton D, Nakanishi K, Meth-Cohn O, and Cane DE. Elsevier Science Ltd.: Kidlington, Oxford, UK. pp. 97-153.  Ying CC. 1991. Genetic resistance to the white pine weevil in Sitka spruce. BC Ministry of Forests Res. Note 106. Victoria BC Canada 17p.   21 2 TERPENOID METABOLITE PROFILING IN SITKA SPRUCE IDENTIFIES ASSOCIATION OF DEHYDROABIETIC ACID, (+)-3-CARENE AND TERPINOLENE WITH RESISTANCE AGAINST WHITE PINE WEEVIL1   2.1 Introduction White pine weevil (Pissodes strobi) is an insect pest that attacks Sitka spruce (Picea sitchensis Bong.), a commercially and ecologically important tree species in the Pacific Northwest. Currently, resistant tree genotypes are identified for breeding in the BC Ministry of Forests and Range (BCMFR) in replicated seedling and clonal field trials which have been augmented with weevils (Alfaro et al. 2008). Although effective for identifying resistant tree genotypes, this process takes several years for tree propagation and planting, augmenting with weevils, and monitoring attack rates. Methods for predicting resistance could save time, effort, and money in choosing resistant parent genotypes for tree breeding programs, as well as provide insight into the underlying mechanisms of resistance.  Several previous studies have suggested chemical or physiological indicators of spruce resistance to weevil. For example, it has been well established that traumatic resinosis and induced terpene synthase gene expression occurs in response to attack by white pine weevil (Alfaro 1995; Byun McKay et al. 2003; Miller et al. 2005), and that resinosis is a significant cause of weevil death (Overhulser and Gara 1981). Larger and more numerous bark resin ducts have also been consistently associated with resistant spruce provenances (Alfaro et al. 2004). Attempts to associate terpene profiles with resistance have been made in several spruce species. Nault et al. (1999) tested for association of monoterpene   1 A version of this chapter has been submitted for publication. Robert, J., Madilao L.L., White, R., Yanchuk A., King, J. and Bohlmann J. Terpenoid metabolite profiling in Sitka spruce identifies association of dehydroabietic acid, (+)-3-carene, and terpinolene with resistance against white pine weevil.   22 compounds and a number of unknown compounds with spruce resistance using a mixed species group (white spruce and Engelmann spruce) of eight resistant parent trees and eight susceptible parent trees and a random selection of open-pollinated progeny (25 trees from each parent tree family). These authors concluded that although a number of unknown chemicals were highly accurate at predicting resistance in the parent trees, the model was poor at predicting resistance of progeny within families. Tomlin et al. (1996) showed that putatively resistant Sitka spruce trees have higher amounts of diterpene resin acids.  This study builds on a foundation of research conducted by the BCMFR.  Researchers at BCMFR, in collaboration with the Canadian Forest Service (CFS), have initiated a program to screen for resistant tree genotypes throughout the range of Sitka spruce. From these screening efforts, two pockets of highly resistant Sitka spruce populations were identified in the Haney (east of Vancouver in the lower mainland of coastal BC) and Big Qualicum (east Vancouver Island) areas. Replicated clonebanks have been established with these Sitka spruce genotypes from Haney, Big Qualicum, as well as other areas throughout BC and into Washington, that have been repeatedly screened for resistance to white pine weevil (King and Alfaro 2009). I sampled 422 trees representing 111 genotypes from a clonebank located at Cowichan Lake on Vancouver Island for metabolite profiling of oleoresin terpenoids to determine the relationship of mono- and diterpenoid compounds, which typically comprise more than 90% of the resin terpenoids in spruce stems (Martin et al. 2002; Miller et al. 2005; Zulak et al. 2009), with the resistance rating for each genotype.  Based on improved methods for the qualitative and quantitative analysis of a full range of monoterpenes (including stereochemistry) as well as diterpenes, diterpene aldehydes, diterpene alcohols and diterpene acids, I aimed to identify the associations of specific terpene compounds with tree genotypes that have been thoroughly tested for resistance. My results show that a fitted discriminant function distinguishes between resistant and susceptible trees from the   23 Cowichan Lake clonebank. Dehydroabietic acid, a diterpene resin acid, was identified as a strong indicator of Sitka spruce resistance against weevils.  Two monoterpenes, (+)-3- carene and terpinolene were also associated with resistance in genotypes originating from the Haney area, an area which may have had historically high weevil pressure.  2.2 Materials and methods  2.2.1 Clonebank and resistance rating The Cowichan Lake clonebank (central Vancouver Island 48°50’N, 124°06’W) was planted in 1994 in the coastal western hemlock very dry maritime variant 2 biogeoclimatic zone (Medinger an Pojar 1991). The provenance of the majority of the tree genotypes come from the Haney area east of Vancouver (49°14’N : 122°36’W) and the Big Qualicum area on Vancouver Island (49°14’N : 124°37’W). The remaining tree genotypes come from a number of locations along the coast of British Columbia and Washington: Queen Charlotte Islands (QCI), 53°55’N : 132°05’W, West Vancouver Island, (WVI) 49°47’N : 126°20’W, Hoquiam, Washington (HOQ), 47°05’N : 124°03’W, areas which are known to contain susceptible trees, as well as trees from the interior spruce hybrid zone (HYB), 54°05’N : 124°03’W, and Squamish (SQ), 49°53’N : 123°15’W (Figure 1).    24 QCI SQ HYB WVI H BQ HOQ British Columbia Cowichan Lake clonebank Washington, USA 12 5’ W 50’N 100 km 0  Figure 2-1 Locations of origin for tree genotypes planted at Cowichan Lake clonebank. The filled circles indicate locations of origin including the Queen Charlotte Islands (QCI), Sitka spruce and interior spruce hybrid zone (HYB), west Vancouver Island (WVI), Big Qualicum (BQ), Squamish (SQ), Haney (H), and Hoquiam (HOQ). The filled triangle shows the sampling location at the Cowichan Lake clonebank. The grey hatched area indicates the high weevil hazard area. This map was modified from King et al. (2004).  Clonebank trees (ramets) were propagated as replicate grafts with five ramets planted per genotype. The resistance rating for the clonebank genotypes resulted from many screening trials in a number of locations. Details of the screening and evaluation are given in King and Alfaro (2009), but briefly: the number of successful weevil attacks were measured as a percentage mean annual attack rate in screening trials that were augmented by hand   25 with high, uniform levels of adult weevils. Mean annual attack rates were assessed over several years after augmentation while weevil populations remained high. The mean annual attack rates for each genotype were converted to a nominal resistance rating value between zero and fifty. A resistance rating of 0 is equivalent to no observed attacks during the sampling period whereas a resistance rating of 50 is equivalent to the maximum number of attacks possible. The maximum number of attacks for a given tree would be an attack every second year (50% mean annual attack rate) as apical shoot leaders rarely will grow in the year directly following an attack (King and Alfaro 2009). The resistance rating values for the 111 genotypes investigated in this study were grouped into three categories for analysis: resistant (72 genotypes with index values between 0 and 12), intermediate (11 genotypes with index values between 12 and 24) and susceptible (28 genotypes with index values between 24 and 50).  2.2.2 Sample collection for metabolite profiling I collected samples over two consecutive days (May 31 and June 1, 2006).  These dates were chosen to capture metabolite profiles during the time of year that coincides with weevil host selection.  I sampled 111 genotypes with four replicates where possible (i.e., four ramets) per genotype that have previously been rated for resistance as described above. For several genotypes, fewer ramets were available (15 genotypes had 3 ramets, 7 genotypes had 2 ramets, and 2 genotypes had one ramet). Branch sections approximately 10 cm in length were taken from the upper-most lateral branches of the trees. The second internode (2005 growth) was sampled from each branch. Each ramet was sampled in a different cardinal direction (north, south, east, west). The direction was chosen randomly for each ramet and the time of day was noted. The bark and xylem were separated, the needles were removed, and the samples were immediately frozen on dry ice, then stored at -80°C until extraction for metabolite analysis of the bark. It is important to note that I could not   26 sample the apical leaders of the clonebank trees as removal of leaders would affect tree growth. I therefore sampled the upper-most lateral branches after first assessing the impact of sample location within a tree on terpene profiles as described below.  2.2.3 Assessment of branch age and within-tree sample location effects on terpene profiles The most severe weevil damage on regenerating Sitka spruce occurs as a result of oviposition and subsequent larval feeding and development in the phloem, cambium and outer xylem layers of the apical shoot leader. Since destructive sampling of the apical leader would prevent future use of clonebank trees for other work, I tested whether samples from lateral branches of the top whorl (i.e., branches from the same whorl as the apical leader) would accurately represent the terpene profiles produced in the leaders (the site of weevil infestation). I conducted two sets of sample analyses to identify changes in the terpene profile for a subset of compounds with branch age (1 - 3 years old) or location within the tree (leader, top branches, middle branches and lower branches, or cardinal direction north, west, south, east).  The two sets of additional sample collections and bark terpene analyses were conducted in 2005 prior to the metabolite profiling of the Cowichan Lake clonebank to assess the variability of terpene profiles within a given tree. The first set (108 samples) was collected at the Cowichan Lake site in order to determine the effect of branch age (2003 growth versus 2004 growth, versus 2005 growth at the branch tip) on terpene profile. Age was determined by counting of growth rings in thin sections at the widest part of the internode using a compound microscope. The second set of samples was collected at the University of British Columbia South Campus farm to identify possible differences in terpene profiles with cardinal direction and between the 2005 apical shoot leader and 2005 lateral branch   27 samples from the top of the tree, the middle of the tree, and the bottom of 10 to 12 year-old Sitka spruce trees with no evidence of weevil infestation. Samples were taken from four randomly chosen trees.  Thirteen samples were taken from each tree (leader, top each cardinal direction, middle each direction, and bottom each direction).  2.2.4 Terpene extraction and compound identification Each bark sample was divided into three parts for three separate terpene extractions (technical replicates) for each of the biological replicates. The terpene compounds were extracted and analysed by gas chromatography (GC) to determine the amount of monoterpene and diterpenoid compounds (µg terpenoid / g dry weight tissue). The terpene extraction protocol was based on Lewinsohn et al. (1993) and Martin et al. (2002) with the following adaptations. Approximately 0.2 grams dry weight of bark tissue were extracted using 1.5 mL tert-butyl methyl ether (Chromasolv Plus, for HPLC, 99.9% MTBE, Sigma- Aldrich) containing 100 µg/mL isobutyl benzene (Fluka) and 200 µg/mL dichlorodehydroabietic acid (Helix Biotech) as internal standards. Samples were shaken overnight at room temperature. The next day, the extract was separated into two fractions: 0.5 mL was used for monoterpene analysis, and 0.5 mL was used to derivatise (addition of a methyl group to the carboxylic acid functional group of diterpene resin acids) for diterpenoid analysis. The remaining solid bark tissue was removed, air dried for one week in a fumehood, and weighed. The monoterpene fraction was washed with 0.3 mL of aqueous 0.1M (NH4)2CO3 (pH 8.0). The diterpenoid fraction was derivatised using 200 µL methanol (Fisher Scientific) with 200 µL (trimethylsilyl)diazomethane (2.0M in diethyl ether, Sigma- Aldrich) at room temperature for 20 minutes, and evaporated under compressed nitrogen gas (Praxair) to dryness. The dried extract was then re-suspended in 1 mL anhydrous HPLC grade, inhibitor-free ethyl ether (Sigma-Aldrich).    28 Extracted terpenoid compounds were identified using an Agilent 6890A Series GC system (Agilent Technologies, Palo Alto, CA, USA) with an Agilent 7683 series autosampler and a flame ionization detector (FID). Compound identification was achieved through the comparison of compound retention time with the retention time of authentic standards. Identities of the compounds were confirmed using GC (Agilent 6890A series) coupled with mass spectrometry (MS; 5973N mass selective detector, quadropole analyzer, electron ionization, 70eV). The following GC program was used to separate the monoterpene peaks on a SGE Solgel-Wax capillary column (Mandel Scientific SG-054796, 250µm diameter, 30 m length, and 0.25 µm film thickness): the 40°C initial temperature was increased by 3°C min-1 to 110°C, then increased at 10°C min-1 to 180°C, then finally increased by 15°C min-1 to 260°C and held for 15 minutes (total run time is 50.67 minutes). The injector temperature was set at 250°C, and the initial flow rate was 1.5 mL He min-1. Stereochemistry of the monoterpene compounds was determined where authentic standards were available on a Cylcodex-B chiral capillary column (J&W 112-2532, 250 µm diameter, 30 m length, and 0.25 µm film thickness) using the following temperature program: the 55°C initial temperature was increased by 1°C min-1 to 100°C, then increased at 10°C min-1 to 230°C held for 10 minutes (total run time is 69.00 minutes). The initial injection temperature was set at 230°C, and the initial flow rate was 1.0 mL He min-1. Diterpenoid compounds were separated using an AT-1000 column (Alltech A-13783, 250 µm diameter, 30 m length, and 0.25 µm film thickness). The 150°C initial temperature was held for 1 minute then increased by 1.5°C min-1 to 220°C, then increased at 20°C min-1 to 240°C, then held for 15 minutes (total run time is 63.67 minutes). The injector temperature was set at 250°C, and the initial flow rate was 1.2 mL He min-1.    29 2.2.5 Statistical analysis Statistical analyses were conducted using R statistics packages (http://www.r- project.org). Mixed model ANOVA (LME function from the NLME package in R, see Pinheiro et al. 2009) were conducted using tree resistance group (resistant, intermediate and susceptible) and tree origin group (Haney, Big Qualicum, and the remaining locations were grouped together as ‘other’) as fixed factors. I divided the variance into 3 components (among genotypes or clones, among ramets and within ramets) and thus adjusted for repeated measurements as the random factor. The terpene profile data was log transformed in order to better satisfy the assumptions of normality and homogeneity of variance. Mixed model ANOVA analysis was performed for all identified terpenes.  Linear discriminant function analysis was used to create linear functions from the measured terpenes that maximized the separation of resistant, intermediate and susceptible tree groups in the data set. I used the linear discriminant analysis (LDA) function from the MASS package, VR bundle (Venables and Ripley 2002) to assess combinations of terpenes that best fit resistance category (resistant, intermediate, or susceptible). As discriminant function analysis does not allow for mixed model assumptions, I averaged the technical replicates and then averaged the data from the biological replicates (i.e., ramets) to obtain one average value per tree genotype for each terpene compound measured which I used to complete the discriminant function analysis. As any zero or missing values (values below the detection limit of the GC method) are not well tolerated by the discriminant function model, I excluded terpenes from the discriminant function analysis with more than 10 genotypes having values at or below detection levels. The remaining 27 terpenes [12 monoterpenes: (−)-α-pinene, (+)-α-pinene, (−)-β-pinene, (+)-sabinene, (+)-3-carene, (−)-α-phellandrene, myrcene, α-terpinene, (+)-limonene, (−)-β-phellandrene, γ-terpinene, terpinolene; 3 diterpene alcohols: pimaradienol, levopimaradienol, palustradienol; 5 diterpene aldehydes:   30 dehydroabienal, abietadienal, neoabietadienal, levopimaraienal, and palustradienal;  7 diterpene resin acids: dehydroabietic acid, sandaracopimaric acid, isopimaric acid, abietic acid, neoabietic acid, levopimaric acid, and palustric acid] were used to create discriminant functions.  I also reassigned missing values using a transcan function in the Hmisc package in R (Harrell et al. 2009) that imputes missing values using the best linear combination of the remaining variables. Additional discriminant function analyses were done for trees from the Big Qualicum and Haney areas to determine whether the patterns for the whole data set were the same in the largest subsets of the data.  All of the discriminant functions were assessed for effectiveness of predicting the resistance group of new trees (model cross-validation for generalization error). First, for the complete data set (containing the full 111 tree genotypes), I performed “leave-one-out” validation where one tree is randomly removed from the data set, then the discriminant function is re-calculated from the remaining trees and used to predict the resistance category of the removed tree. I compared the predicted resistance rating for each tree when it is removed with the fitted resistance rating for that tree when it is included in the discriminant function. Secondly, I used the discriminant functions produced from the trees originating in the Haney region (the Haney subset) to predict the resistance group for trees originating in Big Qualicum (Big Qualicum subset). I will refer to the actual data as the observed classification (resistant, intermediate, or susceptible), the discriminant function output as the fitted classification, and the leave-one-out analysis as the predicted classification.      31 2.3 Results  2.3.1 Terpene profiles from lateral branches of the top whorl accurately reflect terpene profiles of the apical leader I found that total terpene amount was not significantly different across branch age (F4,103 = 1.413, p =  0.235). None of the samples taken showed any significant differences among cardinal direction for any of the compounds analyzed. None of the terpene compounds showed any significant differences between the leader and the other sampling locations within a given tree. 3-Carene showed a non-significant downward trend; the amount of (+)-3- carene was highest in apical shoot leader with 416.6 µg/g dry weight (average) and decreased slightly to 405.4 µg/g dry weight in lateral branches of the top whorl where samples were taken for the metabolite analysis of clonebank trees.  Levels of (+)-3-carene were 305.4 µg/g dry weight in the mid branches and 274.8 µg/g dry weight in the lower branches. Overall, my results show that samples taken from the lateral branches of the top whorl are good representation of the apical shoot leader terpene profile.  Results from ANOVA are shown in supplementary tables S1 and S2 (Appendix B).  2.3.2 Metabolite profiling of the Cowichan Lake clonebank identifies the diterpene resin acid dehydroabietic acid as an indicator for resistance At the Cowichan Lake clonebank I sampled the top whorl branches of up to four ramets per genotype (i.e. four biological replicates) for 111 genotypes, and extracted (in triplicate) and identified the monoterpene and diterpenoid compounds for each sample. Each genotype was previously tested for resistance and a resistance index established (see materials and methods) (King and Alfaro 2009). I was primarily interested in the differences between the resistant and susceptible tree groups for each terpene compound individually (mixed model ANOVAs), and I analysed the best combination of terpene compounds to   32 separate the resistant and susceptible groups (discriminant function analysis). Results from mixed model ANOVA analyses (using tree location of origin and resistance group as factors) for each individual terpene compound showed only four significant (p-value < 0.05) differences between the resistant group and the susceptible group (Table 1): palustradienol (p = 0.047), dehydroabietadienal (p = 0.041), neoabietic acid (p = 0.049), and dehydroabietic acid (p = 0.001).                     33 Table 2-1 The average extracted compound (µg/g dry weight ± 1SE) for the resistant and susceptible groups in combination with the output of mixed model ANOVAs testing the difference between resistant and susceptible trees for each terpene compound.  Compound Resistant  (µg/g dry wt) avg. ± 1SE Susceptible (µg/g dry wt) avg. ± 1SE βa SE t df p-value m on ot er pe ne s (-)-α-pinene 1263.3 ± 79.1 1265.6 ± 205.7 0.074 0.116 0.639 106 0.524 (+)-α-pinene 290.8 ± 37.3 191.8 ± 53.8 0.282 0.229 1.229 106 0.222 α-thujene 114.3 ± 8.7 90.0 ± 12.0 -0.028 0.110 -0.252 95 0.801 (-)-camphene 8.5 ± 1.5 8.3 ± 2.6 0.117 0.224 0.525 56 0.602 (-)-β-pinene 1672.6 ± 136.6 1465.6 ± 164.5 0.109 0.133 0.816 106 0.416 (+)-sabinene 236.6 ± 20.0 175.7 ± 27.1 0.323 0.195 1.658 106 0.100 (+)-3-carene 1590.2 ± 172.5 1146.0 ± 313.5 0.153 0.514 0.298 100 0.766 (-)-α-phellandrene 59.8 ± 18.1 35.6 ± 4.7 -0.137 0.142 -0.966 105 0.336 myrcene 1079.9 ± 54.8 1036.3 ± 113.6 0.139 0.099 1.402 106 0.164 α-terpinene 35.5 ± 2.6 31.9 ± 4.6 -0.076 0.160 -0.476 101 0.635 (+)-limonene 597.4 ± 105.2 615.3 ± 95.2 -0.182 0.273 -0.668 106 0.505 (-)-β-phellandrene 2412.1 ± 128.6 2674.9 ± 218.2 -0.074 0.114 -0.644 106 0.521 γ-terpinene 82.5 ± 5.4 63.6 ± 11.0 -0.004 0.110 -0.034 96 0.973 terpinolene 333.3 ± 24.1 261.6 ± 41.8 0.208 0.183 1.138 106 0.258 di te rp en es  pimaradiene 24.3 ± 17.1 10.0 ± 6.3 0.441 0.406 1.087 59 0.282 dehydroabietadiene ND ND ND ND ND 0 ND sandaracopimaradiene 0.3 ± 0.08 0.4 ± 0.2 0.054 0.179 0.302 23 0.765 isopimaradiene 5.1 ± 2.3 5.4 ± 4.2 0.128 0.257 0.499 46 0.620 abietadiene 11.2 ± 5.4 13.8 ± 6.2 -0.242 0.197 -1.228 88 0.223 neoabietadiene 0.5 ± 0.2 0.8 ± 0.6 -0.347 0.366 -0.949 4 0.396 levopimaradiene 0.1 ± 0.1 1.3 ± 0.8 -0.234 0.817 -0.286 4 0.789 palustradiene 7.9 ± 2.0 7.4 ± 1.3 0.230 0.127 1.807 82 0.074 di te rp en e al co ho ls  pimaradienol 113.4 ± 51.6 65.1 ± 31.7 -0.065 0.122 -0.532 103 0.596 dehydroabietadienol 11.6 ± 4.7 7.5 ± 2.1 -0.102 0.255 -0.402 54 0.689 sandaracopimaradienol 2.3 ± 0.8 2.1 ± 0.8 -0.101 0.320 -0.316 35 0.754 isopimaradienol 59.8 ± 27.9 46.5 ± 18.2 0.097 0.193 0.502 83 0.617 abietadienol 92.8 ± 26.0 32.2 ± 5.6 0.026 0.123 0.214 99 0.831 neoabietadienol 91.2 ± 42.1 31.7 ± 23.5 -0.074 0.231 -0.322 82 0.748 levopimaradienol 145.9 ± 50.7 104.1 ± 59.6 -0.008 0.092 -0.089 106 0.929 palustradienol 253.3 ± 61.1 94.2 ± 11.9 0.250 0.124 2.008 103 0.047* di te rp en e al de hy de s pimaradienal ND ND ND ND ND 0 ND dehydroabietadienal 79.8 ± 23.8 18.6 ± 2.6 0.240 0.116 2.074 102 0.041* sandaracopimaradienal 84.8 ± 40.0 40.7 ± 28.6 1.547 1.346 1.150 14 0.270 isopimaradienal 180.1 ± 151.4 28.7 ± 7.6 0.065 0.288 0.227 44 0.822 abietadienal 229.7 ± 55.9 125.1 ± 35.8 0.128 0.111 1.162 106 0.248 neoabietadienal 126.5 ± 28.1 97.1 ± 26.3 0.077 0.094 0.815 105 0.417 levopimaradienal 344.6 ± 90.2 176.5 ± 62.9 0.065 0.115 0.561 106 0.576 palustradienal 1787 ± 398.7 958.7 ± 317.6 0.123 0.107 1.156 106 0.250 di te rp en e ac id s pimaric acid ND ND ND ND ND 0 ND dehydroabietic acid 3318.6 ± 1047.9 1374.5 ± 617.6 0.361 0.107 3.378 106 0.001* sandaracopimaric acid 1654.3 ± 401.6 947.8 ± 303.8 0.116 0.074 1.566 106 0.120 isopimaric acid 8536.4 ± 2280.1 4956.5 ± 1503.0 0.073 0.093 0.790 106 0.431 abietic acid 11186.7 ± 2472.1 6226.7 ± 1285.5 0.057 0.095 0.599 106 0.550 neoabietic acid 8735.6 ± 1806.1 5129.9 ± 1477.8 0.182 0.091 1.991 106 0.049* levopimaric acid 11413.7 ± 2705.2 6720.1 ± 2445.2 0.138 0.103 1.337 106 0.184 palustric acid 19922.8 ± 4961.7 12659.4 ± 5158.9 0.202 0.105 1.931 106 0.056 a β = model coefficient, SE = the standard error of β, t = the test statistic, df = degrees of freedom, ND = not detected. Note: * denotes statistically significant result     34 Of these four compounds, only dehydroabietic acid had a strong association (p = 0.001) with resistance after the location of tree origin was taken into account. The remaining ANOVA results (statistics for terpene differences between resistant and intermediate, and between susceptible and intermediate) are given as supplementary tables S3 and S4. As a general result from this analysis, monoterpenes appear to distinguish between regions of tree origin (see supplementary table S5 for mixed model ANOVA data), while diterpenes distinguish between resistant and susceptible groups.  In the discriminant function analysis, 27 terpene compounds contributed to the separation of resistant and susceptible groups (Table 2). The compound showing the largest weight in LD1 (the discriminant function best separating the resistant and susceptible groups) was dehydroabietic acid, the only highly significant terpenoid compound to also occur in the mixed model ANOVAs (see above). When dehydroabietic acid, as well as neoabietic acid, levels are high in resistant trees (designated by a positive weight in Table 2), there is a corresponding drop in two diterpene aldehydes (levopimaradienal and dehydroabietadienal), as well as in (−)-α-phellandrene and abietic acid (indicated by the negative weight for these compounds in Table 2). Sandaracopimaric acid (positive weight) and abietic acid (negative weight) show the largest weights in the discriminant function (LD2) separating the intermediate group from the susceptible group.        35 Table 2-2 Coefficients of the linear discriminants (LD1 and LD2) for the 27 compounds used in the discriminant function analysis. Compound LD1 Compound LD2 levopimaradienal -1.50058986 abietic acid -2.08278202 dehydroabietadienal -1.21983163 (-)-β-phellandrene -1.3988539 (-)-α-phellandrene -1.12200972 α-terpinene -1.00848466 abietic acid -1.07446742 neoabietic acid -0.9240242 sandaracopimaric acid -0.93987298 levopimaradienal -0.76195922 palustradienal -0.60630648 pimaradienol -0.75029968 terpinolene -0.20540128 dehydroabietic acid -0.55000289 levopimaradienol -0.12538966 palustric acid -0.534147 (-)-β-phellandrene -0.07937527 palustradienol -0.48978165 (+)-limonene -0.06226044 myrcene -0.37794516 levopimaric acid 0.03391461 neoabietadienal -0.32760077 (-)-β-pinene 0.08457227 levopimaradienol -0.32462968 (+)-3-carene 0.08854134 (+)-α-pinene -0.30556736 isopimaric acid 0.10548601 (+)-3-carene -0.28888068 abietadienal 0.11082087 (+)-limonene -0.26986608 (+)-sabinene 0.12764798 (-)-β-pinene -0.07741131 palustradienol 0.15081359 dehydroabietadienal -0.05619622 palustric acid 0.30426414 γ-terpinene 0.37517959 γ-terpinene 0.37704501 terpinolene 0.52648974 (-)-α-pinene 0.38843515 (-)-α-phellandrene 0.75964186 pimaradienol 0.42143555 abietadienal 0.77471572 myrcene 0.43302498 (+)-sabinene 0.87399357 (+)-α-pinene 0.51111143 isopimaric acid 0.94057216 α-terpinene 0.63015485 (-)-α-pinene 1.06793438 neoabietadienal 0.76487241 levopimaric acid 1.41265284 neoabietic acid 1.73166261 palustradienal 1.86713042 dehydroabietic acid 2.04592391 sandaracopimaric acid 2.05091082 Note: The magnitude of the LD value represents the weight of that compound in the creation of the discriminant function. A minus sign in front of the coefficient indicates how it is related to the rest of the terpenes for that discriminant function. For example, when dehydroabietic acid levels are high, low levels of levopimaradienal are associated with the separation of resistant and susceptible groups.     36 2.3.3 Accuracy of prediction of resistance groups based on terpenoid metabolites The fit of the discriminant function model was accurate for the resistant group with 66 of 72 resistant genotypes correctly classified (91.6% correct) (Figure 2-2 and Table 3).  Susceptible Resistant Intermediate Big Qualicum Haney Other  Figure 2-2 Discriminant function plot. Plot of the discriminant functions (LD1 and LD2) for actual data designated by the filled symbols and using blue for susceptible, red for intermediate, and green for resistant tree genotypes. The category predicted by the model is shown by the colour of the outer circle around each data point. The discriminant functions separated the three resistance categories with LD1 showing the best separation between resistant and susceptible genotypes, and LD2 showing the best separation of the intermediate genotypes.  The fit for the intermediate group (63.6% correct) and the susceptible group (57.1% correct) was less accurate (Table 3).   37 Table 2-3 The observed resistance group (obs.) compared to the predicted resistance group (pred.) generated by the discriminant function model created from the entire data set as compared to the “leave-one-out” cross validation. The percentage of trees correctly identified (% corr.) using each model is also shown.  Susceptible Intermediate Resistant  Obs. Pred. % corr Obs. Pred. % corr Obs. Pred. % corr Full model  28 16 57.1 11 7 63.6 72 66 91.6 “leave-one-out” cross validation 28 11 39.3 11 0 0 72 50 69.4    LD1 showed the best separation between resistant and susceptible groups and LD2 separates the susceptible trees from those in the intermediate group. Although the clonebank includes a mixture of genotypes from areas of high weevil hazard (the Haney region and Big Qualicum region) and other regions in British Columbia mainland, Vancouver Island, the north-western United States as well as from areas with no weevil presence (the Queen Charlotte Islands), my results show that discriminant functions created from the monoterpene and diterpenoid profiles can distinguish between resistant, intermediate and susceptible trees for the set of genotypes represented in the Cowichan lake clonebank. In order to test the generalization error (tests the accuracy of the model when used to predict the resistance group of new genotypes), I predicted the resistance group for each tree after removing it from the data set. Table 3 compares the fitted classifications made by the entire model for each tree versus the predicted classification made for each tree from the remaining data set. The reduction in the number of correctly predicted trees in each resistance group in the cross-validation shows that the discriminant functions will lose accuracy when generalized to new tree genotypes.    38 2.3.4 Identification of the monoterpenes (+)-3-carene and terpinolene, along with dehydroabietic acid, as indicator for resistance in genotypes from the high weevil hazard Haney area For a further analysis I focused on trees of two large subsets represented in the Cowichan Lake clonebank, genotypes originating from the Haney and Big Qualicum areas. I analysed these subsets separately using both the discriminant function analysis and the mixed model ANOVA in order to determine whether terpenes that distinguish resistant genotypes in the entire data set were different from those in the subsets.  Results from the discriminant function fitted to genotypes from the Haney area showed almost perfect separation between resistant, intermediate and susceptible groups. The model fitted 32 of 33 resistant trees correctly (96.9% correct), 7 of 8 intermediate trees correctly (87.5% correct), and 6 of 6 susceptible trees correctly (100% correct). The coefficients of the discriminant function are listed in supplementary Table S6. The predicted classification after leave-one-out cross validation was less accurate, predicting 23 of 33 resistant trees correctly (69.7% correct), 1 of 8 intermediate trees correctly (12.5% correct), and 2 of 5 susceptible trees correctly (40.0% correct) suggesting that the model is over-fit at this sample size. The ANOVA results show two monoterpene compounds, (+)-3-carene (p = 0.024) and terpinolene (p = 0.031), that are significantly different between the resistant and the susceptible groups in the Haney region (Table 4). Again, dehydroabietic acid was strongly associated with the resistant group (p < 0.001).      39 Table 2-4 Output of mixed model ANOVAs testing the difference between resistant and susceptible trees for each terpene compound for the Haney region only.  Compound βa SE t df p-value m on ot er pe ne s (-)-α-pinene -0.199 0.204 -0.979 44 0.333 (+)-α-pinene 0.433 0.409 1.057 44 0.296 α-thujene 0.060 0.228 0.262 41 0.795 (-)-camphene 0.250 0.311 0.803 25 0.429 (-)-β-pinene 0.233 0.255 0.916 44 0.364 (+)-sabinene 0.585 0.302 1.936 44 0.059 (+)-3-carene 1.855 0.795 2.334 43 0.024* (-)-α-phellandrene 0.058 0.220 0.263 44 0.794 myrcene 0.124 0.172 0.719 44 0.476 α-terpinene 0.352 0.237 1.485 43 0.145 (+)-limonene -0.581 0.464 -1.252 44 0.217 (-)-β-phellandrene -0.120 0.200 -0.601 44 0.551 γ-terpinene -0.033 0.155 -0.215 40 0.831 terpinolene 0.611 0.275 2.223 44 0.031*  di te rp en es  pimaradiene 0.610 0.991 0.616 20 0.545 dehydroabietadiene ND ND ND 0 ND sandaracopimaradiene -0.136 0.194 -0.701 9 0.501 isopimaradiene 0.407 0.509 0.799 17 0.435 abietadiene -0.027 0.440 -0.061 36 0.952 neoabietadiene -0.347 0.366 -0.949 4 0.396 levopimaradiene ND ND ND 0 ND palustradiene 0.094 0.284 0.330 34 0.744  di te rp en e  al co ho ls  pimaradienol 0.403 0.358 1.127 42 0.266 dehydroabietadienol -0.470 0.394 -1.192 23 0.246 sandaracopimaradienol -0.276 0.613 -0.450 14 0.660 isopimaradienol 0.715 0.449 1.590 35 0.121 abietadienol -0.148 0.267 -0.556 42 0.581 neoabietadienol -0.771 0.571 -1.350 36 0.185 levopimaradienol 0.049 0.191 0.259 44 0.797 palustradienol 0.168 0.239 0.701 43 0.487  di te rp en e al de hy de s pimaradienal ND ND ND 0 ND dehydroabietadienal 0.134 0.229 0.587 43 0.560 sandaracopimaradienal ND ND ND 0 ND isopimaradienal 0.653 0.588 1.110 16 0.283 abietadienal 0.219 0.299 0.732 44 0.468 neoabietadienal -0.140 0.180 -0.779 43 0.441 levopimaradienal 0.342 0.324 1.056 44 0.297 palustradienal 0.096 0.224 0.430 44 0.669  di te rp en e  ac id s pimaric acid ND ND ND 0 ND dehydroabietic acid 0.707 0.188 3.764 44 0.000* sandaracopimaric acid 0.222 0.138 1.615 44 0.114 isopimaric acid 0.258 0.193 1.336 44 0.189 abietic acid 0.084 0.192 0.439 44 0.663 neoabietic acid 0.009 0.164 0.052 44 0.958 levopimaric acid 0.300 0.170 1.764 44 0.085 palustric acid 0.319 0.193 1.653 44 0.105 a β = model coefficient, SE = the standard error of β, t = the test statistic, df = degrees of freedom, ND = not detected. Note: * denotes statistically significant result    40  Table 5 shows the actual amounts for dehydroabietic acid, (+)-3-carene and terpinolene for the resistant and susceptible genotypes from the Haney area.  Table 2-5 The average amount ± 1SE (µg /g dry weight) of terpenoid compounds associated with resistance for trees originating in Haney, British Columbia.  Resistant trees Susceptible trees dehydroabietic acid 2008.2 ± 683.7 661.9 ± 111.1 (+)-3-carene 1957.9 ± 245.7 874.2 ± 389.2 terpinolene 387.5 ± 30.5 235.6 ± 59.5   Genotypes from the Big Qualicum area were fitted completely using one discriminant function as there were no intermediate trees in the Big Qualicum subset. The model fit for the resistant versus susceptible trees was 100% correct. Again, the predicted classification after leave-one-out cross validation was less accurate predicting 16 of 28 resistant trees correctly (57.1% correct), 0 of 0 intermediate trees correctly (100.0% correct), and 3 of 7 susceptible trees correctly (42.9% correct) suggesting that the model is over-fit for this sample size.  The coefficients of the discriminant functions are listed in supplementary Table S7. There were no terpenoid compounds that showed a significant difference between resistance groups for the Big Qualicum subset (see supplementary table S8).  In order to further assess the generalization error, I used the discriminant functions created from the Haney subset to predict the tree resistance in the trees from the Big Qualicum subset. Prediction of the Big Qualicum data set using the Haney discriminant functions was much less accurate; the model classified 20 of 28 resistant trees correctly   41 (71.4% correct) but predicted 0 of 7 of susceptible trees correctly (0% correct) and predicted 4 of 0 intermediate trees. Thus, although the prediction accuracy of the resistant trees was better than random, predictors identified for the Haney subset could not be transferred as predictors for resistance or susceptibility for trees originating in the Big Qualicum area.  2.4 Discussion I used a metabolite profiling approach that involves high resolution analysis, including stereochemistry, of monoterpenoids and diterpenoids to identify possible predictors for resistance of Sitka spruce against its major insect pest, the white pine weevil.  With the genotypes represented in the clonebank, three compounds in particular stood out in the analysis: dehydroabietic acid is strongly associated with resistance regardless of tree origin, and (+)-3-carene and terpinolene are associated with resistance in genotypes from the Haney region, an area with highly resistant tree genotypes.  The monoterpene profiles extracted matched previous literature for composition and amount. Nault and Alfaro (2001) also extracted α-pinene, sabinene, β-pinene, myrcene, 3- carene, limonene and β-phellandrene as well as three more unknown terpene compounds. I identified small amounts of α-thujene, (−)-camphene, α-terpinene, γ-terpinene, and terpinolene in addition to the stereochemistry of previously identified monoterpenes. Only α- pinene occurred as a mix of stereoisomers in the bark extract although (−)-α-pinene was the dominant stereoisomer, with (+)-α-pinene comprising approximately 25% of the total α- pinene identified. Tomlin et al. (1996) identified all seven diterpene resin acids. I did not find evidence of pimaric acid in the samples taken from Cowichan Lake.    42 2.4.1 Dehydroabietic acid The diterpene resin acid dehydroabietic acid was strongly associated with resistance in this study regardless of tree origin. Diterpene resin acids are major components in the defensive oleoresin secretion of Sitka spruce and other spruce species (Martin et al. 2002; Miller et al. 2005; Zulak et al. 2009). The effects of diterpene resin acids have been studied in the conifer response to attack by insects (for review, Keeling and Bohlmann 2006a), and Wagner et al. (1983) demonstrated reduced feeding and growth for larch sawflys, Pristiphora erichsonii Hartig (Hymenoptera: Tenthredinidae) exposed to increasing doses of dehydroabietic acid, abietic acid, neoabietic acid, isopimaric acid, and sandaracopimaric acid. Wagner et al. (1983) show that although consumption of tissue spiked with dehydroabietic acid does not drop significantly, the efficiency of the conversion of ingested material to body substance and the insect growth rate was significantly reduced. Tomlin et al. (1996) also showed previously that putatively resistant Sitka spruce trees have higher amounts of diterpene resin acids.  2.4.2  (+)-3-Carene and terpinolene In addition, two monoterpene compounds, (+)-3-carene and terpinolene, were significantly associated with resistance in genotypes originating from the Haney region. Provenance trials described in King and Alfaro (2009), suggest that pockets of resistance occur in Sitka spruce. Our results suggest that resistant trees in the Haney region may have a unique terpene profile that contributes to resistance. Several studies show evidence that (+)-3-carene, a monoterpene common in conifer oleoresin, may be an important semiochemical for both white pine weevil and other species.  For example, Storer and Speight (1996) suggest that reduced Dendroctonus micans larval dry weight and survival is associated with increased α-pinene, β-pinene and 3-carene content in Norway spruce bark. Rocchini et al. (2000) suggest that 3-carene may be repellent to ovipositing females or may   43 reduce larval development in lodgepole pine trees that are resistant to attack by pitch moth, Synanthedon novaroensis. Bichao et al. (2003) show that monoterpenes are an important component of compounds sensed by a European weevil, Pissodes notatus and that Pissodes notatus has abundant numbers of olfactory neurons that respond selectively to 3- carene. Toxicity of 3-carene has not been conclusively demonstrated but a few studies address its toxic effects. Lastbom et al. (2003) report that 3-carene is a lung irritant in guinea pigs and potentially in human beings. 3-Carene is also used as the starting material for the artificial synthesis of potent insecticidal compounds (Dhillon et al. 1991). Terpinolene has also been implicated in plant resistance to insect attack.  For example, Salom et al. (1994) suggest that terpinolene acts as an anti-feedant for the pales weevil, Hylobius pales (Herbst). Park et al. (2003) demonstrate significantly increased mortality in two species of adult rice weevil (Callosobruchus chinensis and Sitophilus oryzae) after exposure to volatile terpinolene. Both terpinolene and (+)-3-carene showed repellency and toxicity to adult flour beetles, Tribolium castaneum, and maize weevils, Sitophilus zeamais (Wang et al. 2009). The toxicity of volatile monoterpenes is attributed to the infiltration of these compounds into the insect respiratory system. It is possible that these monoterpenes act as fumigant toxins, volatile deterrents, gustatory toxins, or gustatory deterrents in addition to the toxic effects of dehydroabietic acid.  2.4.3 Genomic and molecular underpinnings of terpene profiles Knowledge of the genomic and molecular foundation for diverse and dynamic terpenoid chemical profiles in conifers has been much advanced in recent years (Keeling and Bohlmann 2006b).  Two large conifer-specific gene families encoding the terpene synthases (TPS) of the TPSd  group (Martin et al. 2004) and cytochromes P450 dependent monooxygenases of the CYP720B family (Ro et al. 2005; Hamberger and Bohlmann 2006) are the major contributors to the formation of a diverse set of terpenoid chemicals in conifer   44 defense. At the biochemical level, many of the spruce TPS enzymes form multiple products from a single substrate. For example, a (+)-3-carene synthase from Norway spruce produces several different cyclic and acyclic monoterpenes with (+)-3-carene (78%) and the mechanistically related terpinolene (11%) the two most abundant products of this enzyme (Fäldt et al. 2003). In white spruce, a genomic clone for a (+)-3-carene synthase gene (Hamberger et al. 2009) has been identified and cDNAs for a small family of Sitka spruce (+)-3-carene synthase like genes have been cloned, all with (+)-3-carene as a major product, and terpinolene as a secondary product. In our study, (+)-3-carene and terpinolene are associated with resistance only in the Haney region. In general, at the transcriptome and proteome levels, differential gene and protein expression of TPS family members contribute to the plasticity and dynamics of terpenoid profiles in spruce (Miller et al. 2005; Zulak et al. 2009).  Similarly, conifer diterpene synthases can form multiple products (Keeling et al. 2008), however a diterpene synthase that forms relevant amounts of dehydroabietadiene, the precursor for dehydroabietic acid, has not yet been cloned at the cDNA or genomic level. The conversion of dehydroabietadiene to dehydroabietic acid via dehydroabietadienol and dehydroabietadienal involves a multifunctional CYP720B4 gene which has recently been cloned from Sitka spruce (B. Hamberger, T. Ohnishi, J. Bohlmann, unpublished results).  2.4.4 Applications for forest management and tree breeding Kersten et al. (2006) recently devised a method to identify underivatised resin acids from conifer tissue rapidly and accurately using a simple methanol extraction procedure coupled with high-performance liquid chromatography (HPLC). Using GC-MS or GC-FID of derivatised samples or using HPLC for analysis of underivatised samples, it may be possible to use dehydroabietic acid levels in screening for resistance, for parent tree selection, or for   45 screening of offspring produced in Sitka spruce breeding programs.  In addition, rapid GCMS analysis for (+)-3-carene and terpinolene could be used for breeding materials from the Haney area.  In future work, creating discriminant functions from an even larger data set (including other clonebanks for example) may decrease the generalization error in using terpene profiles an indicator of resistance. My analysis was conducted using a data set comprised of almost one third of genotypes of widely varying locations of origin (from the Queen Charlotte Islands off of Canada’s north-west coast to Squamish, Vancouver Island, and Hoquiam in the northern United States), and yet the fitted resistance was still almost 92% accurate, and the predicted resistance was almost 70% accurate.  Predicting the terpene profiles of offspring in breeding programs is currently difficult due to the high diversity of the TPSd (Martin et al. 2004) and CYP720B (Hamberger and Bohlmann 2006) gene families in conifers in combination with multiple potential levels of regulation for the expression of genes and the function of enzymes as discussed above. Relevant new gene sequences, such as Sitka spruce (+)-3-carene synthase or Sitka spruce CYP720B4, can be used in future studies to investigate the possible association of gene expression with resistance phenotypes.  2.5 Conclusions The results of this study show that Sitka spruce genotypes in the Cowichan lake clonebank that were identified as resistant, intermediate, or susceptible were distinguishable using the monoterpene and diterpenoid profile. Dehydroabietic acid was strongly associated with tree resistance in the Cowichan Lake clonebank. Investigating any toxicity of this compound, and the other diterpene resin acids, for weevil larvae and adults would enrich   46 our understanding of terpenes as defense compounds in Sitka spruce. In addition,(+)-3- carene and terpinolene were associated with resistant trees originating from the Haney area in B.C. The resistant tree genotypes from Haney may possess or express these genes or alleles that allow for increased production of these compounds. Whether this is responsible for some of the resistant phenotype is a question for future study. The terpene profiles can be interpreted using the molecular-genetic information generated from the functional characterization of the genes involved in the production of the spruce terpene profile. This information will ultimately be applied in forest management practices and tree breeding programs.    47 2.6 References  Alfaro RI. 1995. An induced defense reaction in white spruce to attack by the white pine weevil, Pissodes strobi. Can. J. For. Res. 25(10):1725-1730.  Alfaro RI, King JN, Brown RG, and Buddingh SM. 2008. Screening of Sitka spruce genotypes for resistance to the white pine weevil using artificial infestations. For. Ecol. Manage. 255(5-6):1749-1758.  Alfaro RI, vanAkker L, Jaquish B, and King J. 2004. Weevil resistance of progeny derived from putatively resistant and susceptible interior spruce parents. For. Ecol. Manage. 202(1-3):369-377.  Bichao H, Borg-Karlson AK, Araujo J, and Mustaparta H. 2003. Identification of plant odours activating receptor neurons in the weevil, Pissodes notatus. (Coleoptera, Curculionidae) J.Comp. Physiol. A 189:203-212.  Byun McKay A, Hunter W, Goddard K, Wang S, Martin D, Bohlmann J, and Plant A. 2003. Insect attack and wounding induce traumatic resin duct development and gene expression of (−)-pinene synthase in Sitka spruce. Plant Physiol. 133:368-378  Dhillon RS, Gautam VK, Singh S, and Singh J. 1991. From 3-carene to the most biologically active insecticides – a convenient and efficient synthesis of (+)-(1R)-cis-chrysanthemic acid. Indian Journal of Chemistry Section B – Organic Chemistry including Medicinal Chemistry 30:574-578.    48 Fäldt JF, Martin D, Miller B, Rawat S, and Bohlmann J. 2003. Traumatic resin defense in Norway spruce (Picea abies): Methyl jasmonate-induced terpene synthase gene expression, and cDNA cloning and functional characterization of (+)-3-carene synthase. Plant Mol. Biol. 51(1):119-133.  Hamberger B and Bohlmann J. 2006. Cytochrome P450 mono-oxygenases in conifer genomes: discovery of members of the terpenoid oxygenase superfamily in spruce and pine. Biochem. Soc. Trans. 34(6):1209-1214.  Hamberger B, Hall D, Yuen M, Oddy C, Hamberger B, Keeling CI, Ritland C, Ritland K, and Bohlmann J. 2009. Targeted isolation, sequence assembly and characterization of two white spruce (Picea glauca) BAC clones for terpenoid synthase and cytochrome P450 genes involved in conifer defense reveal insights into a conifer genome. BMC Plant Biol. 9:106.  Harrell FE, and with contributions from many other users. 2009. Hmisc: Harrell Miscellaneous. R package version 3.6-1. http://CRAN.R-project.org/package=Hmisc  Keeling CI and Bohlmann J. 2006a. Diterpene resin acids in conifers. Phytochem. 67:2415- 2423.  Keeling CI and Bohlmann J. 2006b. Genes, enzymes, and chemicals of terpenoid diversity in the constitutive and induced defense of conifers against insects and pathogens. New Phytol. 170:657-675.    49 Keeling CI, Weisshaar S, Lin RPC, and Bohlmann J. 2008. Functional plasticity of paralogous diterpene synthases involved in conifer defense. Proc. Natl. Acad. Sci. USA 105:1085-1090.  Kersten PJ, Kopper BJ, Raffa KF, and Illman BL. 2006. Rapid analysis of abietanes in conifers. J. Chem. Ecol. 32:2679-2685.  King JN and Alfaro RI. 2009. Developing Sitka spruce populations for resistance to the white pine weevil: summary of research and breeding program. B.C. Min. For. Range, For. Sci. Prog., Victoria, B.C. Tech. Rep. 050. www.for.gov.bc.ca/hfd/pubs/Docs/Tr/Tr050.htm.  King JN, Alfaro RI, and Cartwright C. 2004. Genetic resistance of Sitka spruce (Picea sitchensis) populations to the white pine weevil (Pissodes strobi): distribution of resistance. Forestry 77:269-278  Lastbom L, Boman A, Johnsson S, Camner P, and Ryrfeldt A. 2003. Increased airway responsiveness of a common fragrance component, 3-carene, after skin sensitization – a study in isolated guinea pig lungs. Toxicol. Letters 145(2):189-196.  Lewinsohn E, Savage T, Gijzen M, and Croteau R. 1993. Simultaneous analysis of monoterpenes and diterpenoids of conifer oleoresin. Phytochem. Anal. 4(5):220-225.  Martin D, Fäldt J, and Bohlmann J. 2004. Functional characterization of nine Norway spruce TPS genes and evolution of gymnosperm terpene synthases of the TPS-d subfamily. Plant Physiol. 135:1908-1927.    50 Martin D, Tholl D, Gershenzon J, and Bohlmann J. 2002. Methyl jasmonate induces traumatic resin ducts, terpenoid resin biosynthesis, and terpenoid accumulation in developing xylem of Norway spruce stems. Plant Physiol. 129:1003-1018.  Meidinger D and Pojar J. 1991. Ecosystems of British Columbia: Special Report No. 6., Crown Publications, Victoria, BC. 330p.  Miller B, Madilao LL, Ralph S, and Bohlmann J. 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 putative octadecanoid pathway transcripts in Sitka spruce. Plant Physiol. 137:369-382.  Nault JR and Alfaro RI. 2001. Changes in cortical and wood terpenes in Sitka spruce in response to wounding. Can. J. For. Res. 31:1561-1568.  Nault JR, Manville JF, and Sahota TS. 1999. Spruce terpenes: expression and weevil resistance. Can. J. For. Res. 29(6):761-767.  Overhulser DL and Gara RI. 1981. Occluded resin canals associated with egg cavities make by shoot infesting Pissodes. For. Sci. 27:297-298.  Park IK, Lee SG, Choi DH, Park JD, and Ahn YJ. 2003. Insecticidal activities of constituents identified in the essential oil from leaves of Chamaecyparis obtuse against Callosobruchus chinensis (L.) and Sitophilus oryzae (L.). J Stored Prod. Res. 39(4):375- 384.    51 Pinheiro J, Bates D, DebRoy S, Sarkar D, and the R Core team. 2009. nlme: Linear and Nonlinear Mixed Effects Models. R package version 3.1-93.  Ro D-K, Arimura G-I, Lau SYW, Piers E, and Bohlmann J. 2005. Loblolly pine abietadienol/abietadienal oxidase PtAO is a multifunctional, multi-substrate cytochrome P450 monooxygenase. Proc. Natl. Acad. Sci. USA 102:8060-8065.  Rocchini LA, Lindgren BS, and Bennett RG. 2000. Effects of resin flow and monoterpene composition on susceptibility of lodgepole pine to attack by the Douglas-fir pitch moth, Synanthedon novaroensis (Lep., Sesiidae). J. Appl. Entomol. 124:87-92.  Salom SM, Carlson JA, Ang BN, Grosman DM, and Day ER. 1994. Laboratory evaluation of biologically based compounds as antifeedants for the pales weevil, Hylobius pales (Herbst)(Coleoptera: Curculionidae). J. Entomol. Sci. 29:407-419.  Storer AJ, and Speight MR. 1996. Relationships between Dendroctonus micans Kug. (Coleoptera: Scolytidae) survival and development and biochemical changes in Norway spruce, Picea abies (L.) Karst., phloem caused by mechanical wounding. J. Chem. Ecol. 22(3):559-573.  Tomlin ES, Borden JH, and Pierce D. 1996. Relationship between cortical resin acids and resistance of Sitka spruce to the white pine weevil. Can. J. Bot. 74:599-606.  Venables WN, and Ripley BD. 2002. Modern Applied Statistics with S. Fourth Edition. Springer, New York. ISBN 0-387-95457-0    52 Wagner MR, Benjamin DM, Clancy KM, and Schuh BA. 1983. Influence of diterpene resin acids on feeding and growth of larch sawfly, Pristiphora erichsonii (Hartig). J. Chem. Ecol. 9(1):119-127.  Wang JL, Li Y, and Lei CL. 2009. Evaluation of monoterpenes for the control of Tribolium castaneum (Herbst) and Sitophilus zeamaise Motschulsky. Nat. Prod. Res. 23(12):1080- 1088.  Zulak KG, Lippert DN, Kuzyk M, Domanski D, Chou T, Borchers CH, and Bohlmann J. 2009. Targeted proteomics using selected reaction monitoring (SRM) reveals the induction of specific terpene synthases in a multi-level study of methyl jasmonate treated Norway spruce (Picea abies). Plant J. 60(6):1015-30.    53 3 BEHAVIORAL, PHYSIOLOGICAL AND REPRODUCTIVE RESPONSE OF WHITE PINE WEEVIL TO RESISTANT AND SUSCEPTIBLE GENOTYPES OF SITKA SPRUCE2   3.1 Introduction Sitka spruce (Picea sitchensis), a commercially important conifer growing on British Columbia’s west coast, is particularly susceptible to attack from white pine weevil (Pissodes strobi), an insect pest that occurs throughout Canada. The lifecycle of the weevil on a Sitka spruce host involves, in brief, the following events: Adult weevils emerge in the early spring from the duff of the forest floor in search for the apical shoot leader of young host trees where they feed and mate.  In May and early June, female weevils lay their eggs through oviposition holes into the bark of the apical shoot. The larvae hatch and consume or damage the cortex and phloem as well as the cambium zone and outer xylem layers, eventually killing the infested apical stem section and girdling water and nutrient transport to the developing shoot tip. Stem deformation occurs when lateral branches below the attacked leader compete for apical dominance. Weevil attack causes growth losses, a decrease in lumber recovery, and potentially tree and stand death (Alfaro 1989). Because of its extreme susceptibility to attack by white pine weevil, and although ecologically and economically highly valuable, Sitka spruce is rarely planted in large numbers in its natural range of distribution in the Pacific Northwest.  Pockets of natural resistance of Sitka spruce against weevils have been identified in coastal British Columbia (BC), and resistant genotypes have been successfully identified   2 A version of this chapter has been submitted for publication. Robert, J. and Bohlmann J. Behavioral, Physiological and Reproductive Response of White Pine Weevil to Resistant and Susceptible Genotypes of Sitka Spruce.   54 and confirmed in replicated seedling and clonal field trials of the Sitka spruce breeding program of the BC Ministry of Forests and Range (BCMFR) (Alfaro et al. 2008, King and Alfaro 2009). From these trials, the genotype H898 stood out as being almost completely resistant to weevil attack (King and Alfaro 2009).  The H898 genotype originates from the Haney area of the lower mainland of coastal BC east of Vancouver which may have been subject to high weevil pressure.  In this study, I characterize the behavioral, physiological, and reproductive response of weevils (feeding patterns, host choice, ovary development, egg laying behavior, and larval development) to this highly resistant genotype in comparison to a highly susceptible genotype, the Q903 genotype using a series of choice and no-choice experiments.  The Q903 genotype originates from the Queen Charlotte Islands off the coast of BC which are free of any known weevil pressure.  This study required a large number of clonally propagated sapling trees which were available for these two contrasting genotypes, highly resistant H898 and highly susceptible Q903, in form of grafts.  A distinguishing chemical feature of the H898 and Q903 genotypes is the presence (in H898) and absence (in Q903) of the monoterpene (+)-3-carene in the metabolite profiles of these trees.   A recent large- scale metabolite analysis of 111 Sitka spruce genotypes from several geographic locations of the Pacific Northwest showed an association between (+)-3-carene and resistant trees originating in the Haney area (Robert J, White R, Madilao L, Yanchuk A, King J and Bohlmann J; manuscript submitted).  I therefore also tested whether volatile (+)-3-carene may have a role in the weevil host selection between resistant and susceptible trees.  The results of this study suggest that the highly resistant Sitka spruce genotype H898 has defense mechanisms that deter both male and female weevils during host selection and mating, cause delayed ovary development in females, and prevent egg development,   55 hatching, or development of larvae.  Volatile (+)-3-carene does not appear to cause changes in insect behavior prior to feeding.  3.2 Materials and methods  3.2.1 Plant materials and maintenance The BCMFR (Cowichan Lake Research Station, Vancouver Island, BC) supplied three- year-old grafted Sitka spruce clones of the highly resistant H898 genotype originating from the Haney area east of Vancouver (49°14’N : 122°36’W) and the highly susceptible Q903 genotype originating from an area with no history of weevil attack, the Queen Charlotte Islands on BC’s west coast (53°55’N : 132°05’W) (King and Alfaro, 2009).  Trees were maintained in 1 gallon pots outside on the University of British Columbia (UBC) campus. One month prior to experiments, trees were moved into the ambient temperature and light zone of the UBC greenhouse (Miller et al. 2005). In order to best simulate the weevil’s normal host selection, mating and oviposition conditions, the no-choice experiments were conducted in April-May 2007, the choice experiments (with mixed females and males) were conducted in May 2008, and the male only experiment was conducted in May 2009.  3.2.2 Insect rearing and maintenance Weevils were collected by harvesting apical shoot leaders of infested Sitka spruce trees before insect emergence in the fall. Leaders were collected from two BCMFR research plantations in the Campbell River area on Vancouver Island (49°57’N, 125°16’W). Adult weevils were left to emerge in 16L ventilated plastic pails placed in a controlled environment chamber (Conviron chamber 8TC10, 2006) set to 22°C with an 18h:6h light-dark cycle. Every second day emergent adult weevils were removed from the pails and transferred to fresh Sitka spruce cuttings of uncharacterized mixed origin in 2.25L plastic pails. The   56 weevils remained for 5 weeks in the emergence chamber before being transferred to a controlled environment chamber (Conviron) set for a 6°C day and a 4°C night regime. Fresh spruce clippings were added as food source at 10-14 day intervals. The ends of the food branch sections were immersed in heated paraffin wax to reduce water loss. In addition, moistened Whatman filter paper was placed in the bottom of each pail and each pail was misted with water every second day. For all experiments, weevils were separated into males and females according to Harman and Kulman (1966).  3.2.3 Assessment of constitutive and weevil-induced tree defense response I characterized constitutive and weevil-induced defenses for H898 and Q903 when exposed to weevil feeding and oviposition by assessing the constitutive resin canals (resin canals present in the bark tissue), the development of traumatic resin ducts (resin canals produced in the cambium shortly after weevil attack and ultimately embedded in the xylem) in response to weevil feeding, and by characterizing the constitutive and weevil-induced monoterpene profiles for each genotype. Following earlier work on Sitka spruce terpenoid defenses (Alfaro et al. 2002; Byun MacKay et al. 2003; Miller et al. 2005) I measured the number and area of constitutive cortical resin ducts in the phloem tissue (outer stem tissue) and the number and area of weevil-induced, newly formed traumatic resin ducts in the xylem (inner stem tissues) of the apical shoot leader.  I also measured the bark thickness in the cross-sections of tree leaders and measured the quantitative and qualitative monoterpene profiles for each genotype over time of exposure to weevils.  3.2.3.1 Assessment of constitutive and traumatic resin ducts Two adult female weevils and two adult male weevils were caged on the apical shoot leader and highest internode of each tree. I sampled three weevil-attacked trees and three   57 control (unattacked) trees per genotype (H898 and Q903) before weevil exposure and after 22 days of continuous weevil exposure.  A small (1 cm) section from the base of the apical leader was cut into 1 mm thick slices and embedded in resin in three steps over four days: fixation, dehydration and infiltration of resin. Each tissue sample was fixed immediately after harvesting in 1 mL 4% formaldehyde in 50 mM PIPES (sodium salt, minimum 99% titration, SIGMA) buffer (pH 7.2) and incubated overnight in vacuo at room temperature. The next day, fixation was completed by washing twice with 1 mL PIPES buffer for 10 minutes each wash. The tissue was then dehydrated in 1 mL absolute ethanol in nine thirty minute wash intervals beginning with a 10% ethanol in 50mM PIPES buffer (pH 7.2) and working up to 90% ethanol solution. The dehydration step was then completed with two hour-long washes with 1 mL absolute ethanol, and then the samples were left in 1 mL absolute ethanol overnight. The next day, resin infiltration was conducted in thirty minute wash intervals beginning with 1 drop of LR white resin (hard grade acrylic resin, London Resin Company Ltd.) per 1 mL absolute ethanol up to 5 drops resin per milliliter of ethanol. Then, six one-hour washes were conducted beginning with 15% resin (by volume) in ethanol up to 60% resin in ethanol. Samples were left in 60% resin solution over night. The infiltration was continued the next day in three, one-hour washes beginning with 70% resin in ethanol up to 90% resin in ethanol. Two two-hour washes with 100% resin completed the infiltration. The samples were left one final night in 100% resin. The next day, the samples were washed in 100% resin before pouring into molds and baking for overnight at 60°C to solidify the resin in preparation for sectioning.  Each sample was cut into 600 nm thick sections which were viewed using a Zeiss Axioskop 2 MOT compound microscope and photographed using a Zeiss AxioCam HRc camera outfitted with a TV2/3" C>0.63x lens. Two photos, one at 5X magnification and the   58 other at 20X magnification, were taken of each section. The images were analysed using AxioVS40 V 4.6.1.0 (Carl Zeiss Imaging Solutions, 2002-2007). The inner bark thickness (phloem plus primary cortex, excluding periderm) was measured on all samples taken by averaging the thickest part of the inner bark with the thinnest part of the inner bark on each section. The number of traumatic resin ducts produced in response to the weevil attack for the two genotypes was determined from the photos of each cross section sampled after 22 days of weevil feeding. In order to adjust for the varying size of the cross sections, the circumference of the section was measured at the cambium in order to calculate the number of ducts per millimeter of circumference.  3.2.3.2 Analysis of monoterpene profiles Extraction and analysis of monoterpene compounds was conducted for each Sitka spruce genotype, H898 and Q903, at specified timepoints (see below) of continuous weevil exposure. Both the presence and amount of monoterpene (µg / g dry weight tissue) was analysed using gas chromatography (GC) (Agilent 6890A series). The extraction method used is based on Lewinsohn et al. (1993) and conducted as described in Martin et al. (2002) with the following modifications. As in the original protocol, ~0.2 grams dry weight tissue sample were immersed in 1.5 mL tert-butyl methyl ether (Chromasolv Plus, for HPLC, 99.9% MTBE, Sigma-Aldrich), containing 100 µg/mL isobutyl benzene (Fluka) as an internal standard, and shaken overnight at room temperature. The next day, the extract was washed with 0.3 mL of 0.1M (NH4)2CO3 (pH 8.0). The extracted tree tissue was removed, dried at room temperature in the fumehood for one week, and then weighed.  Identification of monoterpene compounds was achieved through the comparison of compound retention time with the retention time of commercially available authentic standards. Identities of the compounds were confirmed using GC coupled with mass   59 spectrometry (MS) (5973N mass selective detector, quadropole analyzer, electron ionization, 70eV).  The following program was used to separate monoterpenes on a SGE Solgel-Wax capillary column (Mandel Scientific SG-054796, 250 µm diameter, 30 m length, and 0.25 µm film thickness): the 40°C initial temperature was increased by 3°C min-1 to 110°C, then increased at 10°C min-1 to 180°C, then finally increased by 15°C min-1 to 260°C held for 15 minutes (total run time is 50.67 minutes). The initial injection temperature was set at 250°C, and the initial flow rate was 1 mL He min-1. Stereochemistry of the compounds was determined where authentic standards were available on a Cylcodex-B capillary column (J&W 112-2532, 250 µm diameter, 30 m length, and 0.25 µm film thickness) using the following temperature program: the 55°C initial temperature was increased by 1°C min-1 to 100°C, then increased at 10°C min-1 to 230°C held for 10 minutes (total run time is 69.00 minutes). The initial injection temperature was set at 230°C, and the initial flow rate was 1 mL He min-1. Response factors were calculated for each compound and the compounds were quantified using a known concentration of internal standard, isobutyl benzene. The amount of compound (µg) per gram dry weight of the tissue extracted was calculated and used as the final value for comparisons.  3.2.4 Assessment of weevil responses to resistant and susceptible tree genotypes and to (+)-3-carene volatiles I conducted a series of choice and no-choice experiments to assess the responses of weevils to resistant and susceptible tree genotypes.  In the no-choice assays, weevils were caged on either individual H898 or Q903 trees. In the choice experiments weevils were free to move between the resistant H898 and susceptible Q903 genotypes. I tracked individual weevil movement over time and assessed the feeding, ovary development, oviposition, and larval development on resistant versus susceptible trees. I also conducted a series of Y-tube   60 bioassays in order to assess the effect of volatile monoterpenes, especially (+)-3-carene, on host attractiveness to adult weevils.  3.2.4.1 Assessment of insect ovary development Ovary development was recorded based on Pernal and Currie (2000) who developed a method for assessment of ovary development for honey bees that was originally derived from Velthus (1970). Assessment of ovary development was based on observations of the following structures and events in the female insect reproductive system. The germanium functions in the formation of the egg; the egg matures and enlarges through the vitellarium and, at maturity, the egg is moved into the lateral oviduct. The two lateral oviducts feed into a common oviduct at the base of the ovaries where the egg is fertilized by sperm stored in the spermatheca (Khan and Musgrave 1969). Information on weevil ovarian physiology was obtained from Perez-Medonza et al. (2004) who studied ovarian development in the rice weevil.  Ovary development was scored in dissected insects using a dissecting microscope (Wild M38, Heerbrugg Switzerland, 40x magnification) as one of three categories: low development, moderate development, or mature. Low development includes undeveloped ovaries (small transparent ovarioles, close to each other, no distinction between germarium and vitellarium) as well as the early initiation of oogenesis (cells beginning to swell at the germarium and follicles moving into the vitellarium); moderate development includes follicles that are still in the vitellarium, they are round or bean-shaped and follicular epithelium is still visible; and mature ovaries were identified by highly developed (mature) sausage-shaped eggs in the lower vitellarium, in the lateral oviduct or the common oviduct. The ovary development category was assigned based on the highest level of development noted   61 during examination of the ovaries as the previous stages are usually visible if the later stages are present.  3.2.4.2 Assessment of feeding and oviposition in no-choice assays This experiment was designed to characterize weevil feeding patterns and oviposition on the resistant H898 versus susceptible Q903 trees when given no choice. For the experiment, weevils were removed from 4°C, placed in petri dishes lined with moistened paper towel, and starved for 48 h prior to placing them on trees. Two adult female weevils and two adult male weevils were caged on the leader and first interwhorl (previous year’s leader growth) of an individual tree. Three trees per genotype (H898 and Q903) were analysed for the number of feeding and oviposition holes after 2, 5, 10, 15 and 22 days of continuous weevil exposure.  3.2.4.3 Assessment of ovary development in no-choice assays  The experimental design above was repeated but with an extended timecourse (sampling after 2, 7, 14, 21 and 28 days of continuous weevil exposure) and a larger number of weevils per tree (3 adult females and 2 adult males per tree) in order to better assess ovary development, eggs and larval development of weevils caged on the resistant versus the susceptible tree genotype. For the weevils recovered at each timepoint, weevil weight (males and females) and ovary development were recorded. At the beginning of the experiment, 15 female weevils were dissected to ensure that no ovary development had occurred prior to the experiment after 48 hours starvation and before feeding on the experimental tree tissue.    62 3.2.4.4 Assessment of feeding and oviposition in choice assays The objective of this experiment was to determine whether adult weevils will occupy, feed, and oviposit preferentially when given a choice between trees of the highly susceptible Q903 genotype and the highly resistant H898 genotype. Six trees (3 resistant H898 and 3 susceptible Q903) were placed in a 1 m width x 1.5 m length x 1.5 m height mesh cage. The trees were placed in two rows with 3 trees in each row. The genotypes were altered systematically. Each tree was covered with a mesh bag cinched at the graft and covering the pot and soil in order to prevent weevils from hiding in the soil or feeding on the rootstock. Trees were watered every 2-3 days during the experiment.  Weevils were removed from 4°C, placed in petri dishes lined with moistened paper towel, and starved for 48 hours prior to placing them on trees. The experiment was conducted in two variations, one with a mix of adult males and females, and one with adult male weevils only.  In the experiment with males and females mixed, 3 female weevils and 2 male weevils were placed on each tree in the mesh cage at the beginning of the experiment (18 female weevils and 12 male weevils in total). Individual female weevils were identified with a color code of two colored spots of oil-based paint (Testor Model Paint) applied to the elytra. The location of each female weevil was recorded at 2-3 day intervals for a total of 23 days (Day 2,5,7,9,12,14,16,19,21,23). The number of males observed on each tree was also recorded at each timepoint. In order to minimize disturbance to the weevils, those that were not easily observable on the tree (or with slow, gentle movement of the branches) were recorded as missing-in-action (MIA) for that sampling day. On final day (day 23) of the timecourse, the number of feeding and oviposition holes was counted on the leader and the first interwhorl below the leader. The leader and the interwhorl tissue were dissected in order to record the number of eggs and larvae beneath the bark.    63 In the experiment with males only, the experimental conditions were identical as described above except that 5 adult male weevils were placed on each tree at the beginning of the experiment. Each male was marked with one spot of yellow paint to increase visibility. Again, the number of males on each genotype was recorded at each timepoint (Day 2,5,7,9,12,14,16,19,21,23) and the number of feeding holes was recorded for each genotype at the end of the experiment on Day 23.  3.2.4.5 Assessment of weevil responses to (+)-3-carene volatiles using Y- tube assays The objective of this experiment was to determine the level of volatile (+)-3-carene that may cause behavior changes in white pine weevil and to assess the role of host volatiles in weevil choice between resistant H898 and susceptible Q903 trees. The weevil response to volatile terpenes was assessed using a Pyrex glass Y-shaped olfactometer and a method modified from Danci et al. (2006). The Y-tube was placed at a 20° incline within a cardboard enclosure containing a Phillips full spectrum natural sunshine fluorescent bulb and a Phillips plant and aquarium fluorescent bulb.  To test whether volatile (+)-3-carene has a role in weevil host selection between resistant and susceptible trees, I conducted a set of Y-tube bioassays with individual monoterpenes in order to determine if, and at which level, these compounds may alter weevil behavior. I also conducted Y-tube bioassays using wounded and unwounded H898 and Q903 and measured the amount of (+)-3-carene released from these trees. I compared the amounts released from trees to the levels of the pure compounds that cause deterrence. Since I used pentane as a solvent for monoterpene dilutions, I first tested the effect of pentane on weevil behavior. As the following assays were conducted with Q903 trees plus   64 pentane on one side (control) versus Q903 trees supplemented with monoterpenes diluted in pentane, it is expected that the solvent should not affect the results of the tests.  A susceptible Q903 tree was placed on either side of the Y in order to ensure that a mixture of authentic host volatiles was present in the background. The treatment, a 10 µL volume  (1S)-(+)-3-carene (99%, Aldrich) at 0.01% (v/v), 0.1% (v/v), 1.0% (v/v), 5.0% (v/v), and 10.0% (v/v) in pentane (spectranalyzed, Fisher Scientific), was applied to a small disc of filter placed randomly at either the right or left side of the tube, and 10 µL of 100% pentane was applied to filter paper on the opposite side. A 5.0% dilution of (1R)-(+)-α-pinene (98%, Aldrich) and (1R)-(−)-limonene (96%, Aldrich) were also assayed. Choice assays were also conducted using H898 versus Q903 trees, and using wounded (a 1mm hole was punched into the bark tissue) H898 and Q903 trees. An air pump with a rubber stopper was attached to the Y-tube in order maintain 0.5 L min-1 air flow through the apparatus.  In each test, insects were placed at the base of the tube and the number of choices toward or away from the treatment was recorded. Each compound and each dilution was tested with 30 male weevils and 30 female weevils that were starved in petri dishes lined with moistened paper towel for 24 hours before the experiment.  In order to assess the level of monoterpene volatiles reaching the test insects from the 5% dilutions (with Q903 background volatiles) and to compare this level to the amount of monoterpene released from wounded and unwounded trees, I collected the monoterpenes from the air flowing through the Y-tube in Poropak Q volatile collection columns. The columns were eluted using 1 mL of tert-butyl methyl ether (Chromasolv Plus, for HPLC, 99.9% MTBE, Sigma-Aldrich) and monoterpene compounds were identified by GCMS (as in section 3.2.3.2).    65 3.2.5 Statistical analyses Tests were conducted using SYSTAT 11.0 (Systat Software Inc. 2004). Where reported, the analyses satisfied the assumptions for both ANOVA and Pearson χ2–squared analysis.   3.3 Results  3.3.1 Assessment of host defense responses in weevil-resistant H898 and - susceptible Q903 Sitka spruce genotypes  The resistant H898 and susceptible Q903 genotypes show major differences in their defense responses at the level of traumatic resin duct formation when exposed to weevils. The H898 genotype showed a significantly larger number of traumatic resin ducts at day 22 of weevil feeding relative to untreated control trees, whereas Q903 showed almost no traumatic resin duct formation (tree genotype: F1, 7 = 102.8, p < 0.001; treatment: F1, 7 = 197.1, p < 0.001). The number and area of constitutive resin ducts were not significantly different between the two tree genotypes (Figure 3-1).    66 N um be r o f d uc ts  / m m  c irc . Ar ea  / du ct Cortical resin ducts Traumatic induced resin ducts 0 0 0 0 H898 Q903 0 1 2 3 0 1 2 3 4 5 0 5000 10000 15000 20000 0 100 200 300 400 500 600 700 B C D weevil-attacked control 4 5 A weevil-attacked control H898 Q903 H898 Q903 H898 Q903 weevil-attacked control weevil-attacked control N um be r o f d uc ts  / m m  c irc . Ar ea  / du ct  Figure 3-1 The number and area of cortical and induced resin ducts for weevil- attacked and control trees resistant versus susceptible tree genotypes. The average number of constitutive cortical resin ducts (A) and induced traumatic resin ducts (B) per millimeter stem circumference (± 1SE) and the average area of cortical resin ducts (C) and induced traumatic resin ducts (D) per duct (± 1SE) for the resistant Sitka spruce genotype (H898) and the susceptible genotype (Q903). Although there are no apparent differences between the resistant and susceptible trees constitutive resin canals, the resistant genotype showed a significantly larger number and area of induced traumatic resin canals after weevil attack.   The leader diameter for H898 was consistently smaller than the diameters measured for the Q903 trees (F1, 82 = 55.7, p < 0.001). However, there was no difference in bark thickness of the leaders of the two genotypes (F1, 69 = 1.3, p = 0.253) (Figure 3-2).   67  H898 Q903 0 1 2 3 4 5 6 7 8 9 Le ad er di am et er  (m m ) H898 Q903 0 0.5 1.0 1.5 2.0 B ar k th ic kn es s (m m ) a b  Figure 3-2 Leader diameter and bark thickness for resistant and susceptible tree genotypes. Panel A shows the leader diameter (mm) and panel B shows the average bark thickness (mm) for Sitka spruce genotypes the resistant genotype (H898) and the susceptible genotype (Q903). Different letters above the bars indicate a significant difference at (p<0.05). Although the susceptible leaders had a larger diameter, there was no significant difference in bark thickness between the resistant and susceptible genotypes.    The monoterpene profiles over 22 days of continuous weevil feeding were variable but not significantly different over time in H898 and Q903 for the compounds I measured, specifically (+)-α-pinene, (−)-limonene, (+)-sabinene and (+)-3-carene (Figure 3-3).   68 H898 Q903 Weevil-attacked Control 0 5 10 15 20 25 Sampling day 0 1000 2000 3000 4000 5000 (+ )- α- pi ne ne  (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 1000 2000 3000 4000 5000 A E (+ )- α- pi ne ne  (µ g/ g dr y w t) (+ )- α- pi ne ne  (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 500 1000 1500 2000 (- )-l im on en e (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 500 1000 1500 2000 B F (- )-l im on en e (µ g/ g dr y w t) (- )-l im on en e (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 50 100 150 200 250 (+ )- sa bi ne ne  (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 50 100 150 200 250 C G (+ )- sa bi ne ne  (µ g/ g dr y w t) (+ )- sa bi ne ne  (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 1000 2000 3000 (+ )- 3- ca re ne  (µ g/ g dr y w t) 0 5 10 15 20 25 Sampling day 0 1000 2000 3000 nd * * D H (+ )- 3- ca re ne  (µ g/ g dr y w t) (+ )- 3- ca re ne  (µ g/ g dr y w t) (+ )- α- pi ne ne  (µ g/ g dr y w t) (+ )- α- pi ne ne  (µ g/ g dr y w t) (+ )- α- pi ne ne  (µ g/ g dr y w t) (- )-l im on en e (µ g/ g dr y w t) (- )-l im on en e (µ g/ g dr y w t) (- )-l im on en e (µ g/ g dr y w t) (+ )- sa bi ne ne  (µ g/ g dr y w t) (+ )- sa bi ne ne  (µ g/ g dr y w t) (+ )- sa bi ne ne  (µ g/ g dr y w t) (+ )- 3- ca re ne  (µ g/ g dr y w t) (+ )- 3- ca re ne  (µ g/ g dr y w t) (+ )- 3- ca re ne  (µ g/ g dr y w t)  Figure 3-3 The resistant and susceptible monoterpene profiles in response to weevil feeding. The average amount of monoterpene measured (µg/g dry weight ± 1SE) on each sampling day for H898 (A-D) and Q903 (E-H). Asterisk indicates a significant difference between treatment and control (p<0.05). nd = not detected. None of the measured monoterpenes changed significantly with weevil feeding over time except for (+)-3-carene which was significantly reduced on day 15 and day 22 in the weevil-feeding treatment. There was no detectable (+)-3-carene in the susceptible genotype Q903.   69  Most of the compounds did not change significantly with continuous weevil feeding in either genotype, but (+)-3-carene decreased significantly at day 15 (F1, 3 = 30.756, p < 0.012) and day 22 (F1, 4 = 35.524, p < 0.004) in the weevil-feeding treatment of H898.  In summary, histological examination of the apical leader cross-sections revealed a strong induced traumatic resin canal response in H898, but similar constitutive resin canal properties and a similar bark thickness. While the monoterpene profiles between the two genotypes differ, they did not change measurably with weevil attack over 22 days. The exception to this trend was (+)-3-carene; this compound was significantly lower in the weevil-treated H898 samples harvested on day 15 and day 22 compared to controls, and was not detectable in the Q903 genotype at any timepoint in control or weevil exposed trees.  3.3.2 Weevil behavior and ovary development affected by resistant and susceptible host trees in no-choice scenarios In order to identify differences in the feeding, oviposition behavior and ovary development, weevils were restricted either to the resistant H898 genotype or the susceptible Q903 genotype and monitored for behavior and ovary development over a period 22 days.  A differential feeding pattern between host genotypes was apparent in particular for the early timepoints of the course of the experiment (Figure 3-4).    70 C um ul at iv e nu m be r o f f ee di ng  h ol es * * Leader Interwhorl A B Da y 0 2 Da y 0 5 Da y 1 0 Da y 1 5 Da y 2 2 0 10 20 30 40 50 60 70 80 90 100 Da y 0 2 Da y 0 5 Da y 1 0 Da y 1 5 Da y 2 2 Sampling day 0 10 20 30 40 50 60 70 80 90 100 Q903 H898 Q903 H898 C um ul at iv e nu m be r o f f ee di ng  h ol es  Figure 3-4 The cumulative number of weevil feeding holes on the leader and interwhorl of the resistant and susceptible tree genotypes. The number of feeding holes observed on H898 and Q903 trees (± 1SE) on the leader (A) and on the interwhorl directly below the leader (B). An asterisk indicates a significant difference between genotypes (p=0.05). The number of feeding holes on the susceptible leader is significantly higher than the number on the resistant leader on day 2, and the number of feeding holes on the resistant internode is significantly higher on day 2.    On day 2, significantly fewer feeding punctures were found on the leader of the resistant H898 trees than were counted on susceptible Q903 trees (F1, 4 = 7.0, p = 0.057). The opposite pattern was observed on the interwhorl directly below the leader tissue (F1, 4 = 17.8, p = 0.013). These differences became indistinguishable by day 10 as feeding accumulated over the timecourse.  There was no significant difference in adult weevil survival between those on H898 versus those on Q903 (χ2 = 0.289, p = 0.865). Although female weevils were significantly heavier than the male weevils (F1, 107 = 15.4, p < 0.001), neither sex showed weight differences depending on the host tree (F1, 107 = 0.185, p = 0.668).   71  Ovary development was also different for weevils restricted to either H898 or Q903 trees. For weevils on the resistant H898 genotype, compared to weevils on the susceptible Q903 genotype, there was a lag in the number of females containing mature eggs up to day 14, followed  by a sharper drop in the number of females with mature eggs after day 14 (Figure 3-5).   Figure 3-5 Weevil ovary development on resistant and susceptible tree genotypes. The percentage weevils with low (gray bars), moderate (white bars), or mature (black bars) ovary development for each sampling day. Weevils were restricted to either H898 host trees (top) or to Q903 host trees (bottom). Weevils feeding on the resistant genotype had fewer mature ovaries by day 7, and fewer mature ovaries on day 21 after peak ovary maturation subsided.   Two days after exposure to host trees, 22% of the females on Q903 and 11% of the females on H898 contained mature eggs. By day seven, 78% of the females caged on Q903   72 contained mature eggs whereas only 25% of the females caged on H898 contained mature eggs. By day 14, 100% of the females caged on Q903 contained mature eggs versus 89% on H898. After 21 days, 62% of the females caged on Q903 still contained mature eggs versus only 29% of females caged on H898.  In summary, when weevils are restricted to either resistant H898 or susceptible Q903 host trees (no choice), I observed an initial avoidance of the resistant leader tissue, which is otherwise a preferred location for weevil activity, and a delay in ovary development.  Without the option of choice, weevils eventually did feed, develop eggs and oviposit on the apical shoot leaders of both genotypes, however.  3.3.3 Weevil behavior and reproductive success affected by resistant and susceptible host trees in choice scenarios Next I tracked individual weevil movement over time in a larger space where weevils were given a choice between resistant H898 and susceptible Q903 trees.  A set of choice experiments were performed in order to assess male and female host selection preferences and to assess oviposition choices and larval development on H898 versus Q903 trees. I monitored the movement of individual weevils three times per week for a total of 23 days. Overall, weevils were highly mobile. On average, a given female weevil was observed on 3.7 different trees. The average number of consecutive sampling days that weevils were found on the same tree was 2.7. No weevils were found on the same tree throughout the complete period of 23 days.  Over the entire timecourse of the choice experiment, a higher number of female weevils were observed on the susceptible Q903 host genotype (Figure 3-6 A). Over 70% of the females had already moved to a susceptible tree by day two and this proportion remained   73 fairly constant until the end of the experiment at day 23.  Movement of male weevils reflected a pattern similar to that of the females for the first half of the timecourse until day 14 (Figure 3-6 B).  However, there was a delay in the number of males showing a preference towards susceptible trees. In contrast to the female weevils, after day 14 the males showed a random distribution on susceptible Q903 and resistant H898 trees. I tested whether the presence of females was influencing male choice of susceptible versus resistant host trees. When males were caged without females (Figure 3-6 C), the distribution was similar to the pattern observed when females were present (Figure 3-6 B).   74 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 Sampling day Pe rc en ta ge  o f m al es  o bs er ve d 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 Sampling day P er ce nt ag e of  m al e w ee vi ls  o bs er ve d Q903 H898 0 10 20 30 40 50 60 70 80 90 100 0 5 10 15 20 25 Sampling day P er ce nt ag e of  fe m al e w ee vi ls  o bs er ve d Q903 H898 Females caged with males Males caged with females Males only A B C Q903 H898 Pe rc en ta ge  o f m al es  o bs er ve d P er ce nt ag e of  m al e w ee vi ls  o bs er ve d P er ce nt ag e of  fe m al e w ee vi ls  o bs er ve d  Figure 3-6 Female and male choice of resistant versus susceptible host tree genotypes. The percentage of female weevils when mixed with male weevils (A), male weevils when mixed with female weevils (B), and male weevils only (C), observed on the susceptible genotype Q903 and on the resistant genotype H898 at each sampling day. The majority of female and male weevils were observed on susceptible trees although males appear to lose this preference at approximately 14 days. The male choice of susceptible trees is independent of the presence of females.    75  Weevil feeding patterns and reproductive success were also different between resistant H898 and susceptible Q903 host trees in the choice tests with female and male weevils combined. Generally, weevils fed less on H898 as assessed by the number of feeding holes (F1, 4 = 7.084, p = 0.056 over the whole tree) (Figure 3-7).  lea der inte rwh orl wh ole  tre e0 100 200 300 400 500 600 700 800 N um be r o f f ee di ng  h ol es Q903 H898 *  Figure 3-7 Cumulative feeding and oviposition punctures for weevils given a choice between resistant and susceptible host tree genotypes. The average number of feeding and oviposition holes (± 1SE) counted on the leader, interwhorl, and over the entire tree for the resistant H898 genotype and the susceptible Q903 genotype over 23 days of continuous weevil exposure when the weevils were given a choice between tree genotypes. When given a choice, weevils feed more on the susceptible trees and on the susceptible leader tissue.   In contrast to successful reproduction, as assessed by development of larvae, on the susceptible Q903 trees after 23 days of weevil exposure, no larvae were found on the resistant H898 trees (Figure 3-8).    76 0 2 4 6 8 10 12 N um be r o f l ar va e lea der inte rwh orl 0 0 * * Q903 H898  Figure 3-8 The number of surviving larvae observed on resistant and susceptible tree hosts for weevils given a choice between resistant and susceptible tree hosts. The number of larvae found in the leader and interwhorl of H898 versus Q903 when the weevils were given a choice between tree genotypes. There were no larvae recovered from the resistant H898 genotype whereas the susceptible leaders averaged nine live surviving larvae on the leader and two surviving larvae in the interwhorl tissue.   In summary, both male and female weevils were able to distinguish between the resistant and susceptible trees and displayed a preference for the susceptible Q903 host genotype.  Both sexes moved frequently, and although feeding holes were present on the resistant trees, weevils were not reproductively successful on resistant H898 trees.  3.3.4 Weevil response to volatile (+)-3-carene In a recent analysis of a large number of Sitka spruce genotypes I found an association between the monoterpene (+)-3-carene and resistance for Sitka spruce trees originating from the Haney area (Chapter 2).  In the case of the H898 and Q903 genotypes, there is a contrasting presence (in H898) and absence (in Q903) of the monoterpene (+)-3-carene in   77 the metabolite profiles of these trees (Figure 3-3 D and H). Females showed a significant attraction to Q903 trees supplemented with pentane when tested against Q903 trees without pentane (χ2 = 4.172, p = 0.041) (Figure 3-9).  As illustrated in Figure 3-9, both female and male weevils showed significant movement away from Q903 trees supplemented with a source of volatile (+)-3-carene at a dilution of 5% (v/v in pentane) (+)-3-carene (females: χ2 = 5.828, p = 0.016; males: χ2 = 5.828, p = 0.016).  10 0%  pe nta ne 0.0 1%  3- ca ren e 0.1 % 3-c are ne 1.0 % 3-c are ne 5%  3- ca ren e 10 % 3-c are ne * ** ns 10 0%  pe nta ne 0.0 1%  3- ca ren e 0.1 % 3-c are ne 1.0 % 3-c are ne 5%  3- ca ren e 10 % 3-c are ne ** ns Females Males Treatment 0 5 10 15 20 25 30 35 N um be r o f w ee vi ls Treatment 0 5 10 15 20 25 30 35 toward away no response A B N um be r o f w ee vi ls N um be r o f w ee vi ls  Figure 3-9 Dilutions of (+)-3-carene causing deterrence in Y-tube bioassays. The number of female weevils (A) and male weevils (B) either moving towards increasing dilutions of (+)-3-carene in pentane versus those that moved away or showed no response. A significant difference (p<0.05) between insect movement toward the treatment versus movement away is designated by an asterisk. Significant numbers of both males and females moved away from a 5% dilution of (+)-3-carene in pentane.   The level of deterrence from 5% (+)-3-carene was very similar to the level of deterrence with 5% (+)-α-pinene (females: χ2 = 3.846, p = 0.050; males: χ2 = 3.000, p = 0.083) and 5% (−)-limonene (females: χ2 = 5.143, p = 0.023; males: χ2 = 13.333, p < 0.001) (Figure 3-10). Neither females (χ2 = 1.286, p = 0.257) nor males (χ2 = 0.040, p = 0.841) showed deterrence   78 behavior against one genotype or the other when given a choice between unwounded or wounded H898 and Q903 trees (Figure 3-10).  * * * ns * * Females Males ns ns 5.0 % 3-c are ne 5.0 % α-p ine ne 5.0 % lim one ne 898  vs .90 3 898  vs .90 3 (wo und ed) 5.0 % 3-c are ne 5.0 % α-p ine ne 5.0 % lim one ne 898  vs .90 3 Treatment 0 5 10 15 20 25 30 35 N um be r o f w ee vi ls Treatment 0 5 10 15 20 25 30 35 toward away no response 898  vs .90 3 (wo und ed) A B N um be r o f w ee vi ls N um be r o f w ee vi ls  Figure 3-10 Comparison of three monoterpenes, the resistant tree genotype and the susceptible tree genotype in Y-tube bioassays. The number of female weevils (A) and male weevil (B) either moving towards or away from a 5.0% dilution of (+)-3-carene compared with 5.0% dilutions of (+)-α-pinene and (-)-limonene, and also compared with tree genotypes H898 and Q903. A significant difference (p<0.05) between insect movement toward the treatment versus movement away is designated by an asterisk. All three monoterpenes showed similar levels of deterrence at the 5% dilution. The weevils did not distinguish between either unwounded or wounded resistant versus susceptible trees in the Y-tube bioassays.   The amount of (+)-3-carene emitted by unwounded trees was undetectable by Poropak Q volatile collection columns and the amount emitted by wounded trees was comparable to volatilization of the 0.1% dilution of (+)-3-carene (Figure 3-11).    79 Treatment 0.00 0.05 0.10 0.15 Am ou nt  o f ( +) -3 -c ar en e (µ g/ L ai r f lo w ) Treatment 0.00005 0.00010 0.00015 0.00020 0.00025 Am ou nt  (+ )-3 -c ar en e (µ g/ L ai r f lo w ) 0.0 1%  3- ca ren e 0.1 % 3-c are ne 1.0 % 3-c are ne 5%  3- ca ren e 10 % 3-c are ne 89 8 v s.9 03 (w ou nd ed ) 0.0 1%  3- ca ren e 0.1 % 3-c are ne 89 8 v s.9 03 89 8 v s.9 03 (w ou nd ed ) A B Am ou nt  o f ( +) -3 -c ar en e (µ g/ L ai r f lo w ) Am ou nt  (+ )-3 -c ar en e (µ g/ L ai r f lo w ) Am ou nt  o f ( +) -3 -c ar en e (µ g/ L ai r f lo w ) Am ou nt  (+ )-3 -c ar en e (µ g/ L ai r f lo w )  Figure 3-11 The amount of volatile (+)-3-carene reaching the insect from each treatment in the Y-tube bioassays. The average amount of (+)-3-carene measured in the Y-tube apparatus (µg/L air flow) from the treatment dilutions (A) and from unwounded and wounded H898 trees compared to known dilutions of (+)-3-carene (B). The volatile (+)-3- carene reaching the insect from an unwounded tree is undetectable by this method, and the amount of volatile (+)-3-carene from a wounded tree is equivalent to the amount of (+)-3- carene volatized from a 0.1% dilution of pure compound.   Thus, although (+)-3-carene is a volatile deterrent for white pine weevils, the level required for deterrent behavior was similar to other monoterpenes, (+)-α-pinene and (−)- limonene, and the levels of (+)-3-carene required to elicit behavior changes was much higher than is emitted from wounded or unwounded trees. Furthermore, both males and females reacted similarly to the volatile terpenes tested.  3.4 Discussion Overall, the results of this work suggest that the highly weevil-resistant Sitka spruce genotype H898 has defense mechanisms that deter both male and female weevils during host selection and mating, that cause delayed ovary development in females, and prevent successful reproduction on H898 trees.  Although the (+)-3-carene has been identified as an   80 indicator of resistance for the H898 genotype and other genotypes from the Haney area, volatile (+)-3-carene as a potential air-borne signal in y-tube tests does not affect host selection behavior when added to the susceptible Q903 genotype.  I measured a number of parameters in order to characterize the constitutive and weevil- induced defense response of the H898 genotype. Histological analysis showed that although leader diameter may have played a role in the weevil’s original choice not to feed on the H898 leader, bark thickness is not significantly different between the two genotypes at any of the timepoints and thus is likely not a factor in the observed feeding patterns. This is in contrast to the suggestion by Manville et al. (2004) that female weevils are primarily using this characteristic to find and oviposit on leader tissue. My data suggest that differential weevil feeding and oviposition between the H898 and Q903 genotypes are caused by variables other than bark thickness. Histological analysis also showed that, in H898, the constitutive tree defense is bolstered by a strong response of induced traumatic resin duct formation. Traumatic resin canals are fully formed 22 days after weevil feeding whereas few induced traumatic resin canals were produced in the Q903 genotype. At the chemical level, the monoterpene profiles generally did not change significantly with continuous weevil feeding over 22 days except that (+)-3-carene was significantly lower in the weevil-attacked trees sampled at day 15 and day 22 in H898. It is possible that this compound is produced constitutively at low levels and then volatilizes from feeding holes at a rate that exceeds de novo biosynthesis.  In contrast, more abundant compounds may be relatively less affected by loss from feeding holes or the loss may be compensated for by differential de novo formation (Miller et al. 2005; Zulak et al. 2009).  The lack of detectable increase of monoterpene accumulation and the apparent decrease of some compounds could be due to increased loss of monoterpenes in form of   81 resin flow or evaporation as volatiles released from weevil feeding holes without being replenished by induced terpenoid biosynthesis (Miller et al. 2005). Importantly, the monoterpene (+)-3-carene was not detected in Q903.  When given a choice between the resistant H898 and susceptible Q903 genotypes, female weevils were mobile and showed a clear preference for the Q903 genotype. This result suggests that female weevils test a number of different potential hosts and may oviposit on more than one. In addition, male weevils also distinguish between resistant and susceptible trees, but more transiently, both in the presence or absence of females. The majority of females remained on susceptible trees for the entire period of my experiments (23 days) whereas males showed a preference for susceptible trees only until day 15. Since males showed similar behavior with or without females, it is unlikely that males are simply following female cues when choosing susceptible trees. It is possible that females are more responsive to an induced defense response in the resistant H898 trees than the males. Induced defense responses have been shown to occur in constitutive and traumatic resin ducts as early as 8 days after methyl jasmonate treatment in Norway spruce (Zulak et al. 2009), and females can be more sensitive to volatile and gustatory cues than males (VanderSar and Borden 1977; Alfaro et al. 1979).  In spite of a strong induced defense response in H898 trees, my results show that weevils are able to feed, develop ovaries and lay eggs on H898 trees, but ovary development is delayed and reduced, and no surviving offspring (larvae) developed on the resistant H898 trees. Sahota et al. (1998) also suggested that weevils feeding on H898 experience ovary regression or reduced ovary development.  My results suggest that although ovary development may be delayed for weevils feeding on resistant H898, it is not blocked. Sahota et al. (1998) propose that reduced feeding on H898 is an indirect result of   82 the effects of resistance on female weevil physiology. However, as male weevils are clearly able to distinguish between resistant and susceptible trees, females were highly mobile, and I observed low initial feeding on the H898 leader tissue in the no-choice experiment, it is unlikely that ovary regression is the overriding defense mechanism in H898. I also showed that feeding punctures do not necessarily correlate with the number of eggs laid, similar to the findings in Nicole et al. (2006), or with the number of surviving larvae.  Thus, interference with egg viability and larval development is likely to be a factor in the resistance of the H898 genotype.  Alfaro et al. (1980) found that the attractiveness or deterrence of certain chemicals can change depending on the concentration being tested. I show a similar dose dependent repellent effect and found that weevils reacted similarly to (-)-limonene, (+)-α-pinene, and (+)-3-carene. Although I did not test combinations of these compounds, it is possible that blends may be more effective over single pure compounds. Although the female weevils were not significantly deterred by volatiles emitted from H898 trees when given the choice between H898 and Q903, a larger, although not significant, number of female weevils moved away from H898. Again, this suggests that although males are capable of distinguishing resistant and susceptible trees, females appear to be more sensitive to volatile and gustatory cues.  3.5 Conclusions The results of this study support a multi-layered defense model in the highly resistant H898 Sitka spruce genotype. The almost complete resistance of H898 could be due to the combination of a strong constitutive defense that is deterrent to both male and female adult weevils, and an induced response that is deterrent to females, causes a lag in female ovary development, and prevents egg development, hatching, or larvae development.  My results   83 are in agreement with multi-genic resistance (Alfaro et al. 1993). Isolating and testing each potential toxic or deterrent effect in the H898 tree genotype is a daunting task. Fumigation or feeding bioassays using potentially toxic compounds in Sitka spruce (dehydroabietic acid for example, see Chapter 2), are immediate experiments that may further illuminate some of these resistance effects. Recently, RNAi knockout lines for a number of genes previously suggested to play a role in white spruce resistance to insect attack (the (+)-3-carene synthase and the levopimaradiene/abietadiene synthase, a diterpene synthase gene for example) have been developed in the Bohlmann laboratory in collaboration with the Arboréa project based in Quebec (www.arborea.ulaval.ca). Removing suspected ‘resistance genes’ within an otherwise unchanged genetic background and assessing the resulting phenotypes using weevil behavioral and physiological responses may eventually be an effective method for understanding resistance in conifers.    84 3.6 References Alfaro RI. 1989. Stem defects in Sitka spruce induced by Sitka spruce weevil, Pissodes strobi (Peck). Proc. IUFRO Working Group on Insects Affecting Reforestation, Vancouver, B.C. pp. 177–185. R.I. Alfaro and S.G. Glover (editors).  Alfaro RI, Hulme M, and Ying C. 1993. Variation in attack by Sitka spruce weevil, Pissodes strobi (Peck), within a resistant provenance of Sitka spruce. J. Entomol. Soc. Brit. Columbia. 90:24-30.  Alfaro RI, King JN, Brown RG, and Buddingh SM. 2008. Screening of Sitka spruce genotypes for resistance to the white pine weevil using artificial infestations. For. Ecol. Manage. 255: 1749-1758.  Alfaro RI, Pierce HD, Borden JH, and Oehlschlager AC. 1979. A quantitative feeding bioassay for Pissodes strobi Peck (Coleoptera: Curculionidae). J.Chem. Ecol. 5(5): 663- 671.  Alfaro RI, Pierce HD, Borden JH, and Oehlschlager AC. 1980. Role of volatile and non- volatile components of Sitka spruce bark as feeding stimulants for Pissodes strobi Peck. (Coleoptera: Curculionidae). Can. J. Zoo. 58: 626-632.  Alfaro RI, Borden JH, King JN, Tomlin ES, McIntosh RL, and Bohlmann J. 2002. Mechanisms of Resistance in Conifers against shoot infesting insects. In: "Mechanisms and Deployment of Resistance in Trees to Insects." M.R. Wagner, K.M. Clancy, F. Lieutier, and T.D. Paine (eds.). Kluwer Academic Press, Dordrecht, the Netherlands. Pages 101-126.   85  Byun McKay A, Hunter WL, Goddard K-A, Wang SX, Martin D, Bohlmann J, and Plant AL. 2003. Insect attack and wounding induce traumatic resin duct development and gene expression of (−)-pinene synthase in Sitka spruce. Plant Phys. 133: 368-378.  Danci A, Schaefer PW, Schopf A, and Gries G. 2006. Species-specific close-range sexual communication systems prevent cross-attraction in three species of Glyptapanteles parasitic wasps (Hymenoptera: Braconidae). Biological Control, 39: 225-231.  Harman DM and Kulman HM. 1966. A technique for sexing live white-pine weevils, Pissodes strobi. Ann. Entomol. Soc. Am. 59(2): 315-317.  Khan NR and Musgrave AJ. 1969. Observations on the functional anatomy of the reproductive organs of Sitophilus (Coleoptera: Curculionidae). Can. J. Zoo. 47: 665-669.  King JN and Alfaro RI. 2009. Developing Sitka spruce populations for resistance to the white pine weevil: summary of research and breeding program. B.C. Ministry of Forests and Range, Forest Science Program, Victoria, B.C. Technical Report 050. www.for.gov.bc.ca/hfd/pubs/Docs/Tr/Tr050.htm.  Lewinsohn E, Savage T, Gijzen M, and Croteau R. 1993. Simultaneous analysis of monoterpenes and diterpenoids of conifer oleoresin. Phytochem. Anal. 4:220-225.  Manville JF, Sahota TS, and Hollmann J. 2004. Geotaxis and phototaxis are note determinant factors for white pine weevil (Col., Curculionidae) oviposition location on intact trees and severed treetops. J. Appl. Entomol. 128(5): 365-368.   86  Martin D, Tholl D, Gershenzon J, and Bohlmann J. 2002. Methyl jasmonate induces traumatic resin ducts, terpenoid resin biosynthesis, and terpenoid accumulation in developing xylem of Norway spruce stems. Plant Phys. 129:1003-1018.  Miller B, Madilao LL, Ralph S, and Bohlmann J. 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 putative octadecanoid pathway transcripts in Sitka spruce. Plant Phys. 137: 369-382.  Nicole MC, Lavallée R, Bauce É, Charest M, and Séguin A. 2006. Stimulating effect of bark polar fraction from the terminal leader of Norway spruce, Picea abies, on white pine weevil, Pissodes strobi, feeding and oviposition. J. Appl. Entomol. 130(5): 284-289.  Perez-Medoza J, Throne JE, and Baker JE. 2004. Ovarian physiology and age-grading in the rice weevil, Sitophilus oryzae (Coleoptera: Curculionidae). J. Stored Prod. Res. 40: 170-196.  Pernal SF and Currie RW. 2000. Pollen quality of fresh and 1-year-old single pollen diets for worker honey bees (Apis mellifera L.). Apidologie, 31: 387-409.  Sahota TS, Manville JF, Peet FG, White EE, Ibaraki AI, and Nault JR. 1998. Resistance against white pine weevil: effects on weevil reproduction and host finding. Can. Entomol. 130: 337-347.    87 VanderSar TJD and Borden JH. 1977. Aspects of host selection behavior of Pissodes strobi (Coleoptera: Curculionidae) as revealed in laboratory feeding bioassays. Can. J. Zoo. 55: 405-414.  Velthus HHW. 1970. Ovarian development in Apis mellifera worker bees. Entomol. Exp. Appl. 13: 377-394.  Zulak KG, Lippert DN, Kuzyk M, Domanski D, Chou T, Borchers CH and Bohlmann J. 2009. Targeted proteomics using selected reaction monitoring (SRM) reveals the induction of specific terpene synthases in a multi-level study of methyl jasmonate treated Norway spruce (Picea abies). Plant J. 60(6):1015-30.    88 4 RESISTANT AND SUSCEPTIBLE SITKA SPRUCE CHEMOTYPES SHOW DIFFERENTIAL EXPRESSION OF MULTIPLE (+)-3-CARENE SYNTHASE-LIKE GENES.3   4.1 Introduction Traumatic resinosis, the release and de novo production of terpene-saturated oleoresin, occurs in response to attack by white pine weevil (Alfaro 1995; Miller et al. 2005), and this response is a significant cause of weevil death (Overhulser and Gara 1981). Mono- and diterpenoid compounds typically comprise more than 90% of the oleoresin terpenoids in spruce stems (Martin et al. 2002; Miller et al. 2005; Zulak et al. 2009). The terpene synthase (TPS) enzymes responsible for the biosynthesis of oleoresin terpenoid compounds arise from a large and highly diverse TPS gene family. The larger terpenoid family (consisting of mono-, sesqui- and diterpene synthase genes) is divided into six subfamilies of which the TPSd family contains all conifer TPS genes, but no genes from angiosperms (Bohlmann et al. 1998, Martin et al. 2004) (Figure 4-1). The (+)-3-carene synthase gene sequence, a monoterpene synthase gene, falls into the TPSd1 class of the TPSd subfamily.    3 A version of this chapter will be submitted for publication. Hall, D., Robert, J., Keeling, C.I., Jancsik S., Hamberger, B., Lara Quesada, A., Kuzyk, M., Domanski, D., Borchers, C.H., and Bohlmann J. Comprehensive analysis of (+)-3-carene biosynthesis in two chemotypes of Sitka spruce.   89  Figure 4-1 The conifer TPS-d subfamily. The conifer TPS-d subfamily is divided into three groups (TPS-d1, TPS-d2, and TPS-d3). The (+)-3-carene synthase identified in Norway spruce is designated by the red rectangle. Figure is modified from Martin et al. 2004.   I showed in Chapter 2 that two monoterpenes, (+)-3-carene and terpinolene, as well as dehydroabietic acid, a diterpene, were associated with Sitka spruce resistance to white pine weevil in the Haney region of British Columbia. Several additional studies show evidence that (+)-3-carene, a monoterpene common in conifer oleoresin, may be an important semiochemical for both white pine weevil and other insect species. Storer and Speight (1996) suggest that reduced Dendroctonus micans larval dry weight and survival is associated with increased α-pinene, β-pinene and 3-carene content in Norway spruce bark. Rocchini et al. (2000) suggest that 3-carene may be repellent to ovipositing females or may reduce larval development in lodgepole pine trees that are resistant to attack by pitch moth,   90 Synanthedon novaroensis. Bichao et al. (2003) show that monoterpenes are an important component of compounds sensed by a European conifer weevil, Pissodes notatus, and that this species has abundant olfactory neurons that respond selectively to 3-carene.  Previously, a Norway spruce (+)-3-carene synthase was cloned and characterized by Fäldt et al. (2003). The Norway spruce (+)-3-carene synthase is a multi-product enzyme producing predominantly (+)-3-carene (78%), but also significant amounts of terpinolene (11%) as well as smaller amounts of sabinene, myrcene, γ-terpinene, α-pinene, β- phellandrene, α-terpinene, and limonene (as 5% or less of the total monoterpenes produced). In addition. Miller et al. (2005) showed the induction of terpene synthase gene transcripts encoding 3-carene synthase in Sitka spruce stems exposed to white pine weevil feeding. A toxic mechanism for (+)-3-carene has not been demonstrated, but a few studies address its toxic effect in animals. Lastbom et al. (2003) report that 3-carene is a lung irritant in guinea pigs and potentially in human beings. 3-Carene is also used as the starting material for the artificial synthesis of potent insecticidal compounds (Dhillon et al. 1991).  Although terpene compounds exist as highly diverse mixtures in conifer oleoresin, little is known about the mechanisms that determine presence or absence of individual terpene compounds in resistant or susceptible Sitka spruce genotypes. Uncovering the connection between terpene synthase genes and the terpene phenotype of resistant trees may have direct application to tree breeding programs designed to enhance tree resistance to insect attack. As part of the tree breeding program for Sitka spruce, the British Columbia Ministry of Forests and Range in collaboration with the Canadian Forest Service has developed a screening program for resistance in Sitka spruce (King and Alfaro 2009; Alfaro et al. 2008). From these screening trials, one genotype (referred to as Haney 898 or H898) stood out as being almost completely resistant to weevil attack. The H898 genotype produces (+)-3-   91 carene as a component of oleoresin. I contrasted this resistant clone with a highly susceptible genotype, Q903, a clone that produces only trace amounts of (+)-3-carene and that originates from the Queen Charlotte Islands in British Columbia where white pine weevil does not naturally occur (Humble et al. 1994). The objective of this study was to identify the (+)-3-carene synthase gene(s) in Sitka spruce and to assess the (+)-3-carene synthase gene expression, enzyme activity and resulting terpene profile using a case study of two genotypes of Sitka spruce: one highly resistant genotype (H898) that produces (+)-3-carene and one highly susceptible genotype (Q903) that produces only trace amounts of (+)-3- carene. Three (+)-3-carene synthase genes were cloned and functionally characterized in each tree genotype showing a new level of diversity in the terpene synthase gene family where only one (+)-3-carene synthase gene had previously been cloned in Norway spruce. In addition, I showed differential (+)-3-carene synthase gene expression between H898 and Q903.  4.2 Materials and methods The British Columbia Ministry of Forests and Range (Cowichan Lake Research Station) supplied three-year-old grafted clones of the highly resistant spruce genotype (genotype H898 originating from the Haney area east of Vancouver 49°14’N : 122°36’W) and a susceptible genotype (genotype Q903 originating from an area with no history of weevil attack, the Queen Charlotte Islands on British Columbia’s west coast 53°55’N : 132°05’W) (King and Alfaro 2009). In order to increase control over environmental variables, the experimental trees were treated and sampled under greenhouse conditions at the University of British Columbia. Trees were maintained in 1 gallon pots outside on the University of British Columbia (UBC) campus. One month prior to experiments, trees were moved into the ambient temperature and light zone of the UBC greenhouse (as in Miller et al. 2005).    92 4.2.1 cDNA isolation and functional characterization of (+)-3-carene synthase Preliminary work to identify the sequence for the (+)-3-carene synthase gene in Sitka spruce was performed by Dr. Barbara Miller, a former post-doctoral fellow in our laboratory. Dr. Miller isolated RNA from the bark of two tree genotypes (H898 and Q903) after inducing the trees with 0.01% methyl jasmonate (95% Sigma-Aldrich) for two days (method is described in Miller et al. 2005). Dr. Miller used gene specific primers based on a (+)-3- carene synthase gene characterized in Norway spruce (Fäldt et al. 2003), Ss-5prime.F (5’ – GTAGTCCATAAGGAGCAGAAATG- 3’) and Ss-Car_69.4.R (5’ – TTACATAGGCACAGGTTCAAGAACGG- 3’), to obtain six candidate sequences from H898 and eight candidates from Q903. To maximize the likelihood of uncovering all possible (+)-3- carene synthase-like candidates, I repeated Dr. Miller’s amplification using gene specific primers Ss-Car-pEF (5’ –CACCATGTCTGTTATTTCCATTGTGCCG- 3’) and a more specific reverse primer CarUTR_gap.R@2035 ( 5’ – CCTAATAGGCTGAAAAGTACAATAATAAACCACATAACCTG- 3’). This approach identified 11 additional candidate sequences in H898 and 3 additional candidate sequences in Q903. The fragments were cloned into the pDRIVE cloning vector (Qiagen). The plasmids were transformed into alpha-select gold efficiency competent cells (Bioline) for subsequent screening using colony PCR. Plasmids containing the correct insert size (candidate sequences greater than 1.8 kilobases) were isolated (QiaPrep Spin Miniprep kit, Qiagen) and fully sequenced. Each candidate was sequenced at least twice using vector primers as well as internal gene-specific primers.  4.2.2 Sequence confirmation with RACE Because there were multiple (+)-3-carene synthase-like candidates in each genotype, I conducted rapid amplification of cDNA 3’ ends (3’-RACE) for three putative groups of (+)-3- carene synthase-like sequences in both genotypes (see Table 6). I targeted the untranslated   93 regions (3’ UTR) in order to confirm the candidate sequence groups using UTR sequence information in addition to the sequence of the open reading frame.  Table 4-1 3’-RACE gene-specific primer combinations. 3’RACE gene-specific outer primer 3’-RACE gene-specific inner primer GAACGATTGGGGATCGACAGAC CAACTAAATTGGGAGCTGCTTAG GAACGATTGGGGATCGACAGAC GAAAGACAATCCCGGATCCACAC GAACGATTGGGGATCGACAGAC GTGACACACGGTGTTACCAGAAT GAGGCAGATCAGCAGCATTCT CAACTAAATTGGGAGCTGCTTAG GAGGCAGATCAGCAGCATTCT GAAAGACAATCCCGGATCCACAC GAGGCAGATCAGCAGCATTCT GTGACACACGGTGTTACCAGAAT GAAGCCCTACAAACGATTCCAGTC CAACTAAATTGGGAGCTGCTTAG GAAGCCCTACAAACGATTCCAGTC GAAAGACAATCCCGGATCCACAC GAAGCCCTACAAACGATTCCAGTC GTGACACACGGTGTTACCAGAAT   Nested 3’-RACE was conducted using the Ambion FirstChoice RLM-RACE kit with the following modifications. The reverse transcription in the first step was conducted using Superscript III First Strand Synthesis Supermix (Invitrogen). BioMix Red (BIOLINE) was used as the DNA polymerase and the annealing temperature used in the thermocycler protocol was 58°C. Three gene-specific outer primers and three gene-specific inner primers were used (Table 6).  Primer sequences were designed based on the known (+)-3-carene synthase sequence from Norway spruce (Fäldt et al. 2003), on the new candidate gene sequences, and on Sitka spruce EST database searches (Ralph et al. 2008). The RACE fragments were cloned into the pDRIVE vector and transformed into BIOLINE Alpha-Select Gold Efficiency chemically   94 competent cells. Plasmids from selected colonies were amplified directly using the Qiagen PlasmidAmp Kit and sequenced. The RACE sequences were grouped with ORF and UTR sequences from the full-length cDNA candidates in order to establish the most likely number of (+)-3-carene synthase-like genes.  4.2.3 Functional characterization of (+)-3-carene synthase-like genes Functional characterization of full-length cDNAs was based on a procedure by Fäldt et al. (2003) and optimized by Dr. C. Keeling, research associate, and Harpreet Dullat, research assistant, in our laboratory using an E. coli expression vector.  Sticky-end PCR (Zeng 1998) was used to amplify full-length candidate cDNAs for ligation into expression vectors. PCR products were gel purified and ligated into either pET100 (Invitrogen) or pET28b(+) (Novagen) expression vectors containing N-terminal polyhistadine tags according to the manufacturer’s protocols.  OverExpress™C41(DE3) competent cells (used to over-express enzymes that may be toxic to other E.coli cells) containing pRARE 2 plasmids [originally found in Rosetta™ 2(DE3) competent cells with high levels of rare tRNAs needed to express plant cell proteins] were transformed with either pET100 or pET28b(+) vectors containing the full-length (+)-3- carene synthase-like cDNA candidates. Transformed E. coli were grown at 37°C overnight (16-20 hrs) on Luria-Bertani broth (LB) plates containing chloramphenicol (Cm, [50]) and the antibiotic of the pET vector (i.e. ampicillin, Amp [100] for pET100, kanamycin, Kan [50] for pET28b(+)). Three 5 mL cultures of LB containing chloramphenicol and the pET-specific antibiotic were each inoculated with a single colony from the transformation and placed in a 37°C incubator shaking at 220 rpm overnight. The next day, 1mL of these cultures was used to inoculate 50 mL terrific broth (TB) cultures (containing the appropriate antibiotics). The TB   95 cultures were incubated at 37°C shaking at 220 rpm for 3-4 hours until the optical density (OD) at 595 nm measured between 0.5 and 0.8. Protein expression was then induced by adding IPTG (Isopropyl β-D-1-thiogalactopyranoside, Fisher Scientific) to 0.2 mM final concentration and the cultures were incubated overnight at 16°C shaking at 220 rpm. The next day, the cultures were centrifuged for 30 min at 4°C at 3270 xg. The supernatant was discarded and the pellet weight was recorded. Pellets were stored at –80°C for subsequent enzyme extraction and functional characterization.  Frozen cell pellets were thawed on ice and vortexed in lysis buffer (1 mL cell lysis buffer / g cell pellet) to resuspend. Cell lysis buffer was made fresh every day; it contained 25 mL His Trap binding buffer (20 mM NaPO4, 500 mM NaCl, and 90 mM imidazole (99%, Acros Organics) adjusted to pH 7.4 and stored at 4°C), 25 mg lysozyme (from chicken egg white, Sigma-Aldrich), 1 mg DNase (from bovine pancreas, Sigma-Aldrich), 250 µL 100 mM MgCl2 in dH20, and 125 µL 200 mM phenylmethanesulfonyl fluoride (PSMF, Sigma-Aldrich) in isopropanol.  The resuspended pellets in buffer were transferred to pre-cooled 15 mL Falcon tubes using an additional 0.5 mL cell lysis buffer per gram cell pellet to rinse the original lysis tube. The resuspended pellets were kept on ice for 30 minutes vortexing occasionally at low speed. The suspension was then sonicated using a Bronson 250 sonicator for two minutes with the duty cycle set to 10%. The mixture was transferred to two microcentrifuge tubes and centrifuged at 24,000 xg for 20 minutes at 4°C. The cleared lysate was passed through His SpinTrap columns (GE Healthcare) according to the manufacturer’s protocol. Protein was eluted in one fraction using 300 µL of HisTrap elution buffer (20 mM NaPO4, 500 mM NaCl, and 500 mM imidazole adjusted to pH 7.4 and stored at 4°C) and used immediately in the single-vial terpene synthase enzyme assay.   96  The single-vial terpene synthase enzyme assays were adapted by Dr. Chris Keeling and Dr. Dawn Hall, based upon previous work by O’Maille et al. (2004), to conduct the entire procedure in a single 2 mL gas chromatography vial. 50 µL of the semi-purified enzyme was placed into the vial with 450 µL of an aqueous monoterpene synthase buffer containing 25 mM HEPES (99.5%, Sigma-Alrich), 100 mM KCl, 10 mM MnCl2, 10% glycerol, with 5 mM dithiothreitol (DTT) added fresh with each use). 25 µL of the substrate was added, then the assay buffer was gently overlaid with a 500 µL layer of pentane (HPLC grade, Fisher Scientific). The recombinant enzymes were tested with the monoterpene synthase substrates geranyl diphosphate (GPP, Echelon Biosciences Inc.) with a final concentration of 138 µM and neryl diphosphate (NPP, Echelon Biosciences Inc.) with a final concentration of 138 µM, as well as with the sesquiterpene substrate, farnesyl diphosphate (FPP, Sigma- Aldrich) with a final concentration of 116 µM, and the diterpene substrate, geranyl geranyl diphosphate (GGPP, Sigma-Aldrich) with a final concentration of 111 µM. The vials were placed in a 30°C water bath for 60 minutes, vortexed for 10-20 sec to extract the terpene compounds into the pentane overlay, then centrifuged for 30 min at 4°C at 1000 xg. The spin was repeated, if necessary, until the two phases were well separated and the top layer was homogeneous and clear.  Compounds in the pentane extract were identified using gas chromatography (Agilent 6890A series) coupled with mass spectrometry (5973N mass selective detector, quadropole analyzer, electron ionization, 70eV) through the comparison of compound retention time with the retention time of commercially available authentic standards as well as comparison with mass spectral libraries (Wiley7Nist05).    97 The following program was used to identify and separate the monoterpene peaks on a DB-WAX capillary column (J&W 122-7032, 250 µm diameter, 30 m length, and 0.25 µm film thickness): the 40°C initial temperature was held for 3 minutes, then increased by 8°C min-1 to 240°C and held for 10 minutes (total run time is 38 minutes). The initial injection temperature was set at 240°C, and the initial flow rate was 1 mL He min-1. Stereochemistry of the compounds was determined where authentic standards were available on a Cylcodex-B chiral capillary column (J&W 112-2532, 250 µm diameter, 30 m length, and 0.25 µm film thickness) using the following temperature program: the 55°C initial temperature was increased by 1°C min-1 to 100°C, then increased at 10°C min-1 to 230°C held for 10 minutes (total run time is 69 minutes). The initial injection temperature was set at 230°C, and the initial flow rate was 1 mL He min-1.  4.2.4 Amino acid alignment and enzyme tertiary structure In order to identify any amino acid changes occurring in the active site between the candidate genes, amino acid sequences were modeled on the three-dimensional structure of a recently crystalized monoterpene synthase, 4S-limonene synthase from spearmint, Mentha spicata, (Hyatt et al. 2007) using DeepView / Swiss-PdbViewer version 3.7 © 1995- 2001 (Guex and Peitsch 1997) accessible from the SWISS-MODEL server located at: http://swissmodel.expasy.org//SWISS-MODEL.html. Trapp and Croteau (2001) suggest that monoterpene synthase tertiary structure is highly conserved and therefore spearmint limonene synthase provided an appropriate skeleton for predicting the tertiary structure of the (+)-3-carene synthase-like enzymes in Sitka spruce.    98 4.2.5 Comparison of (+)-3-carene synthases in H898 and Q903 Sitka spruce genotypes Experiments were designed to compare terpene synthase gene expression patterns, enzyme activities and terpene product profiles between control and methyl jasmonate- treated trees for tree genotype H898, a highly resistant genotype, and genotype Q903, a highly susceptible genotype. I chose to use methyl jasmonate as a defense response elicitor (Martin et al. 2002; Martin et al. 2005; Phillips et al. 2006; Zulak et al. 2009) in order to ensure consistent defense responses across biological replicates. Four trees (biological replicates) per genotype per treatment were sampled over 5 timepoints (2, 7, 14, 21, 28 days) for a total of 40 trees per genotype.  Treated trees were sprayed with 50 mL 0.1% (v/v) methyl jasmonate (95%, Sigma- Aldrich) in 0.1% (v/v) Tween20 (Fisher Scientific) solution. Control trees were sprayed with 0.1% Tween20 solution only. The treated trees were kept separately from control trees and all trees were allowed to dry overnight. Selected trees from each timepoint were randomly distributed within blocks of treated and control trees in the greenhouse space. Treated and control tree blocks were separated by 2 metres in order to prevent an induced defense response in control trees as a result of proximity to methyl jasmonate-treated plants. During sampling, the leader and stem were clipped, the needles removed, and the bark and xylem tissue were flash frozen together in liquid nitrogen. The frozen samples were then analysed for the timing of (+)-3-carene-like terpene synthase transcript accumulation, terpene synthase enzyme activity, and the terpene product profile.  4.2.5.1 (+)-3-Carene synthase transcript expression patterns Quantitative real time RT-PCR (qRT-PCR) was employed in order to identify differential (+)-3-carene synthase expression between the Sitka spruce genotypes. Probes for qRT-   99 PCR were designed based on unique areas of each (+)-3-carene synthase-like sequence identified as well as for a Sitka spruce α-pinene synthase (WS0291_K15, C. Keeling, unpublished data) for comparison.  4.2.5.2 RNA extraction protocol and cDNA synthesis I used a method for RNA extraction that was developed in our laboratory by Kolosova et al. (2004). It was originally based on the work of Wang (2000) and Chang et al. (1993). The RNA extraction protocol was modified for small amounts of tissue (~100 mg per extraction) because of the limited quantity of leader tissue available for my project. The final RNA pellet was resuspended in 15 µL DEPC-treated H20 then further purified using an RNeasy mini kit (Qiagen). In order to prevent genomic DNA contamination, I treated the on-column RNA with an RNase-free DNase set (Qiagen) according to the manufacturer’s protocol. The amount of total extracted RNA ranged from 19 µg to 64 µg per leader sample.  For each sample, 3 µg of total DNase-treated RNA was added to a Sprint RT complete- double preprimed first-strand cDNA synthesis kit (Clontech). This kit reduced human pipetting error over the 88 biological replicates as only water and template were added to the prepared reaction tube, and it produced high quality cDNA suitable for qRT-PCR.  RNA quality was assessed using an Agilent 2100 Bioanalyser. RNA integrity numbers averaged 7.4 indicating sufficient quality for qRT-PCR (Udvardi et al. 2008).  Four ubiquitously expressed reference genes were tested as internal controls for the qRT-PCR reaction: elongation factor 1-alpha (ELF), translation initiation factor EIF-5A (TIF), polyubiquitin (UBQ), and tubulin alpha (TUA). These genes are used routinely in our laboratory as reliable and consistent positive controls of gene expression. I tested the primers with leader tissue from both H898 and Q903 methyl jasmonate-treated tissue as   100 well as the control Tween20 treated tissue in order to choose the least variable positive control. I chose to use TIF as the positive control for each of the plates due to its highly consistent expression. The average Ct value for TIFF (± 1 standard deviation) across both genotypes and treatments was 19.87 ± 0.31.  4.2.5.3 Primer design and q RT-PCR protocol Primers (Table 7) were designed using the following criteria to ensure transcript-specific binding: Tm: 60-64°C, G/C content: 40-60%, primer length: 20-24bp, amplicon length: 80- 300bp, no more than 3 G/C nucleotides in the first 5bp of the 3’ foot, first 3’ nucleotide is a G or C, no stretches of more than 3 repeats of the same nucleotide, and no palindromes (Udvardi et al. 2008).  Table 4-2 Gene-specific qRT-PCR primers (oriented 5’ to 3’). Stop codons are highlighted in bold. UTR group RT-PCR forward primer RT-PCR reverse primer (template sequence) CAR_1: Q09 H09 3-carene synthase-like CACCACTTCTACAAATACCGAG [in last exon]  GATATGTGGTTACATAGGCAC (GTGCCTATGTAACCACATATC) [incl STOP and into UTR]  CAR_2: H02 H08 3-carene synthase-like CACACGGTGTTACCAGGCAG [over intron splice site] GTACATCTGATATGTTGATACGC (GCGTATCAACATATCAGATGTAC) [early UTR]  CAR_3: Q270 3-carene synthase-like GCTTATGAAACCTGACAACAAC [in last exon] GAAACGGTGTAGCCATCACG (CGTGATGGCTACACCGTTTC) [in last exon]  SAB: Q05 H05 sabinene synthase-like GATGTTACCAGGCAGATAGAGA [over intron splice site] CAGTTACATAGGGAGAGGTTC (GAACCTCTCCCTATGTAACTG) [incl STOP and into UTR]  PIN: pinene synthase  CAACCCAAACAGCAGTGTTC [in last exon]  CAGTTACAAAGTCACAGGATC (GATCCTGTGACTTTGTAACTG) [incl STOP and into UTR]   Plate setup was designed to compare one tree from each treatment, timepoint, and genotype (complete block design) on one plate as recommended in Rieu and Powers   101 (2009). For example, a ‘day 2’ plate contained one ‘day 2’ methyl-jasmonate-treated tree and one ‘day 2’ control tree from H898 and one ‘day 2’ methyl-jasmonate-treated tree and one ‘day 2’ control tree from Q903 replicated for each of the five primer pairs (Table 7) and the reference gene. Each well was replicated twice within a plate. For every sample and primer set, I performed a no-reverse-transcription control (i.e. pooled total RNA from one genotype and treatment for the sampling day as a template) and a no-template control (i.e. water only).  The thermocycler (DNA Engine Opticon 2) program started at 95°C for 15 minutes, then cycled through 1 minute at 94°C to denature, 30 seconds at 60°C to anneal the primers, then 30 seconds at 72°C to elongate the amplicons. The fluorescence was read before repeating the cycle. The cycle was repeated 40 times. Melting curves were also generated for each sample and primer pair by ramping from 65°C to 95°C in 0.2°C increments. Data was compiled using Real-time PCR Miner software (Zhao and Fernald 2005).  4.2.5.4 qRT-PCR calculations To calculate the relative quantity (RQ) of each transcript including the housekeeping gene, the following calculation was used (Zhao and Fernald 2005):  RQ = (1/(1+E)Ct)  where “E” is the average efficiency of a given primer pair across all reactions on a given plate and “Ct” is the number of cycles to a threshold fluorescence level. Each sample was then normalized (NRQ) to the expression of the reference gene:  NRQ = RQsample / RQreference   102   The average NRQ and standard error estimates were calculated from 4 biological replicates per genotype per treatment per timepoint for each primer pair.  4.2.5.5 Statistical analysis Where they violated the assumption of homogeneity of variance and normality, data were log transformed to satisfy the assumptions of ANOVA. All statistics were conducted using SYSTAT 11.0 (Systat Software Inc. 2004). The Ct values from the duplicate sample wells on a plate were averaged and used to calculate the NRQ for that sample. Efficiency values for the NRQ calculation were averaged from the all of the efficiency values for that primer pair on the plate (Rieu and Powers 2009). ANOVAs were performed to identify differences between methyl jasmonate-treated trees and control trees at each sampling day for each genotype. Because the treated and control trees were paired as a consequence of the qRT-PCR plate design, I used the ratio of methyl jasmonate-treated expression to control expression for each sample pair to determine whether the ratio of treatment to control between the treatment and control were significantly different than 1 (no effect).  4.2.5.6 Protein extraction Cell-free extracts of total protein were used to test the enzymatic biosynthesis of (+)-3- carene in the H898 genotype versus the Q903 genotype. The protein extraction protocol was modified for small amounts of tissue (0.1 - 0.2 g fresh weight) from Martin et al. (2002); the original basis for this protocol is from Lewinsohn et al. (1991). Tissue samples were ground to a fine powder in liquid nitrogen; a final grind was conducted with protein extraction buffer in liquid nitrogen. The buffer-tissue mixture was scraped into 2 mL eppendorf tubes then frozen overnight at -80°C. The protein extraction buffer stock solution contained 50 mM HEPES (99.5% Sigma-Aldrich) pH 7.2, 5 mM ascorbic acid (Sigma-Aldrich), 5 mM sodium   103 bisulfite (Sigma-Aldrich), 10 mM MgCl2, 1% (w/v) polyvinylpyrrolidone (PVP, Sigma-Aldrich), 10% (v/v) glycerol (Sigma-Aldrich), 0.1% (v/v) Tween20 (enzyme grade, Fisher Scientific). The day of use, 5 mM dithiothreitol (DTT, Sigma-Aldrich), 1 µL protease inhibitor cocktail (for use in purification of Histidine-tagged proteins, DMSO solution, Sigma-Aldrich P8849), 1% (w/v) polyvinylpolypyrrolidone (PVPP, Sigma-Aldrich), 4% (w/v) amberlite (Sigma-Aldrich), and 0.1% (w/v) activated charcoal (Sigma-Aldrich) was added.  The next day, the buffer-tissue mixture was thawed on ice then shaken at 4°C for 30 minutes. The samples were centrifuged at 13,000 rpm for 10 minutes at 4°C, and the supernatant was desalted using a sephadex PD minitrap G-25 medium chromatography column (GE Healthcare) according to the manufacturer’s spin protocol. The desalting buffer contained 25 mM HEPES (99.5% Sigma-Aldrich) pH 7.2, 100 mM KCl, and 10% glycerol. The total protein content was estimated using a Bradford reagent assay (Bio-Rad) comparing extracted protein amounts with a calibration curve of known commercially available protein concentrations.  4.2.5.7 Monoterpene synthase assays and enzyme activity Enzyme assay and activity protocols were modified by Dr. Dawn Hall, a post-doctoral fellow in our laboratory to optimize for small amounts of protein (100 µg of total protein per assay) based on Martin et al. (2002). Martin et al. (2002) developed their protocol based on previous methods in Lewinsohn et al. (1991), LaFever et al. (1994), and Bohlmann et al. (1997).  Total protein extract (100 µg) was combined in a vial with 9 µL substrate, either geranyl diphosphate (GPP) or neryl diphosphate (NPP) to a final concentration of 50 µM, and aqueous monoterpene synthase buffer (see recombinant enzyme assay protocol for   104 monoterpene synthase buffer composition) to a final volume of 500 µL. The aqueous layer was overlaid with 500 µL layer of tert-butyl methyl ether (CHROMASOLV Plus, for HPLC, 99.9% MTBE, Sigma-Aldrich) containing 100 µg/mL isobutyl benzene (Fluka) as an internal standard. The vials were then placed in a 30°C water bath for 60 minutes. The vials were vortexed for 10-20 seconds to extract the terpene compounds and then centrifuged for 30 min at 4°C at 1000 xg. The spin was repeated, if necessary, until the two phases were separated and clear.  Monoterpene peaks were identified in the MTBE extracts by GCMS on a DB-WAX capillary column using the same GC program as described for the recombinant enzyme assays. Stereochemistry of the compounds was again analysed using the Cylcodex-B chiral capillary column. Because the assays with native enzymes produced monoterpenes in small amounts, I used a combination of single ion monitoring (SIM) in combination with total ion monitoring mass spectrometry scanning. SIM allows the specific detection of characteristic mass-to-charge (m/z) ratios and thus eliminates much of the background noise. I scanned for ion masses that are common for monoterpene compounds (69, 93, 121, 134, and 136). This method allowed me to detect very small (nanogram) amounts of terpenes produced from the protein extracts.  Identified compounds were quantified using response factors calculated from a known concentration of isobutyl benzene, the internal standard. The amount of monoterpene compound (µg) per 100 µg of total protein was calculated and used as the final value for comparisons.    105 4.2.5.8 Monoterpene product profile The monoterpene extraction method was based on Lewinsohn et al. (1993) and conducted as described in Martin et al. (2002) with the following modifications. As in the original protocol, ~0.2 grams dry weight sample tree tissue were immersed in 1.5 mL tert- butyl methyl ether (MTBE) (Sigma Aldridge), containing 100 µg/mL isobutyl benzene (Fluka) as an internal standard, and shaken overnight at room temperature. The next day, the extract was washed with 0.3 mL of 0.1M (NH4)2CO3 (pH 8.0). The sample tissue was removed, dried at room temperature in the fumehood for one week, and then weighed. Extracts were analysed by GCMS for monoterpene content.  The following GC program was used to identify and separate the monoterpene peaks on a SGE Solgel-Wax capillary column (Mandel Scientific SG-054796), 250 µm diameter, 30 m length, and 0.25 µm film thickness): the 40°C initial temperature was increased by 3°C min-1 to 110°C, then increased at 10°C min-1 to 180°C, then finally increased by 15°C min-1 to 260°C held for 15 minutes (total run time is 50.67 minutes). The initial injection temperature was set at 250°C, and the initial flow rate was 1 mL He min-1.  Compounds were identified by comparing compound retention times with the retention time of commercially available authentic standards. Compound identities were confirmed using gas chromatography (Agilent 6890A series) coupled with mass spectrometry (5973N mass selective detector, quadropole analyzer, electron ionization, 70eV).  Stereochemistry of the compounds was determined where authentic standards were available on a Cylcodex-B chiral capillary column (J&W 112-2532, 250 µm diameter, 30 m length, and 0.25 µm film thickness) using the following temperature program: the 55°C initial temperature was increased by 1°C min-1 to 100°C, then increased at 10°C min-1 to 230°C   106 held for 10 minutes (total run time is 69.00 minutes). The initial injection temperature was set at 230°C, and the initial flow rate was 1 mL He min-1.  Identified compounds were again quantified using response factors calculated from a known concentration of isobutyl benzene. The amount of compound (µg) per gram dry weight of the tissue extracted was calculated and used as the final value for comparisons.  4.3 Results  4.3.1 Functional characterization identifies seven full-length (+)-3-carene synthase-like cDNAs Seven candidate genes were functionally characterized (Figure 4-2). Two (+)-3-carene synthases, H02 and H08, and one (+)-sabinene synthase, H05, were found in the candidate cDNA’s of Sitka spruce genotype H898. One (+)-3-carene synthase, Q09, and one (+)- sabinene synthase, Q05, were found among the expressed candidate cDNA’s in genotype Q903.    107 H898 [H08] Q903 [Q09] H898 [H02] H898 [H05] Q903 [Q05] D et ec to r r es po ns e CH3 CH3 CH3 (+)-3-carene (+)-Sabinene CH3 CH3 CH2 CH3 CH3 CH3 Terpinolene 550 600 650 700 750 800 Retention Index A B C D E F G RI: 612 RI: 774 40 60 80 100 120 140 m/z 41 53 67 79 93 105 121 136 41 53 69 77 93 105 121 136 43 53 67 79 93 105 121 136 41 53 69 77 93 105 121 136 RI: 612 41 53 67 77 93 105 121 136 RI: 639 RI: 774 41 53 67 79 93 105 121 136 40 60 80 100 120 140 m/z D et ec to r r es po ns e Enzyme product sabinene Enzyme product 3-carene RI: 639 D et ec to r r es po ns e Enzyme product α-terpinolene Authentic 3-carene Authentic sabinene Authentic α-terpinolene D et ec to r r es po ns e D et ec to r r es po ns e D et ec to r r es po ns e D et ec to r r es po ns e  Figure 4-2 The major products of the Sitka spruce (+)-3-carene synthases and (+)- sabinene synthase. Total ion chromatogram of the products for each of the functionally characterized TPS enzymes encoded by candidate cDNAs (A): the (+)-3-carene synthases from H898 (H08 and H02), the (+)-3-carene synthase from Q903 (Q09), and the sabinene synthases (H05 and Q05). The mass spectra are shown for the major enzyme products including 3-carene (B), sabinene (D), and α-terpinolene (F) and this compared to the mass spectra derived from authentic standards (C, E, G).   In addition, another (+)-3-carene synthase, I02, was functionally characterized in a third Sitka spruce genotype FB3-425, and another (+)-3-carene synthase, F08, was characterized in Interior spruce (Picea glauca x engelmannii) genotype Fal_1028.  The product profiles for the seven functionally characterized genes are summarized in Table 8 and compared to the (+)-3-carene synthase from Norway spruce (Fäldt et al. 2003).     108 Table 4-3 Percentage of each monoterpene product for the functionally characterized 3- carene synthase-like enzymes from Sitka spruce genotypes H898, Q903, and FB3_425 as they compare to the gene cloned from Interior spruce (F08) and the previously identified (+)- 3-carene synthase from Norway spruce (JF67).  (-) -α -p in en e α- th uj en e (-) -β -p in en e (+ )-s ab in en e (+ )-3 -c ar en e m yr ce ne  α- te rp in en e (-) -li m on en e (-) -β -p he lla nd re ne  γ- te rp in en e α- te rp in ol en e Sitka spruce H898 [H05] (+)-sabinene synthase  6.01 1.69 2.37 52.57 1.41 1.88 1.03 0.91 1.64 1.35 29.14 Sitka spruce Q903 [Q05] (+)-sabinene synthase  5.90 1.52 2.38 52.70 1.05 1.66  1.18 2.63 1.49 29.49 Sitka spruce Q903 [Q09] (+)-3-carene synthase  4.17   8.21 52.89 5.46 1.53 1.71 2.82 1.56 21.63 Sitka spruce H898 [H02] (+)-3-carene synthase     10.79 67.78 7.42     14.02 Sitka spruce H898 [H08] (+)-3-carene synthase  1.77   8.23 68.18 2.49 1.47  1.47 1.52 13.47 Sitka spruce FB3-425 [I02] (+)-3-carene synthase  4.08   8.17 53.96 5.35  1.70 5.14 1.30 20.29 Interior spruce [F08] (+)-3-carene synthase  8.19   9.21 53.30 5.19 3.42    20.68 Norway spruce [JF67] (+)-3-carene synthase* 0.90   5.00 78.00 3.00  0.40 0.70 1.00 11.00 * data taken from Fäldt et al. (2003)    Similar to previously characterized conifer monoterpene synthases (Martin et al. 2004). all of the functionally characterized Sitka spruce genes encoded for multi-product enzymes. The product profile for each recombinant enzyme was dominated by either (+)-3-carene or   109 (+)-sabinene, followed by α-terpinolene. Other monoterpenes were detected but they were produced in low amounts comprising less than 5% of the total terpenes produced (Table 8). Functionally characterized gene sequences Q09, H02, H08, I02, and F08 are therefore referred to as (+)-3-carene synthases based on the major product of the encoded recombinant protein, and H05 and Q05 are referred to as (+)-sabinene synthases.  4.3.2 Amino acid sequence comparisons show three types of (+)-3-carene synthase-like enzymes: CAR_1, CAR_2, and SAB Based on amino acid sequence, the Sitka spruce Q09 (+)-3-carene synthase (from genotype Q903) is more similar to the (+)-sabinene synthases than to the (+)-3-carene synthases identified in the H898 genotype (Table 9).                 110 Table 4-4 Pairwise percent amino acid sequence identities of the functionally characterized cDNAs in H898 and Q903 in comparison with the Norway spruce (+)-3-carene synthase gene. Spruce species:  Genotype: Gene: Major enzyme product: Sitka spruce H898 [H05] sabinene Sitka spruce Q903 [Q05] sabinene Sitka spruce Q903 [Q09] 3-carene Sitka spruce H898 [H02] 3-carene Sitka spruce H898 [H08] 3-carene Sitka spruce FB3-425 [I02] 3-carene Interior spruce  [F08] 3-carene Norway spruce  [JF67] 3-carene  Sitka spruce H898 [H05] (+)-sabinene synthase   100 83 88 90 82 82 88 Sitka spruce  Q903 [Q05] (+)-sabinene synthase    83 88 90 82 82 88 Sitka spruce Q903 [Q09] (+)-3-carene synthase     84 84 100 99 85 Sitka spruce H898 [H02] (+)-3-carene synthase      98 83 83 95 Sitka spruce H898 [H08] (+)-3-carene synthase       84 84 96 Sitka spruce FB3-425 [I02] (+)-3-carene synthase        98 84 Interior spruce [F08] (+)-3-carene synthase         84 Norway spruce [JF67] (+)-3-carene synthase      Amino acid sequence relatedness of (+)-3-carene synthase and sabinene synthase genes characterized in this study is illustrated with a dendrogram in Figure 4-3    111  Figure 4-3 Phylogenic tree of the sequenced (+)-3-carene synthases and the (+)- sabinene synthases. This tree is based on amino acid sequences of the (+)-3-carene and sabinene synthase genes in H898 and Q903 that have been functionally characterized. The tree also includes the FB3_425 Sitka spruce genotype (+)-3-carene synthase, the Interior spruce (+)-3-carene synthase, and the Norway spruce (+)-3-carene synthase (Fäldt et al. 2003). These are compared to a Sitka spruce (−)-pinene synthase (C. Keeling unpublished data) as an outgroup. The scale bar is equivalent to 0.1 amino acid substitutions per site. The (+)-3-carene synthase sequenced from the susceptible genotype (Q903) is more similar to (+)-3-carene synthases sequenced in other genotypes to the (+)-sabinene synthases than to the (+)-3-carene synthases sequenced from the resistant tree genotype (H898) at the amino acid level.    Figure 4-4 shows the amino acid alignment of the functionally characterized Sitka spruce and interior spruce (+)-3-carene and sabinene synthase genes as they compare to the Norway spruce (+)-3-carene synthase and to Sitka spruce pinene synthase for reference. There are 33 amino acid differences unique to the sabinene synthases in this alignment. Of these 33, 18 were conserved changes (i.e. amino acids unique to the sabinene synthases have similar chemistry to those in the 3-carene synthases) and 17 were non-similar changes. Only two of the observed changes appear to be in the active site according to the three-dimensional model based on the tertiary structure of the Mentha spicata limonene synthase; a change from glutamic acid (E) to lysine (K) at position 532 (amino acids after the start methionine), and a change from phenylalanine (F) to leucine (L) at position 599. Thus,   112 it is likely that these amino acids are involved in the change in product from primarily (+)-3- carene to primarily (+)-sabinene.       SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 1                                               50  MSVISIVPLASKSCLYKSLMSSTHELKALCRPIVTLGMCRRGKSVMASMS MSVISIVPLASNSCLYKSLMSSTHELKALCRPIATLGMCRRGKSVMASMS MSVISILPLASKSCLYKSLMSSTHELKALCRPIATLGMCRRGKSVMASKS MSVISIVPLASNSCLYKSLMSSTHELKALCRPIATLGMCRRGKSVMASMS MSVISIVPLASNSCLYKSLMSSTHELKALCRPIATLGMCRRGKSVMASMS MSVISIVPLASKPCLYKSFISSTHEPKALRRPISTVGLCRRAKSVTASMS MSVISIVPLASKPCLYKSFISFTHEPKALRRPISTVGLCRRAKSVTASMS MSVISIVPLVSKPCLYKSFISSTHEPKALRRPISTVGLCRRAKSVTASMS MALVSVAPMASRSCLHKSLSSSAHELKTICRTIPTLGMSRRGKSATPSMS    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 51           RRxxxxxxxxW                       100                                  ___________    _ TGLTTAVSDDGVQRRIGDHHSNLWDDNFIQSLSSPYRASSYGETTNKLIG TSLTTAVSDDGVQRRIGHHHSNLWDDNFIQSLSSPYGASSYAESAKKLIG TSLTTAVSDDGVQRRIGDHHSNLWDDNFIQSLSSPYGASSYGERAERLIG TSLTTAVSDDGVQRRIGHHHSNLWDDNFIQSLSSPYGASSYAESAKKLIG TSLTTAVSDDGVQRRIGHHHSNLWDDNFIQSLSSPYGASSYAESAKKLIG MSSSTALSDDGVQRRIGNHHSNLWDDNFIQSLSSPYGASSYAERAERLIG MSSSTALSDDGVQRRIGNHHSNLWDDNFIQSLSSPYGASSYAERAERLIG MSSSTALSDDGVQRRIGNHHSNLWDDNFIQSLSSPYGASSYAERAERLIG MSLTTTVSDDGVQRRMGDFHSNLWNDDFIQSLSTSYGEPSYRERAERLIG    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 101                                            150  EVKEIFNSLSMADGGLMSPVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNSLSMAAGGLMSPVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNSLSRTDGELVSHVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNSLSMAAGGLMSPVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNSLSMAAGGLMSPVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNRISMANGELVSHVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNRISMANGELVSHVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKEIFNRISMANGELVSHVDDLLQHLSMVDNVERLGIDRHFQTEIKVSL EVKKMFNSMSSEDGELISPHNDLIQRVWMVDSVERLGIERHFKNEIKSAL   SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 151                                            200  DYVYSYWSEKGIGSGRDIVCTDLNTTALGFRILRLHGYTVFPDVFEHLKD DYVYSYWSEKGIGSGRDIVCTDLNTTALGFRILRLHGYTVFPDVFEHFKD DYVYSYWSEKGIGSGRDIVCTDLNTTALGFRILRLHGYTVFPDVFEHFKD DYVYSYWSEKGIGSGRDIVCTDLNTTALGFRILRLHGYTVFPDVFEHFKD DYVYSYWSEKGIGSGRDIVCTDLNTTALGFRILRLHGYTVFPDVFEHFKD DYVYSYWSEKGIGPGRDIVCADLNTTALGFRLLRLHGYTVFPDVFEQFKD DYVYSYWSEKGIGPGRDIVCADLNTTALGFRLLRLHGYTVFPDVFEQFKD DYVYSYWSEKGIGPGRDIVCADLNTTALGFRVLRLHGYTVFPDVFEQFKD DYVYSYWSEKGIGCGRESVVADLNSTALGFRTLRLHGYAVSADVLNLFKD   113     SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 201                                            250  QMGRIACSANHTERQISSILNLFRASLIAFPGEKVMEEAEIFSATYLKEA QMGRIACSANHTERQISSILNLFRASLIAFPGEKVMEEAEIFSATYLKEA QMGRIACSDNHTERQISSILNLFRASLIAFPGEKVMEEAEIFSATYLKEA QMGRIACSANHTERQISSILNLFRASLIAFPGEKVMEEAEIFSATYLKEA QMGRIACSANHTERQISSILNLFRASLIAFPGEKVMEEAEIFSATYLKEA QMGRIACSTNQTERQISSILNLFRASLIAFPWEKVMEEAEIFSTAYLKEA QMGRIACSTNQTERQISSILNLFRASLIAFPWEKVMEEAEIFSTAYLKEA QMGRIACSANQTERQISSILNLFRASLIAFPWEKVMEEAEIFSTAYLKEA QNGQFACSPSQTEEEIRSVLNLYRASLIAFPGEKVMEEAEIFSAKYLEES    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 251                                           300           •            _   ___  _      ▼     ▼   ▼ LQTIPVSSLSQEIQYVLQYRWHSNLPRLEARTYIDILQENTKNQMLDVNT LQTIPVSSLSQEIQYVLQYRWHSNLPRLEARTYIDILQENTKNQMLDVNT LQTIPVSSLSQEIQYVLQYRWHSNLPRLEARTYIDILQENTKNQMLDVNT LQTIPVSSLSQEMQYVLDYRWHSNLPRLETRTYIDILGETTINQMQDVNI LQTIPVSSLSQEMQYVLDYRWHSNLPRLETRTYIDILGETTINQMQDVNI LQTIPVSSLSREIQYVLDYRWHSDLPRLETRTYIDILRENATNETLDMKT LQTIPVSSLSREIQYVLDYRWHSDLPRLETRTYIDILRENATNETLDMKT LQTIPVSSLSREIQYVLDYRWHSDLPRLETRTYIDILRENATNETLDMKT LQKISVSSLSQEIRDVLEYGWHTYLPRMEARNHIDVFGQDTQNSKSCINT    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 301                                            350 ▼               •  __  _• ___ _____•________•_____ EKVLELAKLEFNIFHSLQQNELKSVSRWWKDSGFPDLNFIRHRHVEFYTL EKVLELAKLEFNIFHSLQQNELKSVSRWWKDSGFPDLNFIRHRHVEFYTL KKVLELAKLEFNIFHSLQQNELKSVSRWWKESGFPDLNFIRHRHVEFYTL QKLLELAKLEFNIFHSIQQNELKCISRWWKESGSPELTFIRHRHIEFYTL QKLLELAKLEFNIFHSIQQNELKCISRWWKESGSPELTFIRHRHIEFYTL EKLLELAKVEFNIFNSLQQNELKCVSRWWKESGSPDLTFIRHRQVEFYTL EKLLELAKVEFNIFNSLQQNELKCVSRWWKESGSPDLTFIRHRQVEFYTL EKLLELAKVEFNIFNSLQQNELKCVSRWWKESGSPDLTFIRHRQVEFYTL EKLLELAKLEFNIFHSLQKRELEYLVRWWKDSGSPQMTFCRHRHVEYYTL    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 351                        DDxxD               400 ▼_____   _ ▼_________________•______• ___ __• _  _ VSGIDMEPKHSTFRLSFVKMCHLITVLDDMYDTFGTIDELRLFTAAVKRW VSGIDMEPKHSTFRLSFVKMCHLITVLDDMYDTFGTIDELRLFTAAVKRW VSGIDMEPKHCTFRLSFVKMCHLITVLDDMYDTFGTIDELRLFTAAVKRW ASGIDMEPKHSAFRLSFVKMCHLITVLDDIYDTFGTMDELRLFTSAVKRW ASGIDMEPKHSAFRLSFVKMCHLITVLDDIYDTFGTMDELRLFTSAVKRW VSGIDMEPKRSTFRINFVKICHFVTILDDMYDTFGTIDELRLFTAAVKRW VSGIDMEPKRSTFRINFVKICHFVTILDDMYDTFGTIDELRLFTAAVKRW VSGIDMEPKRSTFRINFVKICHFVTILDDMYDTFGTMDELRLFTAAVKRW ASCIAFEPQHSGFRLGFAKACHIITILDDMYDTFGTVDELELFTAAMKRW           114   SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 401                                            450  • ▼▼     __ ___ ___ __  _    ▼         ___▼______ DPSTTQCLPEYMKGVYIVLYETVNEMAKEAQKSQGRDTLNYVRQALEAYI DPSTTQCLPEYMKGVYIVLYETVNEMAKEAQKSQGRDTLNYVRQALEAYI DPSTTECLPEYMKGVYTVLYETVNEMAQEAQKSQGRDTLSYVRQALEAYI DRSEIECLPEYMKGVYIILYETVNEMAREARKSQGRDTLNYARLALEEYI DRSEIECLPEYMKGVYIILYETVNEMAREARKSQGRDTLNYARLALEDYI DKSATECLPEYMKGVYIDLYETVNELAREAYKSQGRDTLNYARQALEDYL DKSATECLPEYMKGVYIDLYETVNELAREAYKSQGRDTLNYARQALEDYL DKSATECLPEYMKGVYIDLYETVNELAREAYKSQGRDTLNYARQALEDYL DPSAADCLPEYMKGVYLILYDTVNETSREAEKAQGRDTLDYARRAWDDYL    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 451                                            500 ___________▼__________▼________________ ▼ GAYHKEAEWISTGYLPTFDEYFENGKASSGHRIATLQPTFMLDIPFPHHI GAYHKEAEWISTGYLPTFDEYFENGKASSGHRIATLQPTFMLDIPFPHHI GAYHKEAEWISSGYLPTFDEYFENGKVSSGHRIATLQPTFMLDIPFPHHV GAYLKEAEWISMGYLPTFEEYFKNGKVSSGHRIATLQPILTLDIPFPHHI GAYLKEAEWISMVYLPTFEEYFKNGKVSSGHRIATLQPILTLDIPFPHHI GSYLKEAEWISTGYIPTFEEYLENGKVSSAHRIATLQPILMLDVPFPPHV GSYLKEAEWISTGYIPTFEEYLVNGKVSSAHRIATLQPILMLDVPFPPHV GSYLKEAEWISTGYIPTFEEYLENGKVSSAHRIATLQPILMLDVPFPPHV DSYMQEAKWIATGYLPTFAEYYENGKVSSGHRTSALQPILTMDIPFPPHI    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 501                                            550       _____•_________________ _▼___▼_______ LQEIDFPSKFNDFACSILRLRGDTRCYQADMARGEEASCISCYMKDNPGS LQEIDFPSKFNDFACSILRLRGDTRCYQADMARGEEASCISCYMKDNPGS LQEIDFPSKFNDFACSILRLRGDTRCYQADRARGEEASCISCYMKDNPGS LQEIDFPSKFNELACSILRLRGDTRCYQADRDRGEKASCISCYMKDNPGS LQEIDFPSKFNELACSILRLRGDTRCYQADRDRGEKASCISCYMKDNPGS LQEIDFPSKFNDLAGSILRLRGDTRCYQNDRARGEEASCISCYMKDNPGS LQEIDFPSKFNDLAGSILRLRGDTRCYQNDRARGEEASCISCYMKDNPGS LQEIDFPSKFNDLAGSILRLRGDTRCYQNDRARGEEASCISCYMKDNPGS LKEVDFPSKLNDLASAILRLRGDTRCYKADRARGEEASSISCYMKDNPGA    SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 551                                            600 _  __  __ ___ __         •         ___•_____•▼__▼_ TQEDALNHINNMIEETIKKLNRELLKPDNNVPISSKKHAFDISRGLHHFY TQEDALNHINNMIEETIKKLNRELLKPDNNVPISSKKHAFDISRGLHHFY TQEDALNHINNMIEETIKKLNWELLKPDNNVPISSKKHAFDINRGLHHFY TEEDALNHINGMIEDTIKQLNWELLRPDNNVPISSKKHSFDISRAFHHLY TEEDALNHINGMIEDTIKQLNWELLRPDNNVPISSKKHSFDISRAFHHLY TEEDALNHINGMIEKQIKELNWELLKPDKNVPISSKKHAFNISRGLHHFY TEEDALNHINGMIEKQIKELNWELLKTDKNVPISSKKHAFNISRGLHHFY TEKEALNHINGMIEKRIKELNWELLKPDKNVPISSKKHAFNISRGLHHFY TEEDALDHINAMISDVIRGLNWELLNPNSSVPISSKKHVFDISRAFHYGY             115   SS H898 [H02] carene SS H898 [H08] carene NS [JF67] carene SS H898 [H05] sabinene SS Q903 [Q05] sabinene SS Q903 [Q09] carene SS FB3425 [I02] carene IS [F08] carene SS pinene synthase 601                      628 ________•____•___••_    • NYRDGYTVASNETKNLVIKTVLEPVPM- NYRDGYTVASNETKNLVIKTVLEPVPM- NYRDGYTVASNETKNLVIKTVLEPVPM- RYRDGYTVSSNETKNLVVRTVLEPLPM- RYRDGYTVSSNETKNLVVRTVLEPLPM- KYRDGYTVANSETRNLVIKTVLEPVPM- KYRDGYTVANSETRNLVIKTVLEPVPM- KYRDGYTVANSETRNLVIKTVLEPVPM- KYRDGYSVANIETKSLVRRTVIDPVTL-  Figure 4-4 The amino acid alignment of the functionally characterized 3-carene synthase and sabinene synthase enzymes. The amino acids that differ between the 3- carene synthases and the sabinene synthases are marked with either a ‘•’ indicating a conservative amino acid change or a ‘▼’ indicating a non-similar amino acid change. Those amino acids that are within 20Å of the substrate in the modeled three-dimensional structure are indicated by the pink bar above the alignment. The white arrows in the pink bar represent those amino acids that appear to be directly involved with the substrate in the active site.     The sequence comparisons revealed that I have cloned and functionally characterized three groups of (+)-3-carene-like TPS from the cDNA of both tree genotypes encoding either (+)-3-carene synthases or sabinene synthases: Q09 (+)-3-carene synthase (referred to as CAR_1), H02/H08 (+)-3-carene synthase (referred to as CAR_2), H05/Q05 (+)-sabinene synthase (referred to as SAB).  4.3.3 UTR groupings identify a fourth (+)-3-carene synthase-like sequence: CAR_3 3’-RACE sequencing results confirm that there are multiple copies of (+)-3-carene synthase-like genes in the cDNA pool from both genotypes. The UTR sequences fall into four groups that align with Q09-like (+)-3-carene synthases (CAR_1), H02/H08-like (+)-3- carene synthases (CAR_2), and with HQ05 (+)-sabinene synthases (SAB) as well as a new group called identified in Q903 cDNA that I called Q270 or CAR_3. Although I was able to   116 functionally characterized three groups of (+)-3-carene synthases, I identified the sequence of a fourth (+)-3-carene synthase as well.  4.3.4 Comparison of (+)-3-carene synthases in H898 and Q903 Sitka spruce genotypes Experiments were designed to investigate the gene expression, enzyme activity and terpene profile differences between H898, a highly resistant genotype that produces (+)-3- carene, and Q903, a susceptible genotype that produces only trace amounts of (+)-3- carene, in response to methyl jasmonate treatment (a strong defense response elicitor).  I designed primers to conduct qRT-PCR for the three functionally-characterized (+)-3- carene synthase-like groups (CAR_1, CAR_2, and SAB) as well as for the putative (+)-3- carene synthase sequence (CAR_3). The amplification of gene transcripts for each of the four (+)-3-carene synthase-like groups was successful, and the genes were differentially expressed between the two tree genotypes (Figure 4-5).   117 0 10 20 30 0 0.3 0.6 0.9 1.2 Fo ld  c ha ng e 0 10 20 30 0 10 20 30 0 10 20 30 0 10 20 30 0 10 20 30 Days after treatment 0 10 20 30 CAR_2 0 10 20 30 CAR_3 0 10 20 30 SAB 0 10 20 30 PIN 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 CAR_1 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 0 0.3 0.6 0.9 1.2 nd nd Methyl jasmonate treatment Tween control H898 Q903 * * * * * * * * * * * * * * * * * * * * Fo ld  c ha ng e Fo ld  c ha ng e  Figure 4-5 Gene expression patterns for the (+)-3-carene synthases, the (+)-sabinene synthase and (−)-α-pinene synthase in resistant and susceptible tree genotypes. The average gene expression fold change (± 1SE) over the translation initiation factor (TIF) reference gene expression for each (+)-3-carene synthase (CAR_1, CAR_2, CAR_3) and the (+)-sabinene synthase (SAB) gene sequences using the α-pinene synthase (PIN) gene expression for comparison. An asterisk denotes a significant difference (p < 0.05) between methyl jasmonate treatment and the tween control. Transcripts of CAR_2 were not detectable (nd) in Q903, and transcripts of CAR_3 were not detectable (nd) in H898.   118  All three (+)-3-carene synthases (CAR_1, CAR_2, and CAR_3), the sabinene synthase (SAB), as well as the α-pinene synthase gene (PIN) were induced by the methyl jasmonate treatment. CAR_1, SAB and PIN were expressed in both genotypes whereas CAR_2 transcripts were detected only in H898 and CAR_3 transcripts were detected only in Q903. Samples taken on day 2 after methyl jasmonate treatment showed the largest increase in relative gene expression (Figure 4-5). CAR_1 was induced in both genotypes. In H898, CAR_1 was significantly induced with methyl jasmonate at day 2 (F1,5 = 781.73, P < 0.001) and day 7 (F1,5 = 31.12, P = 0.003), and in Q903, CAR_1 was also significantly induced on day 2 (F1,5 = 36.07, P = 0.001) and day 7 (F1,5 = 11.41, P = 0.015), as well as at day 21 (F1,5 = 107.63, P < 0.001) and day 28 (F1,5 = 20.04, P = 0.004). CAR_2 was significantly induced only in H898 on day 2 (F1,5 = 16.71, P = 0.006). CAR_2 was not detected above background levels in Q903. CAR_3 showed the opposite pattern, it was not detected in H898, but showed significant induction in Q903 on day 2 (F1,5 = 7.75, P = 0.032), day 7 (F1,5 = 119.68, P < 0.001), and day 14 (F1,5 = 29.73, P = 0.002). The SAB gene was induced by methyl-jasmonate treatment in both genotypes. In H898, SAB was significantly induced at only on day 2 (F1,5 = 26.84, P = 0.002), whereas in Q903, SAB was significantly induced on day 2 (F1,5 = 13.58, P = 0.010), day 7 (F1,5 = 30.73, P = 0.001), and day 14 (F1,5 = 23.50, P = 0.003). Finally, PIN was also induced in both tree genotypes. In H898, PIN was significantly induced on day 2 (F1,5 = 40.99, P = 0.001) and day 14 (F1,5 = 6.67, P = 0.042), day 7 was significant only at p <  0.10 (F1,5 = 5.887, P = 0.060). In Q903, PIN was significantly induced on day 2 (F1,5 = 22.92, P = 0.003), day 7 (F1,5 = 12.03, P = 0.013), day 14 (F1,5 = 9.44, P = 0.022).    119 4.3.5 Comparison of monoterpene synthase enzyme activities in H898 and Q903 Cell-free enzyme assays (Figure 4-6) conducted with GPP as the substrate showed methyl jasmonate-induced increases in the formation of (−)-α-pinene, peaking on sampling day 7 in both genotypes (H898: F1,6 = 23.54, P = 0.003; Q903: F1,6 = 6.52, P = 0.043) and on day 14 in H898 (F1,6 = 14.94, P = 0.012). The enzyme assays showed the formation of only trace amounts of (+)-sabinene or (+)-3-carene in both genotypes and the formation of these compounds in enzymes assays did not differ between treatment and control trees.  4.3.6 Comparison of monoterpene metabolite profiles in H898 and Q903 Metabolite extracts (Figure 4-6) showed significant increases in (−)-α-pinene in methyl jasmonate-treated H898 on day 7 (F1,6 = 8.61, P = 0.026), day 21 (F1,6 = 13.62, P = 0.010), and day 28 (F1,6 = 6.18, P = 0.047), and in Q903 on day 14 (F1,6 = 6.76, P = 0.041), and day 28 (F1,6 = 6.73, P = 0.041). Both genotypes showed a significant increase in (+)-sabinene in the methyl jasmonate-treated trees on day 7 (H898: F1,6 = 6.65, P = 0.042; Q903: F1,6 = 19.98, P = 0.004), day 14 (H898: F1,6 = 11.38, P = 0.015; Q903: F1,6 = 266.35, P < 0.001), day 21 (H898: F1,6 = 18.8, P = 0.005; Q903: F1,6 = 14.50, P = 0.009), and day 28 (H898: F1,6 = 614.08, P = 0.009; Q903: F1,6 = 104.92, P < 0.001). Although levels of (+)-3-carene extracted from H898 remained high (for example the day 7 average ± 1SE for the control was 960.15 ± 97.5 µg/g dry weight and for the methyl jasmonate treatment was 1312.54 ± 337.2 µg/g dry weight), there was no significant difference between the amount of (+)-3-carene between the treatment and control. Although I detected only very small amounts of (+)-3-carene in Q903 (the day 7 average ± 1SE for the Q903 control was 7.17 ± 0.9 µg/g dry weight and for the methyl jasmonate treatment was 12.51 ± 1.6 µg/g dry weight), the difference in (+)-3-carene   120 between treated and control trees was still significant (see Figure 4-6) on day 7 (F1,6 = 7.88, P = 0.031), day 14 (F1,6 = 8.90, P = 0.025), day 21 (F1,6 = 6.28, P = 0.046), and day28 (F1,6 = 21.98, P = 0.003).   121   0 10 20 30 0 50 100 150 (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) 0 10 20 30 0 1000 2000 3000 4000 (- )- α- pi ne ne  (u g/ d dr y w t) 0 10 20 30 Days after treatment 0 50 100 150 0 10 20 30 0 1000 2000 3000 4000 H898 Q903 Enzyme product Metabolite extract Q903 H898 (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) (- )- α- pi ne ne  (u g/ d dr y w t) 0 10 20 30 0 50 100 150 0 10 20 30 0 50 100 150 (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) Days after treatment H898 Q903 Enzyme product Metabolite extract Q903 H898 0 10 20 30 0 1000 2000 3000 4000 0 10 20 30 0 1000 2000 3000 4000 (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) 0 10 20 30 0 50 100 150 0 10 20 30 0 1000 2000 3000 4000 0 10 20 30 0 50 100 150 0 10 20 30 0 1000 2000 3000 4000 (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t) Days after treatment H898 Q903 Enzyme product Metabolite extract Q903 H898 (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t) Q903 H898 (-)-α-pinene (+)-sabinene (+)-3-carene **** * * * * * * * * * * * * * * * * (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) (- )- α- pi ne ne  (u g/ d dr y w t) (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) (- )- α- pi ne ne  (u g/ d dr y w t) (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t) (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t) (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) (- )- α- pi ne ne  (u g/ d dr y w t) (- )- α- pi ne ne  (n g/ 10 0u g en zy m e) (- )- α- pi ne ne  (u g/ d dr y w t) (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) (+ )- sa bi ne ne (n g/ 10 0u g en zy m e) (+ )- sa bi ne ne (u g/ d dr y w t) (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t) (+ )- 3- ca re ne (n g/ 10 0u g en zy m e) (+ )- 3- ca re ne (u g/ d dr y w t)  Figure 4-6 Enzyme product and metabolite extract for (−)-α-pinene, (+)-sabinene, and (+)-3-carene for resistant and susceptible tree genotypes. The average amount of product (ng ± 1SE) per 100 µg of total protein extract and the average amount (µg/g dry weight ± 1SE) of extracted compound for the susceptible genotype Q903 and the resistant genotype H898. An asterisk denotes a significant difference (p < 0.05) between methyl jasmonate treatment and the tween control. Although metabolite extract levels for (+)-3-carene and (+)-sabinene are present in H898 and present in trace amounts in Q903, enzyme products are essentially not detectable in enzyme assays. Enzyme assays and metabolite extractions showed the presence of (−)-α-pinene.   122 In summary, I identified four (+)-3-carene synthase-like genes within two genotypes of Sitka spruce. I functionally characterized three of these candidates and found that they are multi-product enzymes with either (+)-3-carene or (+)-sabinene as the major products. The results of the qRT-PCR analysis for each of these genes reveals that although the Q903 genotype produces only trace amounts of (+)-3-carene, it possesses and transcribes genes capable of producing this compound. The (+)-3-carene synthase genes that are expressed in Q903 (CAR_1 and CAR_3), however, are different than those expressed in H898 (CAR_1 and CAR_2), the resistant genotype. Finally, the amount of (+)-3-carene produced in total protein extracts from either genotype were much lower than expected and do not account for the high levels of (+)-3-carene found in tissue extracts from H898.  4.4 Discussion A diverse array of specialized terpenoid metabolites has evolved as an important component of conifer defense. Conifers, including Sitka spruce, are ancient and extremely long-lived organisms, often living over 250 years, that are literally rooted in one spot. An individual tree may be exposed to many generations of attacking pests and pathogens. A large and highly variable defense strategy is more likely to out-maneuver relatively quickly-evolving attacking organisms. The conifer genome is estimated to be 200-400 fold larger than the Arabidopsis genome (Hamberger et al. 2009; Bousquet et al. 2007; Murray 1998), and a number of large multi-gene families have been identified in conifers (Martin et al. 2004; Raj Ahuja and Neale 2005; Hamberger and Bohlmann 2006). The diversity of the (+)-3-carene synthase-like genes is representative of the large terpene synthase TPSd family and may be an advantage in facing the unique challenges of an ancient, long-lived, and stationary organism.    123 In this study, I uncovered a new level of genetic diversity within an already extremely diverse family of terpene synthase genes. Gene duplication and neofunctionaliziation appear to be important in the enormous sequence diversity of terpene synthases (Bohlmann et al. 1998a; Trapp and Croteau 2001; Martin et al. 2004). Aubourg et al. (2002) found evidence of repeated gene duplication events in the terpene synthases that produce secondary metabolites in Arabidopsis thaliana. Although Arabidopsis terpene synthase genes are spread across the five chromosomes, 18 of the 40 Arabidopsis genes identified as terpene synthases were encompassed by six clustered locations on the Arabidopsis chromosomes. Prenyltransferases, the enzymes responsible for the synthesis of the precursor for terpene formation, were found clustered with terpene synthase genes suggesting that sections of the genome containing more than one step in the pathway were duplicated as well (Aubourg et al. 2002). Plomion et al. (1996) found that the 3-carene phenotype is largely accounted for by one large quantitative trait locus in maritime pine (Pinus pinaster), and thus it is possible that multiple (+)-3-carene synthases are grouped in one location. More recently, targeted sequencing of genomic BAC DNA clones containing a (+)-3-carene synthase-like sequence showed only a single gene in 172 kbp demonstrating a relatively low gene density compared to the 4.5 kbp per gene in Arabidopsis (Hamberger et al. 2009). Hamberger et al. (2009) did not find evidence of local tandem gene duplications and suggest that whole genome or chromosome segmental duplications may be responsible for terpene synthase gene family diversity in conifers.  In addition to increased genetic diversity of terpene synthases, my study confirms that sequence information alone is not a reliable predictor of terpene synthase product profile (Bohlmann et al. 1997; Keeling et al. 2008). A few amino acid changes resulted in an entirely different product profile between highly similar (+)-3-carene synthase-like   124 gene sequences in Sitka spruce. From the amino acid alignment, there are only a few amino acids within 20 Å of the predicted active site that were different between the functionally characterized sabinene synthase and (+)-3-carene synthase sequences. Two non-similar changes, a change from positively charged glutamic acid (E) to negatively charged lysine (K) and a change from the aromatic phenylalanine (F) to aliphatic leucine (L), may alter the conditions of the active site and result in the formation of sabinene over (+)-3-carene. Keeling et al. (2008) showed that one amino acid substitution was sufficient to completely change the product profile of a levopimaradiene / abietadiene synthase, a diterpene synthase, to the production of isopimaradiene and sandaracopimaradiene.  4.4.1 Enzyme activity and terpene profile: an area for further research. The differential (+)-3-carene phenotype between H898 and Q903 demonstrates the complex nature of Sitka spruce terpene profiles; single genes producing predominantly one compound are not simply turned on or off, but each terpene compound may be produced by multiple enzymes as either the major product or as minor products (Martin et al. 2004; Keeling et al. 2008). In addition, duplicated copies producing the same major products may vary in expression levels, enzyme activity, or enzyme efficiency. Although it produces only trace amounts of (+)-3-carene, the Q903 Sitka spruce genotype expresses genes capable of producing (+)-3-carene as the major compound in vitro and the transcript expression levels are induced with methyl jasmonate treatment. I suggest that the lack of a capability to produce (+)-3-carene occurs as a result of reduced enzyme activity / kinetic efficiency in the enzymes that are primarily expressed in Q903 (Q09, CAR_3). Recent data suggests that the kinetic efficiency of this form of (+)-3- carene synthase may be lower than the forms expressed in H898 (Hall D, Robert J, Keeling CI, Jancsik S, Madilao L, Lara Quesada A, Hamberger B, and Bohlmann J,   125 unpublished results). It is possible that degenerative changes to duplicated (+)-3-carene synthase-like genes occurred in Q903 without evolutionary consequence as this genotype originates in an area with no history of weevil attack (there are no weevils currently on the Queen Charlotte Islands).  The enzyme activities measured in the cell-free extracts from tree tissues for the (+)- 3-carene synthase-like enzymes do not account for the (+)-3-carene and sabinene phenotypes identified from the metabolite extractions. It is possible that although the enzyme assays showed activity and induction for the (−)-α-pinene synthase, the assays were not optimal for (+)-3-carene synthesis. (+)-3-Carene synthase-like enzymes may be poor competitors for substrate in a cell-free extract, or transcript expression may not result in translation of active (+)-3-carene synthase enzymes. Recent selective reaction monitoring (SRM) for protein identification in Sitka spruce extracts shows that CAR_1, CAR_2, CAR_3, and SAB enzymes are produced and occur in tree tissues (Hall et al., unpublished results). Further experiments to calculate enzyme kinetic parameters (kcat, km, and catalytic efficiency) and to assess substrate competition are underway to elucidate the low production of (+)-3-carene in the cell-free enzyme assays (Hall et al., unpublished results). Alternatively, constant, low levels of (+)-3-carene synthase activity could account for the high levels of constitutive (+)-3-carene accumulating in H898.  4.5 Conclusions Sitka spruce are extremely long-lived, ancient organisms with very large genomes. If plants indeed possess a “basic genetic toolkit of ancient origin” (Flagel and Wendel 2009), then it stands to reason that an old and long-lived species would have highly sophisticated versions of this basic toolkit. The occurrence of multiple (+)-3-carene synthase genes indicates that the terpene synthase gene family may be larger than   126 previously thought. In addition, alternative forms of duplicated terpene synthases are differentially expressed between tree genotypes. If duplicated versions encode for enzymes with different activities, then the large variation in terpene profiles may be due to a number of factors in addition to the presence and number of genes. Determining enzyme kinetic factors and the localization of (+)-3-carene synthases within the tree stem and within the cell, would help to further illuminate the phenotypic difference in (+)- 3-carene production between H898 and Q903. These questions suggest another layer of complexity when breeding for resistance and explains, in part, why the mechanisms behind the production of terpene compounds, the role of terpenoids in tree defense, and the inheritance of terpenoids remain so difficult to predict.    127 4.6 References  Alfaro RI. 1995. An induced defense reaction in white spruce to attack by the white pine weevil, Pissodes strobi. Can. J. For. Res. 25(10):1725-1730.  Alfaro RI, King JN, Brown RG, and Buddingh SM. 2008. Screening of Sitka spruce genotypes for resistance to the white pine weevil using artificial infestations. For. Ecol. Manage. 255:1749-1758.  Aubourg S, Lecharny A, and Bohlmann J. 2002. Genomic analysis of the terpenoid synthase (AtTPS) gene family of Arabidopsis thaliana. Mol Genet Genomics. 267:730-745.  Bichao H, Borg-Karlson AK, Araujo J, and Mustaparta H. 2003. Identification of plant odours activating receptor neurons in the weevil, Pissodes notatus. (Coleoptera, Curculionidae) J.Comp. Physiol. 189:203-212.  Bohlmann J, Meyer-Gauen G, and Croteau R. 1998. Plant terpenoid synthases: Molecular biology and phylogenetic analysis. Proc. Natl. Acad. Sci. USA. 95:4126- 4133.  Bohlmann J, Steele CL, and Croteau R. 1997. Monoterpene synthases from Grand Fir (Abies grandis). J. Biol Chem. 272:21784-21792.    128 Bousquet J, Isabel N, Pelgas B, Cottrell J, Rungis D, and Ritland K. 2007. Chapter 3: Spruce In Genome mapping and molecular breeding in plants, Volume 7: Forest trees. Edited by Kole C. Springer-Verlag: Berlin Heidelberg. pp. 93-114.  Chang S, Puryear J, and Cairney J. 1993. A simple and efficient method for isolating RNA from pine trees. Plant Mol. Biol. Reptr. 11(2):113.  Dhillon RS, Gautam VK, Singh S, and Singh J. 1991. From 3-carene to the most biologically active insecticides – a convenient and efficient synthesis of (+)-(1R)-cis- chrysanthemic acid. Indian Journal of Chemistry Section B – Organic Chemistry including Medicinal Chemistry 30:574-578.  Fäldt JF, Martin D, Miller B, Rawat S, and Bohlmann J. 2003. Traumatic resin defense in Norway spruce (Picea abies): Methyl jasmonate-induced terpene synthase gene expression, and cDNA cloning and functional characterization of (+)-3-carene synthase. Plant Mol. Biol. 51:119-133.  Flagel LE and Wendel JF. 2009. Gene duplication and evolutionary novelty in plants. New Phytol. 183:557-564.  Guex N and Peitsch MC. 1997. SWISS-MODEL and the Swiss-PdbViewer: an environment for comparative protein modeling. Electrophoresis. 18(15):2714-2723.      129 Hamberger B, Hall D, Yuen M, Oddy C, Hamberger B, Keeling CI, Ritland C, Ritland K, and Bohlmann J. 2009. Targeted isolation, sequence assembly and characterization of two white spruce (Picea glauca) BAC clones for terpenoid synthase and cytochrome P450 genes involved in conifer defense reveal insights into a conifer genome. BMC Plant Biol.  Humble LM, Humphreys N, and Van Sickle GA. 1994. Distribution and hosts of the white pine weevil, Pissodes strobi (Peck), in Canada. In: The White Pine Weevil: Biology, Damage and Management, Proceedings of a Meeting, 19-21 Jan. 1994., Richmond, BC Edited by R.I. Alfaro, G. Kiss, and R.G. Fraser. Can. For. Serv. Pac. For. Cent. FRDA Rep. 226. pp. 68-75.  Hyatt DC, Youn B, Zhao Y, Santhamma B, Coates RM, and Croteau RB. 2007. Structure of limonene synthase, a simple model for terpenoid cyclase catalysis. Proc. Natl. Acad. Sci. USA 104(13):5360-5365.  Keeling CI, Weisshaar S, Lin RPC, and Bohlmann J. 2008. Functional plasticity of paralogous diterpene synthases involved in conifer defense. Proc. Natl. Acad. Sci. USA 105:1085-1090.  King JN and Alfaro RI. 2009. Developing Sitka spruce populations for resistance to the white pine weevil: summary of research and breeding program. B.C. Min. For. Range, For. Sci. Prog., Victoria, B.C. Tech. Rep. 050. www.for.gov.bc.ca/hfd/pubs/Docs/Tr/Tr050.htm.    130 Kolosova N, Miller B, Ralph S, Ellis BE, Douglas C, Ritland K, and Bohlmann J. 2004. Isolation of high quality RNA from gymnosperm and angiosperm trees. Biotechniques 36:821-824.  LaFever RE, Stofer Vogel B, Croteau R. 1994. Diterpenoid resin acid biosynthesis in conifers: enzymatic cyclization of geranyl geranyl pyrophosphate to abietadiene, the precursor of abietic acid. Arch. Biochem. Biophys. 313:139-149.  Lastbom L, Boman A, Johnsson S, Camner P, and Ryrfeldt A. 2003. Increased airway responsiveness of a common fragrance component, 3-carene, after skin sensitization – a study in isolated guinea pig lungs. Toxicol. Letters 145:189-196.  Lewinsohn E, Gijzen M, Savage TJ, and Croteau R. 1991. Defense mechanisms of conifers: Relationship of monoterpene cyclase activity to anatomical specialization and oleoresin monoterpene content. Plant Physiol. 96:38-43.  Lewinsohn E, Savage T, Gijzen M, and Croteau R. 1993. Simultaneous analysis of monoterpenes and diterpenoids of conifer oleoresin. Phytochem. Anal. 4:220-225.  Martin D, Fäldt J, and Bohlmann J. 2004. Functional characterization of nine Norway spruce TPS genes and evolution of gymnosperm terpene synthases of the TPS-d subfamily. Plant Physiol. 135:1908-1927.  Martin D and Bohlmann J 2005. Molecular biochemistry and genomics of terpenoid defenses in conifers. Rec. Adv. Phytochem. 39:29-56.    131 Martin D, Tholl D, Gershenzon J, Bohlmann J. 2002. Methyl jasmonate induces traumatic resin ducts, terpenoid resin biosynthesis, and terpenoid accumulation in developing xylem of Norway spruce stems. Plant Physiol. 129:1003-1018.  Miller B, Madilao LL, Ralph S, and Bohlmann J. 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 putative octadecanoid pathway transcripts in Sitka spruce. Plant Phys. 137:369-382.  Murray BG. 1998. Nuclear DNA amounts in gymnosperms. Ann. Bot. 82(A):3-15.  Overhulser DL and Gara RI. 1981. Site and host factors affecting the Sitka spruce weevil, Pissodes strobi, in western Washington. Envir. Entomol. 10:611-614.  Phillips MA, Bohlmann J, and Gershenzon J. 2006. Molecular regulation of induced terpenoid biosynthesis in conifers. Phytochem. Rev. 5:179-189.  Plomion C, Yani A, and Marpeau A. 1996. Genetic determinism of δ3-carene in maritime pine using RAPD markers. Genome. 39:1123-1127.  Ralph SG, Chun HJ, Kolosova N, Cooper D, Oddy C, Ritland CE, Kirkpatrick R, Moore R, Barber S, Holt RA, Jones SJ, Marra MA, Douglas CJ, Ritland K, Bohlmann J (2008) A conifer genomics resource of 200,000 spruce (Picea spp.) ESTs and 6,464 high-quality, sequence- finished full-length cDNAs for Sitka spruce (Picea sitchensis). BMC Genomics 9: 484    132 Raj Ahuja M and Neale DB. 2005. Evolution of genome size in conifers. Silvae Genetica 54(3):126-137).  Rieu I and Powers SJ. 2009. Real-time quantitative RT-PCR: design, calculations, and statistics. Plant Cell. 21:1031-1033.  Rocchini LA, Lindgren BS, and Bennett RG. 2000. Effects of resin flow and monoterpene composition on susceptibility of lodgepole pine to attack by the Douglas-fir pitch moth, Synanthedon novaroensis (Lep., Sesiidae). J. Appl. Entomol. 124:87-92.  Storer AJ and Speight MR. 1996. Relationships between Dendroctonus micans Kug. (Coleoptera: Scolytidae) survival and development and biochemical changes in Norway spruce, Picea abies (L.) Karst., phloem caused by mechanical wounding. J. Chem. Ecol. 22(3):559-573.  Trapp SC and Croteau RB. 2001. Genomic organization of plant terpene synthases and molecular evolutionary implications. Genetics. 158:811-832.  Udvardi MK, Czechowski T, and Scheible WR. 2008. Eleven golden rules of quantitative RT-PCR. The Plant Cell. 20:1736-1737.  Wang SX, Hunter W, and Plant A. 2000. Isolation and purification of functional total RNA from woody branches and needles of Sitka and White Spruce. Biotechniques 28(2):292.  Zeng G. 1998. Sticky-end PCR: New method for subcloning. Biotechniques 25:206-208.   133  Zhao S and Fernald RD. 2005. Comprehensive algorithm for quantitative real-time polymerase chain reaction. J. Comput. Biol. 12(8):1045-62.  Zulak KG, Lippert DN, Kuzyk M, Domanski D, Chou T, Borchers CH, and Bohlmann J. 2009. Targeted proteomics using selected reaction monitoring (SRM) reveals the induction of specific terpene synthases in a multi-level study of methyl jasmonate treated Norway spruce (Picea abies). Plant J. 60(6):1015-30.   134  5 GENERAL CONCLUSIONS In this thesis, I aimed to identify the role and to investigate the molecular-genetic underpinnings of the formation of terpene compounds in the Sitka spruce defense response to attack by white pine weevil. The three components of this thesis each illustrate a different aspect of tree defense: a broad-scale association study, an insect- based bioassay approach, and molecular level investigation for a specific set of monoterpene synthases. These approaches contributed substantial new information to our current understanding of the Sitka spruce defense response and opened up new questions to be explored in future work.  From the large-scale association study (Chapter 2), I determined that terpene profiles can be used to distinguish resistant tree genotypes. My analysis was conducted using a data set comprised of tree genotypes from widely varying locations of origin (from the Queen Charlotte Islands off of Canada’s north-west coast to Squamish, Vancouver Island, and Hoquiam in the northern United States), and yet the fitted resistance was still almost 92% accurate. Thus, creating discriminant functions from a larger data set could potentially predict the resistance group for parent trees from new areas without needing the weevil-augmented plantations currently used for screening new parent genotypes. Alternatively, a library of discriminant functions tailored to the current high weevil hazard areas will define the current range and variability of resistant terpene profiles.  Predicting the terpene profiles for trees in breeding programs  is currently difficult due to the high diversity of the TPSd (Martin et al. 2004) and CYP720B (Hamberger and Bohlmann 2006) gene families in conifers in combination with multiple potential levels of   135 regulation for the expression of genes and the function of enzymes. In addition, obtaining a sufficient number of samples to accurately describe the interplay of resistance characteristics in Sitka spruce is an ongoing challenge. Ultimately, a very large-scale study measuring all known resistance characteristics in all tree genotypes that have been rated for resistance (measuring trees contained in multiple clonebanks) could be highly beneficial for determining the full suite of biologically relevant resistance mechanisms. As climate change is likely to alter the current range limit for Sitka spruce (Aitken et al. 2008), and potentially expand the areas designated as high weevil hazard (Volney and Fleming 2000), it may be possible to identify new parent tree genotypes for breeding programs proactively and rapidly in order to replant genetically resilient trees of an economically valuable native spruce species.  I have also characterized weevil behavior and physiological response (feeding patterns, host choice, ovary development, egg laying behavior, and larval development) to highly resistant and susceptible tree genotypes (Chapter 3). My results suggest that the highly resistant genotype H898 has defense mechanisms that deter both male and female weevils during host selection and mating, that cause delayed ovary development in females, and prevent successful reproduction of weevils on H898 trees. Although I found an association of (+)-3-carene with resistance in Chapter 2, I did not see evidence that this compound is acting as a volatile headspace signal or deterrent. It is interesting to note, however, that (+)-3-carene is significantly reduced at 15 days in trees that were fed on by weevils. This is the same time period in which males lose their preference for susceptible trees. In addition, I showed that the attractive effect of pentane in females is neutralized by relatively low levels (0.01% by volume) of (+)-3-carene. Attractive mixtures of commercially available terpene compounds  based on the volatile emissions of susceptible trees (mixtures of monoterpenes and diterpene resin acids for example)   136 could be produced in order to test attractiveness or deterrence of compounds such as (+)-3-carene within the mixture. This approach may help to elucidate the role of (+)-3- carene in a complex mixture of terpenoid compounds comprising tree oleoresin. Finally, the role of this compound as a possible gustatory deterrent, especially in males, should be investigated. Follow-up studies using this compound, as well as terpinolene and dehydroabietic acid, in fumigation and feeding bioassays may help to identify any toxic or deterrent effects of these compounds on weevils (for example, identifying dose- response curves for both individual compounds and mixtures). Finally, our laboratory has recently developed RNAi lines of white spruce that show decreased production of (+)-3-carene. These trees would be ideal candidates for feeding bioassays where the physiological effects of the weevils could be monitored as well as changes in the tree’s production of (+)-3-carene and other terpene compounds in response to methyl jasmonate.  Lastly, I have identified the first (+)-3-carene and sabinene synthase genes in Sitka spruce (Chapter 4). These genes have very similar sequences, yet the encoded enzymes have different product profiles; this shows a new level of genetic diversity in this spruce terpene synthase defense gene family. In addition, different (+)-3-carene synthase genes are expressed in a resistant tree genotype that produces large amounts of (+)-3-carene versus a susceptible genotype that produces trace amounts of (+)-3- carene. The role of (+)-3-carene in the tree defense response is not straight-forward. Although the transcription of multiple (+)-3-carene synthases was induced, a measureable rise in (+)-3-carene concentration does not occur in tree tissue. Further exploring the enzyme kinetics and localization as well as the possible storage of (+)-3- carene in spruce tissues will provide relevant insight into the role and induction of terpene compounds in the Sitka spruce defense response.   137  The variation of individual terpenoid compounds, such as dehydroabietic acid, (+)-3- carene and terpinolene, among different genotypes of the same species can be explained, in principle, with current knowledge of the molecular and biochemical underpinnings of terpenoid formation in conifers (Keeling and Bohlmann 2006, Zulak and Bohlmann, 2010).  A diterpene synthase that forms relevant amounts of dehydroabietadiene, the precursor for dehydroabietic acid, has not yet been cloned at the cDNA or genomic level, but the conversion of dehydroabietadiene to dehydroabietic acid via dehydroabietadienol and dehydroabietadienal involves a multifunctional CYP720B4 gene which has recently been cloned from Sitka spruce (B. Hamberger, T. Ohnishi, J. Bohlmann, unpublished results). The cloning and characterization of the dehydroabietadiene synthase would be an important step towards uncovering the role of this compound in resistance of Sitka spruce. Followup studies to explore the location, induction, and timing of diterpene synthase gene expression in Sitka spruce could be compared, for example, to this study on (+)-3-carene leading to further insights into the dynamics of the induced terpene response to attack by weevils.  In our study, (+)-3-carene and terpinolene are associated with resistance only in the Haney region. The resistant genotypes from this region may express members of the (+)-3-carene synthase gene family that allow for increased production of these compounds. Whether this is responsible for some of the resistant phenotype can now be tested in future work using methods established for monitoring of individual TPS genes and proteins (Zulak et al. 2009). Allelic variants of these cDNA or candidate genes, and their gene frequencies will also need to be quantified in these populations in order to evaluate the associative effects they may have on phenotypic resistance.    138 This thesis project combined terpene profile information from more than 100 tree genotypes, the characterization of the insect response, and the molecular-level investigation of (+)-3-carene formation to form a picture of highly effective, natural tree resistance. This study is an important step in untangling the tree’s molecular response to attack by white pine weevil and demonstrates the complexity of this response. Knowledge of the genetic mechanisms to explain the enormous diversity of specialized terpenoid metabolites is of great importance for understanding the evolution, diversity and plasticity of the genes that form the fundamentals of plant chemical ecology (Hartmann 2007). An understanding of the potential functions and importance of these compounds in the conifers that dominate many of Canada’s (and the world’s) ecosystems can increase the potential for in the incorporation of their useful properties as we manage the forest resource into the future.   139 5.1 References  Aitken SN, Yeaman S, Holliday JA, Wang T, and Curtis-McLane S. 2008. Adaptation, migration or extirpation: climate change outcomes for tree populations. Evol. Appl. 1: 95-111.  Hamberger B and Bohlmann J. 2006. Cytochrome P450 mono-oxygenases in conifer genomes: discovery of members of the terpenoid oxygenase superfamily in spruce and pine. Biochem. Soc. Trans. 34(6):1209-1214.  Hartmann T. 2007. From waste products to ecochemicals: Fifty years research of plant secondary metabolism. Phytochem. 68:2831-2846.  Keeling CI and Bohlmann J. 2006. Genes, enzymes, and chemicals of terpenoid diversity in the constitutive and induced defense of conifers against insects and pathogens. New Phytol. 170:657-675.  Martin D, Fäldt J, and Bohlmann J. 2004. Functional characterization of nine Norway spruce TPS genes and evolution of gymnosperm terpene synthases of the TPS-d subfamily. Plant Physiol. 135:1908-1927.  Volney WJA and Fleming RA. 2000. Climate change and impacts of boreal forest insects. Agric. Ecosyst. Environ. 82(1-3): 283-294.  Zulak KG and Bohlmann J. 2010. Terpenoid biosynthesis and specialized vascular cells of conifer defense. J. Integrative Plant Biology 52: 86 – 97.   140  Zulak KG, Lippert DN, Kuzyk M, Domanski D, Chou T, Borchers CH, and Bohlmann J. 2009. Targeted proteomics using selected reaction monitoring (SRM) reveals the induction of specific terpene synthases in a multi-level study of methyl jasmonate treated Norway spruce (Picea abies). Plant J. 60(6):1015-30.     141  APPENDIX: Chapter 2 Supplementary Mixed Model ANOVA output  Table S 1 – Output from ANOVA for each compounds identified for differences among cardinal directions. Compound dfnum dfden F-value p-value (+)-α-pinene 3 41 2.1649 0.1068 (+)-3-carene 3 41 0.2227 0.8801 myrcene 3 41 2.1723 0.1059 (-)-limonene 3 41 0.15136 0.9282 (-)-β-phellandrene 3 41 1.1092 0.3564 terpinolene 3 41 0.1578 0.9241 a dfnum = numerator degrees of freedom, dfden = numerator degrees of freedom.       Table S 2 – Output from ANOVA for each compound identified for differences between leader and the top branches. Compound difference SEdiff df t-value p-value (+)-α-pinene 45.0362 107.0660 45 0.4206 0.6760 (+)-3-carene -11.1425 63.4522 45 -0.1756 0.8614 myrcene 35.8031 161.8409 45 0.2212 0.8259 (-)-limonene -198.3025 231.8182 45 -0.8554 0.3968 (-)-β-phellandrene 62.5806 62.5806 45 0.5501 0.5850 terpinolene -2.3850 -2.3850 45 -0.1264 0.9000 a difference = the difference between the leader and the top whorl, SEdiff = standard error of the difference.            142 Table S 3 – Mixed model ANOVA output testing the difference between trees from the intermediate group and the susceptible group. Significant values are in bold. ND = not detected.  Compound β SE t df p-value m on ot er pe ne s (-)-α-pinene 0.088 0.180 0.491 106 0.625 (+)-α-pinene -0.325 0.355 -0.915 106 0.362 α-thujene 0.068 0.165 0.412 95 0.681 (-)-camphene -0.481 0.319 -1.506 56 0.138 (-)-β-pinene 0.053 0.207 0.258 106 0.797 (+)-sabinene 0.494 0.302 1.637 106 0.105 (+)-3-carene 0.209 0.779 0.268 100 0.789 (-)-α-phellandrene 0.237 0.219 1.083 105 0.281 myrcene 0.152 0.154 0.985 106 0.327 α-terpinene 0.194 0.243 0.799 101 0.426 (+)-limonene -0.575 0.422 -1.361 106 0.176 (-)-β-phellandrene 0.113 0.177 0.637 106 0.526 γ-terpinene 0.116 0.161 0.721 96 0.473 terpinolene 0.398 0.283 1.405 106 0.163  di te rp en es  pimaradiene -0.025 0.654 -0.039 59 0.969 dehydroabietadiene ND ND ND 0 ND sandaracopimaradiene -0.162 0.264 -0.613 23 0.546 isopimaradiene -0.178 0.377 -0.472 46 0.639 abietadiene -0.327 0.301 -1.088 88 0.280 neoabietadiene -0.452 0.347 -1.302 4 0.263 levopimaradiene 0.598 0.726 0.825 4 0.456 palustradiene 0.225 0.178 1.263 82 0.210  di te rp en e  al co ho ls  pimaradienol -0.057 0.193 -0.294 103 0.769 dehydroabietadienol -0.073 0.393 -0.185 54 0.854 sandaracopimaradienol -0.958 0.567 -1.691 35 0.100 isopimaradienol 0.269 0.296 0.911 83 0.365 abietadienol 0.025 0.192 0.132 99 0.895 neoabietadienol -0.265 0.352 -0.753 82 0.454 levopimaradienol -0.081 0.144 -0.561 106 0.576 palustradienol 0.327 0.190 1.720 103 0.088  di te rp en e  al de hy de s pimaradienal ND ND ND 0 ND dehydroabietadienal 0.051 0.175 0.293 102 0.770 sandaracopimaradienal 0.293 1.825 0.160 14 0.875 isopimaradienal -0.157 0.428 -0.366 44 0.716 abietadienal 0.027 0.171 0.158 106 0.874 neoabietadienal -0.063 0.146 -0.430 105 0.668 levopimaradienal 0.094 0.178 0.529 106 0.598 palustradienal 0.299 0.167 1.793 106 0.076  di te rp en e  ac id s pimaric acid ND ND ND 0 ND dehydroabietic acid 0.175 0.167 1.047 106 0.298 sandaracopimaric acid 0.105 0.116 0.904 106 0.368 isopimaric acid 0.044 0.145 0.304 106 0.761 abietic acid -0.097 0.149 -0.655 106 0.514 neoabietic acid 0.006 0.142 0.042 106 0.967 levopimaric acid 0.198 0.161 1.228 106 0.222 palustric acid 0.334 0.163 2.048 106 0.043   143 Table S 4 – Mixed model ANOVA output testing the difference between trees from the intermediate group and the resistant group. Significant values are in bold. ND = not detected.  Compound β SE t df p-value m on ot er pe ne s (-)-α-pinene 0.014 0.163 0.088 106 0.930 (+)-α-pinene -0.606 0.320 -1.897 106 0.061 α-thujene 0.096 0.146 0.658 95 0.512 (-)-camphene -0.598 0.274 -2.187 56 0.033 (-)-β-pinene -0.056 0.186 -0.298 106 0.766 (+)-sabinene 0.171 0.272 0.629 106 0.531 (+)-3-carene 0.056 0.697 0.080 100 0.936 (-)-α-phellandrene 0.374 0.197 1.894 105 0.061 myrcene 0.013 0.139 0.092 106 0.927 α-terpinene 0.270 0.217 1.243 101 0.217 (+)-limonene -0.392 0.381 -1.030 106 0.305 (-)-β-phellandrene 0.187 0.160 1.167 106 0.246 γ-terpinene 0.120 0.139 0.858 96 0.393 terpinolene 0.190 0.255 0.743 106 0.459  di te rp en es  pimaradiene -0.467 0.595 -0.785 59 0.436 dehydroabietadiene ND ND ND 0 ND sandaracopimaradiene -0.216 0.225 -0.962 23 0.346 isopimaradiene -0.306 0.324 -0.944 46 0.350 abietadiene -0.085 0.272 -0.312 88 0.756 neoabietadiene -0.105 0.306 -0.342 4 0.750 levopimaradiene 0.832 1.093 0.761 4 0.489 palustradiene -0.004 0.162 -0.027 82 0.979  di te rp en e  al co ho ls  pimaradienol 0.008 0.173 0.048 103 0.962 dehydroabietadienol 0.030 0.370 0.081 54 0.936 sandaracopimaradienol -0.857 0.510 -1.679 35 0.102 isopimaradienol 0.173 0.263 0.657 83 0.513 abietadienol -0.001 0.173 -0.006 99 0.995 neoabietadienol -0.191 0.300 -0.636 82 0.526 levopimaradienol -0.072 0.130 -0.556 106 0.579 palustradienol 0.077 0.171 0.452 103 0.652  di te rp en e  al de hy de s pimaradienal ND ND ND 0 ND dehydroabietadienal -0.189 0.158 -1.190 102 0.237 sandaracopimaradienal -1.255 1.232 -1.018 14 0.326 isopimaradienal -0.222 0.409 -0.543 44 0.590 abietadienal -0.101 0.155 -0.656 106 0.513 neoabietadienal -0.139 0.132 -1.057 105 0.293 levopimaradienal 0.030 0.161 0.184 106 0.854 palustradienal 0.175 0.150 1.165 106 0.247  di te rp en e  ac id s pimaric acid ND ND ND 0 ND dehydroabietic acid -0.186 0.151 -1.237 106 0.219 sandaracopimaric acid -0.011 0.105 -0.108 106 0.914 isopimaric acid -0.029 0.131 -0.224 106 0.823 abietic acid -0.154 0.134 -1.150 106 0.253 neoabietic acid -0.176 0.129 -1.367 106 0.175 levopimaric acid 0.060 0.145 0.411 106 0.682 palustric acid 0.132 0.147 0.896 106 0.372   144 Table S 5 – Mixed model ANOVA output testing the difference between trees from the Haney (H) region versus Big Qualicum (BQ). Significant values are in bold. ND = not detected.  Compound β SE t df p-value m on ot er pe ne s (-)-α-pinene 0.062 0.112 0.553 106 0.582 (+)-α-pinene -0.321 0.220 -1.456 106 0.148 α-thujene 0.000 0.105 -0.004 95 0.997 (-)-camphene 0.022 0.208 0.106 56 0.916 (-)-β-pinene 0.078 0.128 0.605 106 0.546 (+)-sabinene 0.429 0.188 2.289 106 0.024 (+)-3-carene 0.551 0.498 1.107 100 0.271 (-)-α-phellandrene 0.104 0.138 0.756 105 0.451 myrcene 0.261 0.096 2.723 106 0.008 α-terpinene 0.216 0.154 1.401 101 0.164 (+)-limonene -0.350 0.262 -1.334 106 0.185 (-)-β-phellandrene 0.123 0.110 1.117 106 0.266 γ-terpinene 0.301 0.099 3.048 96 0.003 terpinolene 0.422 0.176 2.397 106 0.018 di te rp en es  pimaradiene -0.247 0.397 -0.621 59 0.537 dehydroabietadiene ND ND ND 0 ND sandaracopimaradiene 0.181 0.155 1.168 23 0.255 isopimaradiene -0.250 0.284 -0.882 46 0.382 abietadiene -0.026 0.199 -0.129 88 0.898 neoabietadiene 0.249 0.463 0.539 4 0.619 levopimaradiene -0.126 0.847 -0.148 4 0.889 palustradiene 0.239 0.126 1.899 82 0.061  di te rp en e  al co ho ls  pimaradienol -0.069 0.116 -0.592 103 0.555 dehydroabietadienol 0.225 0.261 0.863 54 0.392 sandaracopimaradienol 0.795 0.311 2.555 35 0.015 isopimaradienol -0.045 0.197 -0.229 83 0.819 abietadienol 0.091 0.117 0.779 99 0.438 neoabietadienol -0.036 0.192 -0.188 82 0.852 levopimaradienol -0.035 0.090 -0.384 106 0.702 palustradienol 0.237 0.123 1.930 103 0.056  di te rp en e  al de hy de s pimaradienal ND ND ND 0 ND dehydroabietadienal 0.092 0.114 0.801 102 0.425 sandaracopimaradienal 0.720 0.951 0.757 14 0.461 isopimaradienal 0.448 0.261 1.716 44 0.093 abietadienal 0.117 0.108 1.090 106 0.278 neoabietadienal 0.166 0.092 1.809 105 0.073 levopimaradienal 0.182 0.111 1.637 106 0.104 palustradienal 0.235 0.103 2.270 106 0.025  di te rp en e  ac id s pimaric acid ND ND ND 0 ND dehydroabietic acid 0.042 0.104 0.403 106 0.687 sandaracopimaric acid 0.144 0.072 2.010 106 0.047 isopimaric acid 0.030 0.090 0.328 106 0.743 abietic acid 0.007 0.092 0.079 106 0.937 neoabietic acid 0.166 0.088 1.876 106 0.063 levopimaric acid 0.181 0.100 1.806 106 0.074 palustric acid 0.139 0.101 1.367 106 0.174   145 Table S 6 – Coefficients of linear discriminants (LD1 and LD2) for the Haney region only. Magnitude of the LD value represents the weight of that compound in the creation of the discriminant function. Compound LD1 Compound LD2 dehydroabietadienal -4.8921195 neoabietic acid -3.75009124 terpinolene -3.8520274 g-terpinene -2.92508269 g-terpinene -3.1639274 a-terpinene -2.26315831 neoabietic acid -2.9957515 abietic acid -2.1210369 levopimaradienal -2.2392576 pimaradienol -1.95743219 palustric acid -1.8242456 palustradienol -1.7142967 sandaracopimaric acid -1.1563468 (-)-a-pinene -1.70997961 abietic acid -0.9623497 (-)-b-phellandrene -1.49940377 abietadienal -0.840444 (+)-3-carene -1.33534449 levopimaradienol -0.1323359 (+)-a-pinene -1.15590847 (-)-α-pinene -0.1270491 dehydroabietic acid -0.98170297 (+)-limonene 0.1667227 (+)-limonene -0.67916096 palustradienal 0.1902308 levopimaradienol -0.54657312 neoabietadienal 0.3793508 myrcene -0.34407422 myrcene 0.3862684 levopimaradienal -0.23728187 (-)-β-phellandrene 0.4046352 palustric acid -0.02518333 (-)-α-phellandrene 0.4082929 (-)-b-pinene 0.47254339 (+)-α-pinene 0.6478381 (+)-sabinene 0.53519472 (-)-β-pinene 0.9989963 levopimaric acid 0.54448421 (+)-3-carene 1.1667911 a-phellandrene 0.67646566 pimaradienol 1.373093 neoabietadienal 1.00600715 palustradienol 1.4863806 abietadienal 1.49846475 a-terpinene 1.9970644 palustradienal 1.6073637 (+)-sabinene 2.1840684 isopimaric acid 2.28578304 isopimaric acid 2.3219065 dehydroabietadienal 3.07514877 levopimaric acid 2.6807463 sandaracopimaric acid 3.1977512 dehydroabietic acid 4.36517 terpinolene 5.22417399             146 Table S 7 – Coefficients of the linear discriminant (LD1) for the Big Qualicum region. Magnitude of the LD value represents the weight of that compound in the creation of the discriminant function. Compound LD1 isopimaric acid -6.53818 (+)-sabinene -4.41273 α-terpinene -3.84422 γ-terpinene -3.30665 sandaracopimaric acid -2.80909 abietadienal -2.8026 (+)-3-carene -2.13962 pimaradienol -1.99635 levopimaric acid -1.72049 levopimaradienol -1.44429 palustradienal -1.12448 (+)-limonene -0.90854 myrcene -0.7835 palustric acid -0.53246 (-)-α-pinene -0.45665 (-)-β-phellandrene 0.047877 palustradienol 0.134335 (+)-α-pinene 0.312244 abietic acid 1.092397 α-phellandrene 1.239826 (-)-β-pinene 1.281941 dehydroabietadienal 1.350867 neoabietadienal 1.583715 levopimaradienal 1.748734 neoabietic acid 2.284138 dehydroabietic acid 8.379155 terpinolene 14.47896            147 Table S 8 – Mixed model ANOVA output testing the difference between trees from the resistant group and the susceptible group for the Big Qualicum region only.  Compound β SE t df p-value m on ot er pe ne s (-)-α-pinene -0.085 0.192 -0.443 33 0.660 (+)-α-pinene 0.172 0.450 0.382 33 0.705 α-thujene -0.084 0.166 -0.505 29 0.617 (-)-camphene -0.077 0.307 -0.249 18 0.806 (-)-β-pinene -0.143 0.233 -0.612 33 0.545 (+)-sabinene 0.050 0.383 0.131 33 0.897 (+)-3-carene -0.906 0.875 -1.035 29 0.309 (-)-α-phellandrene -0.079 0.270 -0.293 32 0.771 myrcene -0.040 0.156 -0.255 33 0.801 α-terpinene -0.166 0.324 -0.514 30 0.611 (+)-limonene -0.244 0.481 -0.507 33 0.616 (-)-β-phellandrene -0.161 0.213 -0.752 33 0.457 γ-terpinene -0.113 0.228 -0.497 30 0.623 terpinolene 0.002 0.361 0.006 33 0.995  di te rp en es  pimaradiene 0.174 0.742 0.234 22 0.817 dehydroabietadiene ND ND ND 0 ND sandaracopimaradiene 0.353 0.295 1.196 7 0.271 isopimaradiene -1.170 1.090 -1.073 10 0.308 abietadiene -0.750 0.379 -1.979 26 0.059 neoabietadiene ND ND ND 0 ND levopimaradiene -0.234 0.680 -0.343 1 0.789 palustradiene -0.030 0.198 -0.150 23 0.882  di te rp en e  al co ho ls  pimaradienol -0.070 0.215 -0.327 32 0.746 dehydroabietadienol -0.493 0.952 -0.518 12 0.614 sandaracopimaradienol -0.352 0.661 -0.532 10 0.606 isopimaradienol 0.029 0.361 0.080 24 0.937 abietadienol -0.132 0.240 -0.547 29 0.588 neoabietadienol -0.073 0.451 -0.162 23 0.873 levopimaradienol -0.273 0.141 -1.940 33 0.061 palustradienol 0.168 0.244 0.688 31 0.497  di te rp en e  al de hy de s pimaradienal ND ND ND 0 ND dehydroabietadienal 0.111 0.221 0.500 31 0.621 sandaracopimaradienal ND ND ND 0 ND isopimaradienal -0.160 0.427 -0.374 16 0.713 abietadienal -0.057 0.154 -0.370 33 0.714 neoabietadienal 0.131 0.129 1.014 33 0.318 levopimaradienal -0.096 0.159 -0.606 33 0.549 palustradienal -0.110 0.179 -0.614 33 0.543  di te rp en e  ac id s pimaric acid ND ND ND 0 ND dehydroabietic acid 0.129 0.204 0.631 33 0.532 sandaracopimaric acid -0.108 0.137 -0.789 33 0.435 isopimaric acid -0.096 0.154 -0.621 33 0.539 abietic acid -0.074 0.139 -0.530 33 0.599 neoabietic acid 0.139 0.147 0.946 33 0.351 levopimaric acid -0.065 0.188 -0.344 33 0.733 palustric acid -0.111 0.203 -0.547 33 0.588 

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0069660/manifest

Comment

Related Items