UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Molecular systematics and population genomics of the tree-pathogenic fungus Grosmannia clavigera Massoumi Alamouti, Sepideh 2013

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

Item Metadata

Download

Media
24-ubc_2014_spring_massoumialamouti_sepideh.pdf [ 44.61MB ]
Metadata
JSON: 24-1.0165679.json
JSON-LD: 24-1.0165679-ld.json
RDF/XML (Pretty): 24-1.0165679-rdf.xml
RDF/JSON: 24-1.0165679-rdf.json
Turtle: 24-1.0165679-turtle.txt
N-Triples: 24-1.0165679-rdf-ntriples.txt
Original Record: 24-1.0165679-source.json
Full Text
24-1.0165679-fulltext.txt
Citation
24-1.0165679.ris

Full Text

 Molecular systematics and population genomics of the tree-pathogenic fungus Grosmannia clavigera    by    Sepideh Massoumi Alamouti    A thesis submitted in partial fulfillment of the requirements for the degree of    Doctor of Philosophy    in    The Faculty of Graduate and Postdoctoral Studies   (Forestry)    The University of British Columbia (Vancouver)    November 2013    ? Sepideh Massoumi Alamouti, 2013  ii Abstract  Ophiostomatoid fungi increasingly damage forests, but understanding their interactions with vectors and hosts is hampered by uncertainty over the validity of genera, relationships between genera, and species boundaries. To address some of these issues, I first generated a new multigene dataset from 67 taxa that represent the genus Grosmannia and other related genera. The multigene phylogeny resolved the Grosmannia fungi into a clade that was separated from previously intermixed species of the genera Ambrosiella and Raffaelea, and that corresponded to distinct ecological niches and vector associates, i.e. bark versus wood-boring beetles.  Second, I generated and used 15 gene genealogies to define species boundaries in G. clavigera. This destructive pine pathogen is vectored by two beetle species: mountain and Jeffrey pine beetles (MPB, JPB).  MPB and its fungal associates have expanded into the largest epidemic in western North American history. I identified two phylogenetic species: Gs and Gc. Gc is present in the phylogenetically close Pinus species ponderosa and jeffreyi, which are infested by localized populations of their respective beetles. In contrast, Gs is an exclusive associate of MPB and its primary host P. contorta, although it is found in other pine species in current epidemic regions. These results suggest that host-tree species and beetle population dynamics are important factors in the genetic divergence and diversity of fungal associates in the beetle-tree ecosystems.  iii  Finally, we generated new genomic sequences for eleven Gs and Gc strains to further assess evidence for divergence in these fungi as they adapt to different pine species, and to find genes that may be involved in species divergence. Aligning these genome sequences to the reference genome, we identified 103,430 SNPs that supported the Gs and Gc lineages and divided each lineage into two subclades. Genome-wide scans identified truncated genes and potential pseudogenes that differed between Grosmannia lineages, as well as seven genes that show evidence of positive selection. The seven genes are involved in secondary metabolism and in detoxifying host-tree defense chemicals (e.g. polyketide synthases, oxidoreductases), and their variants may reflect adaptation to the specific chemistries of P. contorta, ponderosa, and jeffreyi.           iv Preface  A version of chapter 2 has been published as ?Massoumi Alamouti S, Tsui CKM, Breuil C. (2009). Multigene phylogeny of filamentous ambrosia fungi associated with ambrosia and bark beetles. Mycological Research. 113: 822?835?.  I conducted the experimental design, experiments and analyses with guidance from Colette Breuil. Clement Tsui provided some inputs for the Bayesian analysis. I wrote the manuscript with the guidance and assistance from Colette Breuil.    A version of chapter 3 has been published as ?Massoumi Alamouti S, Wang V, DiGuistini S, Six DL, Bohlmann J, Hamelin RC, Feau N, Breuil C. 2011. Gene genealogies reveal cryptic species and host preferences for the pine fungal pathogen Grosmannia clavigera. Molecular Ecology. 20: 2581?2602?.  I conducted the experimental design, experiments and analyses with guidance from Colette Breuil. I received inputs from Nicolas Feau for the analyses and suggestions from Scott DiGuistini and Richard Hamelin for finding the polymorphisms. Vincent Wang helped me with DNA extraction and PCR for the fungal samples. Diana Six conducted the fungal samplings in California. I wrote the manuscript with the guidance and assistance of Colette Breuil and inputs from Nicolas Feau, J?erg Bohlmann, Diana Six and Richard Hamelin.     v A version of chapter 4 will be submitted to a peer-reviewed journal as ?Massoumi Alamouti, et al. 2013. Comparative genomics of the pine pathogens and the beetle symbionts in the genus Grosmannia?.  I conducted the experimental design with guidance from Colette Breuil. Illumina sequencing was carried out in the Genome Science Center (GSC, Vancouver, BC, Canada). Sajeet Haridas generated the de novo genome assemblies. I proposed the improvements of the de novo assemblies (i.e. by mapping to the reference genome) that were generated by Sajeet Haridas. I performed the data analyses (SNP calling, orthologous gene finding, phylogenomics, gene genealogies and the evolutionary selection analyses). The selection analyses using PAML was ran by Nicolas Feau. Gordon Robertson showed me how to generate the graphics for figures 4.1b and 4.4a-b. I wrote the chapter with the guidance and assistance from Colette Breuil and Gordon Robertson and inputs from Nicolas Feau, Mary Berbee, Ye Wang and J?erg Bohlmann.          vi Table of Contents  Abstract ...........................................................................................................................ii Preface ...........................................................................................................................iv Table of Contents ..........................................................................................................vi List of Tables ...............................................................................................................xiii List of Figures..............................................................................................................xiv Acknowledgments........................................................................................................xv Dedication ...................................................................................................................xvii Chapter 1 General introduction and research objectives .......................................1 1.1 Introduction............................................................................................................1 1.2 Symbiosis ..............................................................................................................2 1.2.1 Bark beetle-tree interactions............................................................................3 1.2.2 Bark beetle-fungal interactions ........................................................................9 1.2.3 Fungal-beetle-tree interactions......................................................................12 1.3 Ophiostomatoids ecology, biology and systematics............................................14 1.4 Fungal species concept and recognition .............................................................19 1.4.1 Defining species boundaries and population structures in G. clavigera ........22 1.5 Fungal genomics and their importance in G. clavigera .......................................25 1.5.1 Next generation sequencing and fungal population genomics ......................28 1.6 Single nucleotide polymorphisms........................................................................30 1.7 Divergent adaptive evolution in fungi ..................................................................32 1.8 Overview and purpose of this thesis ...................................................................34  vii Chapter 2 Multigene phylogeny of ophiostomatoid fungi associated with bark and ambrosia beetles...................................................................................................36 2.1 Introduction..........................................................................................................36 2.2 Materials and methods ........................................................................................40 2.2.1 Taxon sampling .............................................................................................40 2.2.2 Morphological investigation ...........................................................................41 2.2.3 DNA extraction, PCR amplification and sequencing .....................................41 2.2.4 Phylogenetic analysis ....................................................................................42 2.3 Results ................................................................................................................44 2.3.1 Sequence analysis ........................................................................................44 2.3.2 Phylogenetic analysis ....................................................................................45 2.3.3 Morphological investigation ...........................................................................49 2.4 Discussion ...........................................................................................................50 2.4.1 Ambrosiella associates of bark beetles are related to the teleomorph genus Ophiostoma ....................................................................................................................52 2.4.2 Ambrosiella associates of ambrosia beetles are related to the teleomorph genus Grosmannia .........................................................................................................54 2.4.3 Morphological features ..................................................................................58 2.5 Tables and figures...............................................................................................62 Chapter 3 Gene genealogies reveal cryptic species and host preferences for the pine fungal pathogen Grosmannia clavigera.............................................................72 3.1 Introduction..........................................................................................................72 3.2 Materials and methods ........................................................................................77  viii 3.2.1 Samples.........................................................................................................77 3.2.2 Isolation .........................................................................................................78 3.2.3 Polymorphism detection ................................................................................79 3.2.4 DNA extraction, primer design and sequencing ............................................80 3.2.5 Sequence alignments and analyses..............................................................80 3.2.6 Gene trees and concatenated data phylogeny..............................................81 3.2.7 Network approaches and evidence for recombination in G. clavigera ..........82 3.3 Results ................................................................................................................83 3.3.1 Polymorphism discovery................................................................................83 3.3.2 Polymorphism validation................................................................................84 3.3.3 Single-gene phylogenies, phylogenetic species recognition and concatenated phylogeny .......................................................................................................................85 3.3.4 Evidence of recombination ............................................................................89 3.3.5 Ecological and morphological characteristics................................................90 3.4 Discussion ...........................................................................................................91 3.4.1 Evolutionarily independent lineages ..............................................................93 3.4.2 Evidence of recombination ............................................................................95 3.4.3 Ecologically distinguishable...........................................................................97 3.5 Tables and figures.............................................................................................105 Chapter 4 Comparative genomics of the pine pathogens and the beetle symbionts in the genus Grosmannia .......................................................................119 4.1 Introduction........................................................................................................119 4.2 Material and methods........................................................................................124  ix 4.2.1 Fungal samples ...........................................................................................124 4.2.2 Illumina paired-end library construction, sequencing and assembly ...........124 4.2.3 Gene predictions and ortholog determination..............................................125 4.2.4 Mapping and variant calling.........................................................................127 4.2.5 Verification of variant calls...........................................................................128 4.2.6 Functional annotations for SNVs .................................................................129 4.2.7 SNVs clustering and phylogenomics ...........................................................130 4.2.8 Gene genealogies and concatenated data phylogeny ................................132 4.2.9 Detecting signature of selection and rate of protein evolution.....................133 4.2.10 Physiological assessments ........................................................................135 4.3 Results ..............................................................................................................136 4.3.1 Genome assembly, orthologs determination and SNV variants ..................136 4.3.2 Functional classification of genomic variants ..............................................137 4.3.3 Divergence classification of genomic variants.............................................139 4.3.4 Clustering and phylogenomic analysis of SNVs ..........................................140 4.3.5 Ecological and physiological assessments..................................................141 4.3.6 Signature of positive and purifying selections in Grosmannia .....................142 4.4 Discussion .........................................................................................................146 4.4.1 Grosmannia draft genomes .........................................................................147 4.4.2 Grosmannia genome-wide SNVs ................................................................148 4.4.3 Grosmannia SNV-phylogenomics ...............................................................151 4.4.4 Grosmannia genes involved in host adaptation and ecological divergence ...... .....................................................................................................................155  x 4.5 Tables and figures.............................................................................................160 Chapter 5 Conclusions ...........................................................................................173 5.1 Ophiostomatoids systematics limitation and status of the genus Grosmannia ....... ..........................................................................................................................175 5.2 Defining species boundaries in Grosmannia clavigera .....................................176 5.2.1 Polymorphism discovery and species recognition.......................................177 5.2.2 Ecologically distinguishable lineags ............................................................178 5.3 Population genomics in Grosmannia.................................................................179 5.3.1 Grosmannia draft genomes and genome-wide characterization of SNVs...179 5.3.2 Genome-wide SNV-phylogeny ....................................................................181 5.3.3 Functional characterization of SNVs and their adaptive contribution ..........182 5.4 Perspective on future work................................................................................185 References. .................................................................................................................187 Appendices .................................................................................................................215 Appendix A: Supplementary information supporting chapter 3 ....................................215 A.1 List of isolates, sampling locations, collection resources and deposited culture collections.....................................................................................................................216 A.2 Primer sequences and gene descriptions for 67 G. clavigera loci screened for polymorphisms .............................................................................................................220 A.3 Polymorphism summaries and diversity indices within the two monophyletic clades in G. clavigera ...................................................................................................224 A.4 Gs morphology compared with those of the G. clavigera holotype ...............225 A.5 Haplotype network .........................................................................................226  xi Appendix B: Supplementary information supporting chapter 4 ....................................228 B.1 Main features of primary genome sequence data .........................................228 B.2 Assembly statistics of Grosmannia genomes ................................................229 B.3 Summary of genBlastG output used for gene annotations of each Grosmannia draft genome, with pairwise homology (PID) with reference gene models ..................230 B.4 Total number of sequence reads and filtering steps used for SNVa calling...231 B.5  Primer sequences used in the SNV validation and in the phylogenetic and population genetic analyses .........................................................................................232 B.6 Grosmannia gene-model summaries.............................................................233 B.7 McDonald-Kreitman (MK) test results and the mean Gs-Gc pairwise rate of protein-coding divergence (dN/dS)...............................................................................234 B.8 PAML ?site-model? test of positive selection for 1,213 Gs-Gc orthologous genes .......................................................................................................................234 B.9 Variants identified using the published Grosmannia genome as the reference sequence......................................................................................................................235 B.10 Gene models containing intra- and interspecific stop-codon variants..........236 B. 11 GO functional enrichment analysis of potential pseudogenes in Gs and Gc strains ....................................................................................................................240 B.12 Map of western North America representing the mountain pine beetle (MPB) distribution. ...................................................................................................................241 B.13 Large-scale synteny between Grosmannia genomes..................................242 B.14 Genetic distance between the twelve Grosmannia genomes. .....................243  xii B.15 Haplotype networks and genealogies of nine gene regions sequenced in 28 Grosmannia strains. .....................................................................................................244 B.16 Distributions of polymorphism-to-divergence ratios (NI) and the rate of protein-coding evolution (dN/dS) across 3,476 Gs-Gc orthologous gene models. ......246                     xiii List of Tables  Table 2.1 Fungal species used in this study ..................................................................62 Table 2.2 Morphological and ecological characters reported for the genera Ambrosiella, Raffaelea and Dryadomyces in the literature or in this study .........................................65 Table 3.1 Fungal isolates used in this study ................................................................105 Table 3.2 Primer sequence and gene description for loci used in phylogenetic and population genetic analyses .........................................................................................108 Table 3.3 Fixed and shared polymorphisms between the two monophyletic clades in Grosmannia clavigera ..................................................................................................109 Table 3.4 Information on phylogenetic dataset sequenced from G. clavigera and its close relatives...............................................................................................................110 Table 4.1 Fungal strains used in this study..................................................................160 Table 4.2 Summary of the genomic and gene coverage data in the eleven sequenced genomes. .....................................................................................................................162 Table 4.3 Genome-wide characterization of fixed and shared polymorphisms between Gs and Gc lineages......................................................................................................163 Table 4.4 The top 42 genes showing evidence of positive selection a.........................164      xiv List of Figures  Figure 2.1 Multigene phylogeny of ophiostomatoid fungi ..............................................69 Figure 2.2 Sporodochium-like formation........................................................................70 Figure 2.3 Annellidic conidiogenous cells......................................................................71 Figure 3.1 Single-locus phylogenies of 15 genes in G. clavigera and its four close relatives ....................................................................................................................113 Figure 3.2 Fungal collection sites and 15?gene phylogeny of Grosmannia clavigera complex ....................................................................................................................115 Figure 3.3 Recombination analysis..............................................................................117 Figure 3.4 Asexual and sexual stage in Gs .................................................................118 Figure 4.1 Grosmannia SNP-phylogenomics, gene content and amino acid similarity. .... ....................................................................................................................167 Figure 4.2 Intra-/interspecific variants in the terpenoid-processing gene cluster.........169 Figure 4.3 Grosmannia species phylogeny correlated with the host-tree species of different phylogenetic lineages and fungal lineage tolerance to a related host-defence chemical ....................................................................................................................171 Figure 4.4 Comparison of divergence and polymorphism ...........................................172      xv  Acknowledgements  Without the support and patience of my advisors, colleagues, friends and family this thesis would not have been possible. It is to them that I owe my deepest gratitude. First I thank my supervisor Colette Breuil who patiently taught me about being critical, prepared and responsible. I appreciate her endless support and guidance throughout all these years. Her knowledge, enthusiasm and energy were motivating and contagious and made my PhD stimulating and productive. To Gordon Roberston for the excellent mentorship, I appreciate all his contributions of time and expertise that helped me to grow professionally and to move on. I?m grateful to my committee members Mary Berbee and Richard Hamelin who have been always encouraging and enthusiastic about my work. I am always thankful to Mary Berbee for being the source of inspiration to me from the very beginning and for teaching me so many things. To Jan Stenlid, Martin Adamson, Kermit Ritland, Loren Rieseberg and Les Lavkulich for accepting to be in my examining committee and for their time and insightful questions. It?s a pleasure to acknowledge Joerg Bohlmann, who always kindly granted his time and provided insightful discussions and guidance to improve my study. An especial acknowledgement to Keith Seifert, Adnan Uzunovic and Diana Six, who have been always supportive and introduced me to their fascinating world of science. To Inanc Birol, Rod Docking, Shaun Jackman, Simon Chan and Greg Taylor who make the Genome Science Center a welcoming place for sharing ideas and a source of good advice and collaboration. To collaborators at the United States Barbara Bentz, Matthew Hansen and Jim Vandygriff  xvi for their kind support with the fieldwork, and to Kathy Bleiker and Adrian Rice for sharing their fungal cultures. I also thank the great folks whom I worked with Vincent Wang, Lynette Lim, Vicente Hernandez, Huang-Ju Chen, Marie Josee, Thomas Wang, Ahmed Kenawy, Pablo Chung, Mohamed Ismail, Fariba Izadi, Kristin Tangen, Clement Tsui, Ljerka Lah, Lily Khadempour, Sajeet Haridas, Ye Wang and Scott DiGuistini. Their excellent work, encouragement and friendship made my PhD life more enjoyable and energizing. To my friends, especially Mahin Narimany and Farah Larki for helping me keep things in perspective and providing a much-needed time of occasional escapes. Special thanks to my amazing siblings, Siavash, Siamack and Sepehr for their love, support and encouragement. In particular, I would like to thank my brother Siavash. His determination and sense of joy have always kept me inspired, and I could not do this without him. My deepest respect to my partner Hedayat Alghassi, who endured this process with unselfishness and never-ending encouragement. His patience, love and support throughout all these years are something for which I will be forever grateful. Mom and Dad, Atiyeh Shaeri and Ahmad Massoumi Alamouti, I love you both and I wish you all the joy that you always ensured for me to experience. You raised me with love and supported me in all my pursuits. I cannot thank you enough.       xvii  Dedication       To my loving mother and father   Atiyeh and Ahmad  1 Chapter 1 General introduction and research objectives  1.1 Introduction  Worldwide, beetles and their fungal associates remain among the most damaging forest pests (Wood 1982; Ayres and Lombardero 2000; Kirisits 2004). Pests that are at low and relatively stable endemic levels can quickly transition into substantial threats over regions far larger than the sites of origin by expanding into large-scale outbreaks or by invading new environments (Bentz et al. 2010). One such example is the mountain pine beetle (MPB; Dendroctonus ponderosae)-fungal outbreak, which in Canada alone has infested over 18.1 million hectares of Pinus contorta forests (http://www.for.gov.bc. ca/hfp/mountain_ pine_beetle/facts.htm). Cataloging the diversity of species and populations is necessary for developing an understanding of how these organisms interact with their host trees, and assessing their potential threat to forest ecosystems. My thesis addresses some of these issues, focusing on the ophiostomatoid (Sordariomycetes, Ascomycota) Grosmannia clavigera. This fungus is an important component of the MPB-fungal outbreak. It is an obligate symbiont of the MPB and can be a destructive pathogen for P. contorta trees. We generated new molecular resources for G. clavigera and some related species, which I used to (1) establish the fungal phylogeny, species boundaries and population structures in both epidemic and localized populations infesting different species of pine, (2) assess the potential role of host trees and/or beetle vectors in fungal evolution and divergence, and (3) identify genes that may be involved in fungal species divergence and/or host adaptation. The outcome of  2 this work provides a foundation for future research on the comparative and functional genomics of this important group of fungi, and of related ophiostomatoid genera. In this introductory chapter I will give a brief overview of the following topics (1) symbiosis; (2) ophiostomatoid biology, ecology and systematics; (3) fungal species concepts and criteria; (4) fungal comparative and population genomics, and (5) the rationale and the objectives of my work.  1.2 Symbiosis  Fungi are heterotrophs and so they must interact with other organisms to obtain their food. This has led fungi to evolve saprophytic, mutualistic or parasitic lifestyles and nutritional modes. Fossils and molecular clock data suggest that fungi have repeatedly evolved partnerships (i.e. symbiosis) with plants, some of which are ancient and even ancestral to terrestrial plants (Selosse and Le Tacon 1998; Heckman et al. 2001). The establishment of plants and other eukaryotes on land was probably facilitated by such partnerships (Selosse and Le Tacon 1998). It is likely that initially some mycelium-forming fungi were saprophytic and, as they interacted with plants, continued to develop the ability to tolerate plant defenses, so that parasitic and, eventually, biotrophic interactions evolved (Taylor and Osborn 1996). While the earliest fungi have been reported from the Precambrian period, the first examples of plant responses to fungi have been traced back to early Devonian (Taylor et al. 1992; Heckman et al. 2001). While plants and fungi continued to form a close association from the early stages of terrestrial colonization, insects likely originated in the Silurian period (Taylor and Osborn  3 1996; Engel and Grimaldi 2004). Therefore, it is likely that fungi first adapted to plants and that interactions with insects developed later. Beetle-tree-fungal symbiosis has been dated as originating between 40 and 85 million years ago. Today, symbiotic associations are widespread in nature, are essential for the functions of ecosystems and the evolution of biodiversity, and involve complex interactions that are still poorly understood (Farrell et al. 2001; Six 2003; Harrington 2005; Jordal and Cognato 2012).    1.2.1 Bark beetle-tree interactions  Bark beetles belong to the Scolytinae subfamily, which includes over 6,000 described species within about 225 genera including many ambrosia and cone beetles that are specialized in exploiting woody plants (coniferous/hardwood trees and shrubs). Most bark beetles are harmless to healthy living trees, but some, in particular those in the genus Dendroctonus, are important pests in coniferous forests, especially in the boreal and temperate regions of the northern hemisphere.  Each beetle species typically colonizes either a few (i.e. often closely related members of the same genus) or a single tree species (e.g. D. rufipennis, D. pseudotsugae and D. jeffreyi); the colonization occurs on specific parts of the host tree, and each tree can be inhabited by a large number of beetles (Wood 1982). However, some bark beetles (e.g. the MPB) can attack more distantly related host species; and can also switch to novel host species when beetle ranges expand into new habitats where ?na?ve? host trees may have not developed sufficient defenses against large-scale attacks (Ayres and Lombardero 2000; Bentz et al. 2010; Safranyik et al. 2010; Cullingham et al. 2011). These potential new  4 interactions are a matter of concern, as they can result in extensive outbreaks and damage in forest ecosystems as the MPB-fungal complexes on Pinus forests in Canada (Kurz et al. 2008; Safranyik et al. 2010; Cullingham et al. 2011).  1.2.1.1 Mountain pine beetle: ecology, biology and systematics   Mountain pine beetles are native to North America and historically found in areas from northern Mexico to western Canada (BC and marginal regions in western Alberta where P. contorta trees are present), and from the Pacific Coast to South Dakota in the USA (Wood 1982; Mock et al. 2007). Although the MPBs? primary hosts are P. contorta, P. ponderosa and P. monticola, the beetles can successfully attack and reproduce in most pine species throughout their range. These pine species include: P. albicaulis, P. strobus, P. flexilis, P. sylvestris and P. lambertiana (Wood 1982; Safranyik and Wilson 2006). Host species have been shown to cause variation in some parameters of the beetles? life-cycle, like survival, phenology, development rate, and body size (Reid 1962; Reid 1963; Safranyik and Linton 1983; Langor 1989; Langor et al. 1990; Safranyik and Wilson 2006).  Mountain pine beetles have also been recorded attacking other coniferous genera such as Picea engelmannii ! glauca, P. mariana and Abies species. Successful reproduction in non-pine hosts has only been reported for the MPB populations infesting Picea species. However, these observations are based on pheromone-baiting of the  5 beetles in epidemic regions and not on natural host-tree colonization and reproduction (Huber et al. 2009; Safranyik et al. 2010). The life cycle phase in which MPBs disperse and find new hosts often occurs during a short period between July and early September, depending on the geographic area (Safranyik and Wilson 2006). In large outbreaks, bark beetles are reported to locate a suitable host tree by randomly landing and testing the tree and its resistance capability (Hynum and Berryman 1980; Wood 1982). After the selection of host trees they use volatiles or pheromones to recruit more beetles, leading to a mass attack that defeats the chemical defenses of individual trees (Raffa and Berryman 1982; Raffa and Berryman 1983). During the last two decades of MPB-fungal outbreaks, the beetles have expanded their geographic ranges and have established in new host species (Cullingham et al. 2011). The beetle-fungal complexes have crossed the Rocky Mountain to the north-central Alberta where forest composition shifts to P. banksiana (jack pine), which is evolutionary close and form a hybrid zone with P. contorta (Bentz et al. 2010; Safranyik et al. 2010). Cullingham et al. (2011) used microsatellites to identify species and hybrids of pine and showed successful MPB-fungal attack and establishment in natural P. banksiana. Because P. banksiana is an important component of the North American boreal forest, it may permit the MPB to expand to the east across the north of Canada.   Outside the MPB epidemic populations, localized populations of the beetle-fungal complexes in the western and southern USA have been studied less extensively (Wood 1982; Mock et al. 2007). Generally, in response to host abundance and climate suitability, the beetle population size can progress through characteristic ?endemic,  6 incipient epidemic, epidemic (i.e., outbreak) and post-epidemic (i.e., declining)? phases (Safranyik and Carroll 2006). During the endemic phase the ?localized populations? are very small (less than 10 trees are attacked per hectare) and beetles typically infest damaged trees or those with compromised defense capacity (Carroll et al. 2006). Localized populations of beetles tend to be specific to some pine species (Wood 1982; Sturgeon and Mitton 1986a; Langor et al. 1990). However, given the right climate and suitable resources, they can erupt into large-scale epidemics causing significant losses of healthy pine stands of different species. Large-scale outbreaks of MPB-fungal complexes have caused mortality of hundreds of millions of trees including the primary host P. contorta and other pine species over large epidemic regions (Kurz et al. 2008; Safranyik et al. 2010; Cullingham et al. 2011).  Two factors contribute to an ongoing MPB-fungal outbreak in western North America: 1) food supply (abundant mature P. contorta resulted from forest management practices such as extensive fire suppression over the past 50 years), 2) climate changes (mild winters lead to low brood mortality and dryer summers increase stress on trees).  1.2.1.2 Mountain and Jeffrey pine beetle taxonomy   Dendroctonus ponderosae was originally described from P. ponderosa in South Dakota (Hopkin 1902). Later, the mountain pine beetle (D. monticolae) was described from P. contorta, P. ponderosa, P. lambertiana and P. monticola while the Jeffrey pine beetle (D. jeffreyi, JPB) was described from P. jeffreyi (Hopkins 1909). The range of D. ponderosae extended south and east into Wyoming, South Dakota, Utah, Colorado,  7 Arizona and New Mexico, and D. monticolae was found in Idaho, Montana and British Columbia, as well as south into California along the west coast. Experimental mating of D. ponderosae and D. monticolae suggested that these two beetles might represent only one species that varied in body size with host and region (Hay 1956). The two species, D. monticolae and D. ponderosae, were synonymized as one single species of mountain pine beetle: ?D. ponderosae? (Wood 1963). Additional evidence (i.e. phenology and karyology) supported the synonomy of D. ponderosae and D. monticolae, and confirmed the species status of D. jeffreyi (Lanier and Wood, 1968; Renwick and Pitman 1979; Z??iga et al. 2002). Later, genetic studies showed differences in MPB colonizing P. contorta and P. ponderosa where they intermixed in localized populations in Colorado and California, and also between populations breeding in two varieties of P. contorta (i.e. var. murrayana and var latifolia) (Stock and Guenter 1979; Stock and Amman 1980; Sturgeon and Mitton 1986a; Kelley et al. 2000). Using mtDNA sequences and AFLP markers, Mock et al. (2007) have shown an isolation-by-distance pattern of separation, and some of the isolated populations (e.g. Arizona populations) were congruent with the original morphological subdivision between D. ponderosae and D. monticolae. While uncertainties around the host specificity and perhaps species boundaries in the MPB remain, this will soon change now that one genome sequence of the beetle is available (Keeling et al. 2013). A rapid and cost-effective re-sequencing of additional strains from distinct localized populations of the MPB would help clarify its population/species boundaries, which would support population genotyping analyses and identifying potential adaptive genomic differences.    8 While it seems that MPB can attack all species of pine across its range, there is no evidence of this beetle on P. jeffreyi ? even in sympatric regions where MPBs and JPBs and their respective hosts are intermixed and the trees are attacked (Higby and Stock 1982; Wood 1982; Six and Paine 1999; Kelley et al. 2000). Instead P. jeffreyi is commonly infested by the JPB. This beetle species is a sibling of the MPB with an almost identical morphology, life cycle and gallery characteristics; however, JPB is a highly specialized beetle that only occur in P. jeffreyi trees in a limited geographic region in the western USA (Higby and Stock 1982; Wood 1982; Six and Paine 1999). The differences in host tree range and the physiological tolerances to various tree defense compounds, as well as a low genetic variation have distinguished these bark beetles as two distinct species (Kelley and Farrell 1998; Six et al. 1999; Kelley et al. 2000; Six 2003). The resistance of P. jeffreyi to the MPB may be related to heptane, which is a volatile chemical-defense unique to this pine species (Mirov and Hasbrouck 1976; Six and Paine 1998; Smith 2000). The JPB can periodically kill healthy trees, but because of its limited host range and restricted geographic distribution, it is economically less important than the MPB. The geographic range of JPBs follows roughly that of its host tree, extending from the northern border of California to the northern Baja peninsula, Mexico (Wood 1982). Based on molecular makers, Kelly et al. (2000) showed lower genetic diversity in JPB than in MPB populations, and they speculated that the specialist JPB has diverged from MPB, a more ?generalist? beetle species.    9 1.2.2 Bark beetle-fungal interactions  Beetles and their associated fungi spend most of their life cycle protected either under the bark (e.g. bark beetles) or inside the wood (e.g. ambrosia beetles). The beetle life cycle usually requires one to three years; however, some beetles have several generations per year (Wood 1982). The MPB and its sister species JPB spend their whole one-year life cycle under the bark except for a short dispersal flight in summer (Whitney 1971; Six et al. 1999). The timing of MPB attacks is related to seasonal temperatures with attacks occurring earlier in warmer areas. If the beetles successfully overcome host tree defenses, the adult beetles construct vertical galleries in the phloem under the bark, where they lay eggs. During this process, spores of the fungal symbionts are transferred to the phloem and the outer surface of sapwood, where they start to grow and reproduce. Generally, the eggs hatch within a few days and eventually late larval instars overwinter underneath the bark. Before emerging, young adult beetles will come in contact with the fungi that have grown from the spores introduced by the parents. When new adult beetles emerge and fly to select and attack new host trees, they transport the slimy fungal spores on their exoskeletons, or in mycangia (organs specialized for transferring fungal spores to new hosts) and/or inside their gut. As beetles attack the trees, the cycle is repeated. Each generation generally takes one year, except at high elevations where it can take two years.  Beetle bodies and their galleries host a variety of microorganisms, mostly bacteria, yeasts and filamentous fungi. Among all fungal symbionts, ophiostomatoids are the  10 most common and relatively well-known associates of both bark and ambrosia beetles (M?nch 1907; Wingfield et al. 1993; Wingfield and Seifert 1993; Jacobs and Wingfield 2001; Kirisits 2004; Harrington 2005; Massoumi Alamouti et al. 2007). The association of ophiostomatoid fungi with particular beetle species can be either specific or more casual (Kirisits 2004). In a specific association, different populations of the beetle consistently carry spores of certain fungal species that are generally not found on any other beetles. For example there has been a historical association of the MPB with two ophiostomatoid species: G. clavigera and O. montium (Rumbold 1941; Robinson 1962; Robinson-Jeffrey and Davidson 1968). Both species are present in the beetle?s maxillary mycangia, but O. montium is isolated more frequently from the beetles? exoskeletons than from the mycangia (Whitney and Farris 1970; Six 2003). Fungal associates of JPBs have been sampled throughout its most geographic range but only from the mycangia. All isolates from JPBs have been identified as G. clavigera and no O. montium has been found (Six and Paine 1997). Although more work is required to assess the diversity of fungal species associated with the JPB, G. clavigera were shown to be the dominant and specific symbiont of this beetle and its sister species MPB (Six 2003).   Recent works by Lee et al. (2005; 2006a) on the epidemic populations of MPB infesting P. contorta, reported two distinguished additional fungal species closely associated with the beetle. When inoculated at high densities, the pathogenic Leptographium longiclavatum grows inside the P. contorta sapwood and can subsequently kill the tree. The non-pathogenic, slow growing Ceratocystiopsis sp. is  11 mainly found inside the beetle galleries. Heterogeneous assemblages of species from different ophiosotmatoid genera are rather common in bark beetle ecosystems (Six 2003; Kirisits 2004; Massoumi Alamouti et al. 2007), possibly because different fungal associates have different roles in the beetles? life cycle.    Beetle- fungal interactions are complex and poorly understood and thus can be controversial. Interactions are diverse, ranging from antagonistic to commensal or mutualistic relationships (Klepzig and Wilkens 1997; Six 2003). In many cases, beetle-fungal symbioses seem to be mutualistic, benefiting both the fungus and the beetle (Ayres and Lombardero 2000). The ophiostomatoid fungi almost completely rely on beetles for dispersal and clearly benefit by being transported to new hosts (Six 2003; Harrington 2005). Besides this protected transport to suitable habitats, beetle dispersal provides protection from desiccation and UV light (Klepzig and Six 2004). The fruiting structures of these fungi usually have long stalks bearing sticky and concave shaped spores that can easily adhere to beetle cuticles for dispersal. These morphological features are considered as adaptations to insect dispersal and to the bark beetle habitat (Francke-Grosmann 1967; Whitney 1971; Malloch and Blackwell 1993).   The evolution of mycangia also indicates that some beetles also benefit from the association with fungi (Paine et al. 1997; Harrington 2005). Based on a combination of evidence and speculation fungal symbionts may 1) increase the availability of nutrients (e.g. nitrogen and sterol) that are rather scarce in host trees so that beetles can complete their life cycle; 2) make the infested trees more favorable for the beetles by  12 detoxifying host defense metabolites and lowering the wood moisture content; and 3) reduce the effectiveness of tree defenses (Lieutier et al. 2009). Given this, the relationship between G. clavigera and its beetle vectors is mutualistic, partly due to the specificity of the association and the development of the beetles? mycangia for the maintenance and protection of fungal spores.  1.2.3 Fungal-beetle-tree interactions   Tree mortality results from complex interactions among the tree, beetles and their fungal associates (Robinson 1962; Yamaoka et al. 1990; Yamaoka et al. 1995; Paine et al. 1997; Lee et al. 2006b; Rice, et al. 2007a,b; DiGuistini et al. 2011; Wang et al. 2013). Pioneer fungal species, like G. clavigera, need to colonize tree tissues that are alive and to overcome tree defenses induced by the beetle attacks. While the suggestion that fungal pathogens help the beetle to colonize host trees is speculative (Six and Wingfield 2011), increasing evidence suggests that the beetle-tree-associated fungi have evolved specific mechanisms that allow them to colonize healthy trees (Hesse-Orce et al. 2010; DiGuistini et al. 2011; Lah et al. 2013; Wang et al. 2013). Fungi that do not display high levels of pathogenicity might be those invading tree tissues later and more slowly, following pathogenic fungal associates (e.g. Ceratocystiopsis sp. associate of the MPB). In the struggle to colonize a host tree, pathogen strategies may involve degradation or conversion of toxic host molecules, transport of host defense molecules out of the cell, modification of cell structures to avoid/exclude/sequester toxic host molecules (Katsir et al. 2008; DiGuistini et al. 2011; Wang et al. 2013). In G. clavigera, a number of genes  13 (e.g. cytochromes P450s, dehydrogenases and monooxygenases) potentially involved in a combination of these strategies for detoxifying defense chemicals from P. contorta have been identified or functionally characterized (i.e. ABC-transporters) (DiGuistini et al. 2011; Wang et al. 2013).  Conifers have evolved a complex chemical defense system that can be deployed against a wide range of pests and pathogens. The system produces both constitutive and inducible chemicals that are mainly composed of oleoresin and phenolics. Oleoresin is primarily composed of monoterpenes and diterpenes, but the chemical composition can vary significantly depending on the species of pine or different populations of the same species. For example, in contrast to other pine species, Pinus jeffreyi have very low amounts of terpene components in their resin. Instead, they contain aldehydes, which are diluted with heptane (Mirov and Hasbrouck 1976). Although tree phytochemistry plays a critical role in host selection by bark beetles (Raffa 2001), it also affects the rate at which fungal spores germinate and grow. In general, both terpenes and phenolics have shown deleterious effects on microorganisms (Delorme and Lieutier 1990; Savluchinske Feio et al. 1999; Hofstetter et al. 2005; Wang et al. 2013).      14 1.3 Ophiostomatoids ecology, biology and systematics  Fungi are the second largest group of eukaryotes with an estimate of over 1.5 million species, of which only 5 to 10% have been described (Hawksworth 2001; Spooner and Roberts 2010). Ophiostomatoid fungi are members of Ascomycota, which is the largest fungal phylum with over 64,000 described species (Kirk et al. 2008). In their sexual phase (teleomorph), ascomycetes produce sac-like structures called asci where the ascospores (sexual spores) are produced. The asci are enclosed in fruiting bodies called ascocarps. After ascospores are released, they germinate and produce a mycelium, from which the conidiophores and conidia  (asexual spores) that characterize the asexual phase (anamorph) will develop. In their wide diversity of forms, conidiophores and conidia vary greatly from asci and ascospores. In filamentous ascomycetes, neither the teleomorph nor the anamorph is considered to be an organism by itself, as neither can exist without a mycelium. Therefore, the taxonomy of a fungus must include both the anamorph and teleomorph, but these phases do not always occur together at the same time or under the same conditions. Also, there are many species where only an asexual phase has been observed (e.g. L. longiclavatum). For many years, anamorphs had an independent taxonomy and a separate name (i.e. anamorph genera) ? but when the teleomorph becomes known its name takes precedence over that of the anamorph (Seifert and Samuels 2000). Currently, while each fungus has only one name, researchers have proposed competing names for anamorphs and teleomorphs, following the ?one fungus one name? proposal by the  15 2011 International Botanical Congress (Hawksworth 2011; Wingfield et al. 2012; Hawksworth et al. 2013).   Ophiostomatoids represent an artificial group of 397 accepted species (De Beer et al. 2013). These fungi are highly adapted for being dispersed by insects and colonizing host plants (Wingfield and Seifert 1993; Spatafora and Blackwell 1994; Six 2003; Harrington 2005; De Beer and Wingfield 2013; Wang et al. 2013). They are distributed in the Northern and Southern hemispheres on a wide variety of host plant substrates (Upadhyay 1993). These fungi are highly pleomorphic, and can grow as either mycelia or as unicellular yeast forms. The fungal sexual phase typically has long-necked, flask-shaped ascocarps called perithecia (Upadhyay 1981; De Beer and Wingfield 2013). The asexual structures come in a variety of forms, all of which produce similar sticky conidia (Upadhyay 1981; Six 2003; De Beer and Wingfield 2013). Some fungi like G. clavigera have yeast-like structures inside the mycangia, but form mycelia when grow in the tree phloem and sapwood (Tsuneda and Currah 2006).   The current classification of the ophiostomatoid group is largely based on ribosomal gene DNA (rDNA) that places the 397 accepted species in 12 genera that are named according to the ?one fungus: one name? proposal (Wingfield et al. 2012; De Beer and Wingfield 2013). Six of these genera are classified in the order Ophiostomatales: Ophiostoma sensu lato (s. l.) (including Pesotum, Sporothrix and Ambrosiella), Ceratocystiopsis, Fragosphaeria, Graphilbum, Raffaelea s. l., and Leptographium s. l. (including Grosmannia, and a few other unresolved groups). The other six are classified  16 in the order Microascales: Ceratocystis s. l. (including Thielaviopsis and Ambrosiella), Graphium sensu stricto, Knoxdaviesia, and Sphaeronaemella, Cornuvesica and Custingophora. Among the currently defined genera, Ophiostoma, Leptographium and Ceratocystis include the largest number of species (134, 94 and 72 spp., respectively). Ceratocystis species are characterized by endogenous conidia (Minter et al. 1983; Gebhardt et al. 2005). In contrast, species in the Ophiostomatales are characterized by a variety or a continuum of anamorphs (e.g. Hyalorhinocladiella, Leptographium, Pesotum and Sporothrix) that form exogenous conidia by building apical walls (Minter et al. 1982; Hausner et al. 1993; Jacobs and Wingfield 2001; Massoumi Alamouti et al. 2009). Ceratocystis species have less specific relationships with their beetle vectors than most genera in Ophiostomatales, particularly Grosmannia and Leptographium species, which are always associated with bark beetles and mainly colonize coniferous host trees (Jacobs and Wingfield 2001; Harrington 2005; Duong et al. 2012).  Traditional taxonomy in ophiostomatoid fungi is complicated by their limited range of morphological characteristics and by their convergent evolution for the insect dispersal (Spatafora and Blackwell 1994). Identification based on these structures has confused the taxonomy of these fungi for many years (reviewed by De Beer et al. 2013). For example, in the early 1970?s Ophiostoma and Ceratocystis were considered to be synonymous, based on the features of their perithecia (Upadhyay and Kendrick 1975, Upadhyay 1981, De Hoog and Scheffer 1984, Wingfield et al. 1993). The genus Ophiostoma contains many species with a variety of ascospore shapes, including Ceratocystiopsis and Grosmannia species that until recently were synonymous with  17 Ophiostoma (Zipfel et al. 2006). Anamorph morphology has been preferred for defining fungal species for two reasons. First many species have no known sexual phase or do not produce fruiting body (ascomata) under artificial condition; for example, more than 20 species of Leptographium are known to have teleomorphs (Zipfel et al. 2006) but the sexual states are not known for at least an additional 70 species (e.g. L. longiclavatum associate of the MPB). Second anamorphs have diverse shapes that are easily observed in artificial media. But defining species using only the morphology of asexual phase can be problematic because many species develop a combination of anamorphs (e.g. synnematous and mononematous Leptographium anamorphs in G. clavigera) or reduced and non-distinctive asexual structures (Tsuneda and Currah 2006). Further, for many species, including G. clavigera, anamorphs can degenerate after repeated subculturing on artificial media (Robinson-Jeffrey and Davidson 1968; Tsuneda and Hiratsuka 1984; Okada et al. 1998; Six 2003). Therefore, for species identification and their classification at higher taxonomic level, morphological characterization needs to be complemented by molecular analysis or sufficient diagnostic markers. Analyses of DNA sequence data have redefined the status of several genera and species and have led to the discovery of many unrecognized species. This trend is likely to continue as more sequence, genomic and metagenomic data become available for this ecologically important fungal group (DiGuistini et al. 2011; Hintz et al. 2011; Massoumi Alamouti et al. 2011; Haridas et al. 2013; Khoshraftar et al. 2013).  Since rDNA was first used to estimate phylogenetic relationships, the classification of genera has evolved rapidly and names of most species have been changed  18 (Upadhyay 1981; Upadhyay 1993; De Beer et al. 2013). rDNA studies have shown that most of the genera in Ophiostomatales are polyphyletic, suggesting that similar morphological characteristics and an intimate association with beetles have originated more than once in Ophiostomatales genera (Spatafora and Blackwell 1994; Cassar and Blackwell 1996; Farrell et al. 2001; De Beer and Wingfield 2013). Phylogenetic analyses of rDNA sequences placed the genera Ophiostoma and Ceratocystis into two different ascomycetes orders, but kept the synonymy of the genera Ophiostoma and Ceratocystiopsis. More recent studies have shown that Ophiostoma, Leptograhium, Raffaelea and Ambrosiella species are polyphyletic; however, the partial rDNA data are not adequate to define monophyletic groups inside these genera (Zipfel et al. 2006; Massoumi Alamouti et al. 2009; Duong et al. 2012; De Beer and Wingfield 2013). Zipfel et al. (2006) used DNA sequence data from combined partial nuclear large subunit (nLSU) rDNA and !-tubulin genes, and suggested three well-supported, sexual-monophyletic clades in Ophiostoma. They re-introduced the teleomorph-genus Ceratocystiopsis to include Hyalorhinocladiella anamorphs and species with short perithecial necks. They proposed grouping together species with Leptographium anamorphs, including G. clavigera and close relatives, and to accommodate these species they re-instated the teleomorph-genus Grosmannia. However, when they tested the monophyly of Leptographium-forming species they did not consider some close relatives including Ambrosiella, Raffaelea and the monotypic genus Dryodomyces that are all associated with the ambrosia (fungal-feeding) beetles (Gebhardt et al. 2005).  Previous nuclear small subunit (nSSU) rDNA phylogenies have shown that these fungi are dispersed within the Leptographium-forming species (Cassar and Blackwell 1996;  19 Jones and Blackwell 1998; Farrell et al. 2001; Rollins et al. 2001; Gebhardt et al. 2005). In chapter 2 we address some of these issues, and extend this work by generating the largest multigene phylogeny for a diverse set of genera in the ophiostomatales. We show that the Leptographium-forming species, which also represent the most common associates of bark beetles, form a clade separated from ambrosia-beetle associates. We also show that neither Ambrosiella nor Raffaelea are well-defined within ophiostomatales and both are likely to represent additional distinct genera (Massoumi Alamouti et al. 2009). Based on a large rDNA phylogeny of 216 Ophiostomatales taxa, De Beer and Wingfield 2013 suggested that there are at least 24 distinct groups (including Ambrosiella and Raffaelea clades resolved by our mutligene phylogeny in chapter 2) that might represent distinct genera. However, the rDNA data are not sufficient to define the unresolved groups, and thus for now they remain in the Ophiostoma s. l., Leptographium s. l. and Raffaelea s. l. until a multigene phylogeny becomes available for a larger number of taxa to define appropriate new combinations for these species.    1.4 Fungal species concept and recognition  Defining a ?species? is fundamental for studying speciation and for understanding the biology and ecology of organisms. It is also essential for practical purposes such as pest controls and quarantine regulations. Despite disagreements about the origin and maintenance of species, there is a consensus view that species are ?segments of separately evolving metapopulation lineages? (Mayden 1997; de Queiroz 1999; 2007).  20 During speciation processes, diverse events occur (e.g. initial separation, diagnosable characters, monophyly and reproductive lineages), and describing each of these events forms the bases for alternative species criteria (de Queiroz 1998). Properties used in each criterion can arise at different times and in different orders that can vary with factors like geography, demography, drift, selection and gene flow. For example, fixation of a nucleotide character in a lineage segment that originates from a large subdivision of the ancestral species may take longer than a lineage originating from a founder event or under strong positive selection. Different species criteria can lead to different conclusions regarding the boundaries and number of recently diverged species in different organisms (Avise 2004).  In fungi the criteria most often used to recognize and delimit species emphasize morphological divergence (morphological species criterion: MSC) and, less commonly, reproductive isolation (biological species criterion: BSC) (Brasier 1987; Hawksworth et al. 1995). Morphologically simple organisms like fungi may become genetically isolated due to selection and/or random genetic drift before morphological phenotypes or mating-behavioral differences have accumulated (Brasier 1987; Taylor et al. 2000). For a number of reasons, BSC cannot be a general method of choice for fungi. BSC focuses on the ability of a fungal species to interbreed in nature or in laboratory conditions. However, not all fungi can be cultured or mated in the laboratory, and no sexual stage is known for at least 20 percent of described fungi (Geiser et al. 1998). For example, G. clavigera can be cultured but not mated, and no sexual state is known for two of its close relatives, L. longiclavatum and L. terebrantis. However, failure to find the  21 sexual stage for a fungus does not mean that it does not exist; sexual reproduction has been demonstrated from crosses of fungi that were originally thought to be asexual (Hull et al. 2000; Magee and Magee 2000), and recombining population structures have been reported using multilocus nucleotide analyses (Burt et al. 1996; Geiser et al. 1998; Pringle et al. 2005; Matute et al. 2006). Some species that lack a sexual stage may have developed other strategies for overcoming the resulting shortfall of genetic recombination (Lynch et al. 1993; Butcher 1995), but it is difficult to conclusively rule out the presence of this stage (Pawlowska and Taylor 2004).   Phylogenetic and population genetic methods that emphasize nucleotide divergence (phylogenetic species criterion: PSC) define species as the ?smallest monophyletic clade of organisms that share a derived character state? (Avise and Ball 1990). PSC became broadly applied in species recognition in the late 1980?s, with the discovery of the Polymerase Chain Reaction (PCR) and the availability of many DNA characters through direct sequencing. These methods have resulted in an increasing number of reported cryptic species and species complexes in all taxonomic groups of living organisms (Bickford et al. 2005). Based on the outcomes of PSC in fungi, it is expected that most fungal species whose current description is based on morphology actually consist of more than one closely related cryptic or sibling species (Taylor et al. 1999; Taylor et al. 2000; Kasuga et al. 2003).   Accuracy in PSC requires information from multiple loci. Relying on the concordance of more than one gene genealogy (i.e. phylogenetic species recognition by  22 genealogical concordance: PSCGC) provides a higher resolution and can avoid subjectively determining the boundaries of species (Avise and Ball 1990; Slatkin and Maddison 1990; Baum and Shaw 1995; Taylor et al. 2000; Pringle et al. 2005; Matute et al. 2006). Methods that summarize population genetic and genealogical patterns across many loci are essential for diagnosing recent evolutionary lineages for which divergence time has not been long enough to detect complete reciprocal monophyly at many loci (Hudson and Coyne 2002; Maddison and Knowles 2006; Knowles and Carstens 2007).  1.4.1 Defining species boundaries and population structures in G. clavigera  During a MPB outbreak in the 1960?s, Robinson described G. clavigera from the beetle and its primary host trees P. contorta and P. ponderosa (Robinson-Jeffrey and Davidson 1968). The fungus was first named Europhium clavigerum (Robinson and Davidson 1968), then Ophiostoma clavigerum, and was renamed G. clavigera [= O. clavigerum (Robinson and Davidson) Harrington] (Zipfel et al. 2006). The description of this species is based on an isolate recovered from the sapwood of P. ponderosa from Cache Creek, BC, Canada (Robinson and Davidson 1968). Later, G. clavigera was also reported from JPBs in western United States (Six and Paine 1997). Grosmannia clavigera is predominantly haploid except for a transient diploid phase occurring during sexual reproduction, which likely requires pairing of the two opposite mating types (heterothallic) (DiGuistini et al. 2011; Tsui et al. 2013). The sexual structures are rarely observed in nature and not produced under artificial conditions (Robinson-Jeffrey and Davidson 1968; Lee et al. 2003; Massoumi Alamouti et al. 2011).   23  Given the rarity of G. clavigera teleomorphs, mating incompatibility among strains cannot be determined, and the morphology of anamorphs is often unstable and unreliable for the identification of this species. The fungus is pleoanamorphic possessing several types of synnematous and mononematous anamorphs (Lee et al. 2003, Six et al. 2003, Tsuneda and Hiratsuka 1984, Upadhyay 1981). Furthermore, Tsuneda and Hiratsuka (1984) have observed holoblastic and annellidic-yeast states for this species. In addition to its pleoanamorphic nature, G. clavigera produces a broad range of conidiophores and conidia size and shape that overlap with those of G. aurea, G. robusta, L. longiclavatum, L. pyrinum, L. terebrantis and L. wingfieldii (Six et al. 2003, Lee et al. 2003). Molecular tools, including restriction fragment length polymorphism (RFLP), DNA fingerprinting and multigene phylogenies, have also shown that these seven Leptographium-forming fungi form a complex of closely related species that are very similar in morphology, found on the same host trees, and associated with bark beetles (Jacob and Wingfield 2001; Lim et al. 2004, Six et al. 2003). Therefore, in G. clavigera morphologically and genetically similar individuals can be recovered either from infected pine trees or from the beetles throughout the beetle distribution. Given the difficulty to identify the fungi in G. clavigera-species complex, it is not surprising to find misidentifications in the literature. The multigene phylogenies (i.e. rDNA and three house-keeping genes); however, have improved the molecular identification of G. clavigera distinguishing the pathogen from most close relatives, except for L. terebrantis (Six et al. 2003; Lim et al. 2004; Roe et al. 2010). Therefore, controversy had remained  24 over the species status of L. terebrantis and its phylogenetic relationship with G. clavigera.  According to morphological criteria and DNA fingerprinting analysis G. clavigera has a wide geographical range across Canada (British Columbia and western Alberta) and the Unites States (Washington, Oregon, and California inland to South Dakota, Colorado and New Mexico) where the major MPB and JPB hosts and other pines are found and attacked by the beetles (Upadhyay 1981; Zambino and Harrington 1992; Six and Paine 1999; Six et al. 2003; Lim et al. 2004). However, it seems reasonable to argue that the association of this fungal species with distinct host beetle and tree species might be an artifact of the morphological species recognition and shortcoming of the molecular tools that have been used in earlier systematic and population studies. Further, AFLP (amplified fragment length polymorphism) markers have suggested that two genetically distinct groups exist in the epidemic populations of G. clavigera associated with the MPBs (Lee et al. 2007). However, the AFLP groups have not been supported by our gene genealogies presented in chapter 3 or by microsatellite markers (Massoumi Alamouti et al. 2011; Tsui et al. 2012).  The beetle-pathogen population structures have been well studied in epidemic regions using different molecular tools (Lee et al. 2006a; Mock et al. 2007; James et al. 2011; Roe et al. 2011; Gayathri Samarasekera et al. 2012; Tsui et al. 2012). Based on the fungal surveys G. clavigera epidemic populations are divided into four groups according to four major geographic regions in northwestern BC, southern BC,  25 northeastern BC/Alberta, and Rockies. A north-south genetic structure is concordant among different studies for both the beetle and the pathogen pointing to the fungal pathogen?s dependence on the beetle for the dispersal. On contrary to the epidemic regions, studies on the fungal localized populations in western and southern USA is limited to few allozyme markers for the California populations of G. clavigara associated with JPBs, showing low a genetic diversity within and among the populations.   1.5 Fungal genomics and their importance in G. clavigera  Sequencing technologies that enable the acquisition of whole genome sequences provide new approaches to address questions related to fungal systematics, evolution and speciation, as well as to identify genomic regions or particular genes involved in fungal divergence, host specialization and/or pathogenicity. Fungi are the eukaryotic group with the greatest number of completely, or nearly completely, sequenced genomes (http://www.ncbi.nlm.nih.gov/genomes/leuk s.cgi) (Stajich et al. 2009). This is not only due to their importance in ecology, medicine, agriculture and biotechnology, but also because their genomes are among the smallest and most compact eukaryotic genomes.  Saccharomyces cerevisiae was the first fungal genome to have been fully sequenced and annotated, making a major contribution to the basic understanding of eukaryote cell physiology, genetics and biochemistry (Goffeau et al. 1996). Today  26 sequencing advances have led to over a hundred fungal genome sequences being available for the phyla Ascomycota, Basidiomycota, Zygomycota and Chytridiomycota. Among the currently sequenced genomes, more than 30 represent important plant pathogens like Cochliobolus, Fusarium, Mycosphaerella, Magnaporthe, Sclerotinia, and Ustilago species, most of which have been sequenced by Sanger sequencing (Dean et al. 2005; Cuomo et al. 2007; Coleman et al. 2009; Ma, H Charlotte van der Does, et al. 2010; Raffaele et al. 2010; Schirawski et al. 2010; Spanu et al. 2010; Stukenbrock et al. 2010; Goodwin et al. 2011; Klosterman et al. 2011; Croll and McDonald 2012; Raffaele and Kamoun 2012; Stukenbrock et al. 2012). Genome analyses of these fungi have shown a remarkable diversity in genome size (i.e. 19?160 Mb), numbers of chromosomes, structural organization, repeat-induced point (RIP) mutations, transposable element activity and protein similarity (Galagan et al. 2005; Raffaele and Kamoun 2012). For examples, the genomes of rice blast fungus Magnaporthe grisea and its non-pathogenic relative Neurospora crassa have an average amino acid identity of only 47% and almost no conserved synteny (Berbee and Taylor 2001; Dean et al. 2005). These observations have been a powerful tool for inferring genome evolution and speciation process in plant pathogens, and to identify the molecular bases of host adaptation and pathogenicity.  Functional and genomic studies in ophiostomatoid fungi are at an early phase and at the time of writing are limited to three pathogens G. clavigera, O. ulmi and O. novo ulmi and one saprophyte O. piceae (DiGuistini et al. 2009; Forgetta et al. 2013; Haridas et al. 2013; Khoshraftar et al. 2013). G. clavigera, owing to its importance as the MPB  27 symbiont and a pine pathogen was the first beetle-tree-associated fungus for which the genome sequence was published. The genome was assembled from a combination of Sanger paired-end (PE) reads (0.3-fold coverage), 454 single reads (7.7-fold coverage), and Illumina paired-end reads (100-fold coverage), resulting in a high-quality draft genome sequence of 32.5 Mb (DiGuistini et al. 2009). Building on the draft sequences, the G. clavigera genome was manually finished, yielding 18 supercontigs with a total length of 29.8 Mb and 8,312 gene models that were supported by expressed sequence tag (EST) and RNA-seq data (DiGuistini et al. 2011). These genomic resources provided major insights into the biology of this fungus and the molecular mechanisms involved in its host-pathogen interactions (Hesse-Orce et al. 2010; DiGuistini et al. 2011; Lah et al. 2013; Wang et al. 2013).   Terpenoids, particularly monoterpenes, are among the most abundant defense chemicals that protect pine trees against pests and pathogens; however, beetle-tree-associated fungi, like G. clavigera, have evolved the ability to tolerate these rather toxic environments and to grow and become established inside the living trees (Keeling and Bohlmann 2006; Boone et al. 2011; Bohlmann 2012). Functional genomics and transcriptomic data have shown that monoterpenes and P. contorta extracts induce a stress response in G. clavigera and activate a ~100-kb cluster of genes. The cluster contains genes involved in !-oxidation pathway, as well as monooxygenases and alcohol/aldehyde dehydrogenases, that may be involved in detoxification and/or metabolism of host-defense chemicals (DiGuistini et al. 2011). A number of other genes that may be important for tree colonization by the pathogen include cytochromes P450  28 and ATP-binding cassette (ABC)-transporter families (Hesse-Orce et al. 2010; DiGuistini et al. 2011; Lah et al. 2013). A pleiotropic drug resistance transporter in G. clavigera has been functionally characterized and shown to be required to control monoterpene levels within the cells, enabling the fungus to grow on media containing these compounds (Wang et al. 2013).    1.5.1 Next generation sequencing and fungal population genomics  Despite (or indeed because of) much progress in the area of genome sequencing and functional genomics, even more information can be gained from sequencing not just one genome per species but rather from sequencing and comparing the genomes of different strains from the same or closely related species. Such work was made feasible by next-generation sequencing (NGS) technologies and supporting analysis methods that focus on addressing questions relevant to the biology/ecology of eukaryotic microorganisms (Nowrousian 2010). Different NGS technologies offer differences in read lengths and numbers of reads (e.g. ~450+ bp reads for Roche 454 and ~50-100-bp reads for Illumina), but all generate millions to hundreds of millions of reads per sequencing run at a substantially lower cost than Sanger sequencing (Mardis 2008; Metzker 2010). Despite the short sequence reads, NGS platforms have greatly facilitated genome sequencing, first for prokaryotes, and, within the last few years, for eukaryotic genomes (Reinhardt et al. 2009; Li et al. 2010; Nowrousian et al. 2010). NGS of different strains of the same species, when a reference genome is available (i.e. ?resequencing?), can take full advantage of the high throughput, because the sequences  29 of the additional isolates can be mapped with high confidence to the reference sequence. This approach works well even for sequencing of large genomes. Sequence reads that are mapped to a reference can be used for detecting genomic variants that include single nucleotide polymorphisms (SNPs), insertions/deletions (indels) or structural variants (Wang et al. 2008; Wheeler et al. 2008; Eck et al. 2009). A number of novel algorithms for mapping NGS reads (i.e. aligning reads to a reference genome) have been developed to address issues like alignment accuracy, uniqueness and confidence, and identifying SNPs and small indels (Li et al. 2009; Li and Durbin 2009; Trapnell and Salzberg 2009). For de novo genome assembly of eukaryotes including the G. clavigera reference genome, NGS was once used in combination with Sanger sequencing (DiGuistini et al. 2009); but today, longer reads, the ability to sequence both ends of a DNA fragment (i.e. paired-end sequencing), and developments in assembly programs have allowed generating de novo draft genomes of eukaryotes like the giant panda and the filamentous fungus Sordaria macrospora (Korbel et al. 2007; Simpson et al. 2009; Li et al. 2010; Miller et al. 2010; Nowrousian et al. 2010; Nowrousian 2010).  Since the introduction of PCR thirty years ago, increasing amounts of data for RFLP, AFLP, microsatellites, and small-scale DNA sequencing have broadened the range of questions open to empirical analysis in evolutionary and population genetic studies. With the recent abundance of genome-wide SNP data, and now the advent of genomes sequenced across populations, evolutionary studies have become a data-driven discipline. Population genomics has enabled comprehensive views of genome-wide patterns of sequence variation within and between closely related species (i.e.  30 polymorphisms and divergence respectively), and of the evolutionary relationships between such species. Further, it has provided new insights into biological and ecological attributes of plant pathogens, and a rich resource for genome-wide assessment of adaptive evolution and functional variations. For filamentous plant pathogens, comparative genomics between different strains of the same species or between the closely related species have revealed genomic features that can have important roles in lifestyle and host ranges of these fungi. Examples include expansion/contraction of specific gene families (e.g. polyketide synthases, lytic enzymes and putative transporters), gene pseudogenizations or deletion, repetitive sequences, and distinct genomic regions (e.g. telomeres, gene clusters and rapidly-evolving genomic islands), or dispensable chromosomes (reviewed by Stukenbrock et al. 2011; Raffaele and Kamoun 2012; Stukenbrock and Bataillon 2012). Other mechanisms that drive genetic diversity and shape the genomes of plant pathogens include DNA point mutations (or SNPs when shared among different strains of the same species) and recombination.  1.6 Single nucleotide polymorphisms   As in many other organisms, SNPs represent a useful kind of genetic variation across the genomes of plant pathogens (Collins et al. 1998; Tyler et al. 2006; Ma et al. 2010; Neafsey et al. 2010; Amselem et al. 2011; Andersen et al. 2011; Desjardins et al. 2011; Klosterman et al. 2011; McCluskey et al. 2011; Stukenbrock et al. 2011; Xue et al. 2012; Condon et al. 2013). Generally, nucleotide changes that result in amino acid  31 replacements are called ?nonsynonymous? polymorphisms, while substitutions in coding regions that do not change an amino acid are called ?synonymous?. Variations can also occur within non-coding regions of the genome such as intergenic regions, upstream (i.e. putative regulatory regions and/or promoters) and downstream of a gene, or within introns (Hartl and Clark 2007). SNPs in the coding regions of genes or in regulatory regions are more likely to cause functional differences than nucleotide variants elsewhere.  Genomic analyses have shown that SNP frequencies can vary among species and strains, and can also vary locally along chromosomes (Raffaele and Kamoun 2012). For example, the polymorphisms between two Fusarium graminearum strains were most frequently found near telomeres and within discrete chromosome regions that are likely involved in plant-fungus interactions (Cuomo et al. 2007). In other cases, SNP frequencies were more homogenous across the genomes, but local biases (e.g., in gene or genomic regions) in the ratio of synonymous to non-synonymous variants have been observed and interpreted as a signature of positive selection (Fedorova et al. 2008; Raffaele et al. 2010; Stukenbrock et al. 2010; Kemen et al. 2011; Stukenbrock et al. 2011; 2013). Therefore, population and comparative genomics approaches offer the possibility not only of capturing the evolutionary history of species, but also of identifying signatures of selection in genes involved in ecological divergence and host specificity.    32 1.7 Divergent adaptive evolution in fungi  In plant pathogens, speciation has been generally associated with ecological divergence (e.g. host shifting or specificity) or with large changes in genomic structure and composition due to hybridization (reviewed by Stukenbrock 2013). Genes or genomic regions important in fungal divergence and speciation can stand out because of an increased fixation of adaptive mutations ?? ecological divergence (e.g. host specificity) in particular will leave footprints of positive selection.   The relative abundance of non-synonymous and synonymous polymorphisms can reflect the effect of natural selection, which is generally expected to remove slightly deleterious non-synonymous variants in coding sequences (i.e. purifying selection). Therefore a significant increase in the ratio of non-synonymous to synonymous changes may reflect protein-coding genes that are favored by natural selection (i.e. positive selection) (Nielsen and Yang 1998; Yang and Swanson 2002; Yang 2007). Some analysis are designed to detect positive selection at individual coding sites within a gene or between lineages, which may be more sensitive than earlier methods that would average the ratio of non-synonymous and synonymous along a gene or between species (Yang and Nielsen 2000; Yang 2007). If genomic data from different strains and related species are available, rates of non-synonymous and synonymous substitutions between species can be assessed and compared with patterns of polymorphisms. Such tests allow the detection of ancient selection in homologous genes of related species (Fedorova et al. 2008; Li et al. 2008; Raffaele et al. 2010; Stukenbrock et al. 2010;  33 Kemen et al. 2011; Stukenbrock et al. 2011; 2013). For example, Stuckenbrock et al. (2010) using population genomic approaches within and between species have shown the role of beneficial mutations in host specialization of the wheat pathogen Mycosphaerella graminicola. Positive selection has been shown to have a significant impact on the evolution of filamentous plant pathogens, particularly on the effector genes that are responsible for modulating host physiology and enabling colonization of plant tissue (Hogenhout et al. 2009; Raffaele and Kamoun 2012).    Comparative and population genomic analyses are becoming more common in fungi. Such work assesses the genomic-wide pattern of variation within and between closely related species, and scans for the signature of positive selection across protein-coding genes. To date, however, most of these studies have analyzed crop pathogens and have not investigated beetle-tree-associated fungi. Due to their obligatory symbiotic association with the beetle vector and colonization of host-tree niches, ophiostomatoid species like G. clavigera represent an interesting group of organisms in which to study ecological divergence and assess genome-wide molecular adaptation. Despite its ecological importance, evolutionary and population studies of this fungal group are limited to a few protein-coding genes and neutral markers (i.e. AFLP and microsatellites), and little is known about the level and pattern of nucleotide polymorphisms and their functional importance in fungal fitness. This limitation is changing as more genome data becomes available (DiGuistini et al. 2011; Massoumi Alamouti et al. 2011; Forgetta et al. 2013; Haridas et al. 2013; Khoshraftar et al. 2013).    34 1.8 Overview and purpose of this thesis  As an important pathogen of P. contorta forests in western North America, G. clavigera forms a tight symbiotic association with two sibling species of bark beetles: MPB and JPB. Despite their close phylogenic relationship, the two beetle vectors infest distinct host-tree species, where they spend most of their whole life under the bark, except during short dispersal flights. Thus, these organisms have a high degree of intimacy with their hosts that could reinforce local adaptive structures and consequent host differentiation of their accompanying fungi. However, important ecological and biological attributes of the symbiotic fungi may not have been detected because their species boundaries had not been clearly defined. Grosmannia clavigera shows morphological, ecological and evolutionary similarities with a number of pine-infesting species, leading to confusing taxonomic and phylogenetic relationships at the interspecific and generic levels. While G. clavigera and its close relatives have been re-instated into the genus Grosmannia inside the Ophiostomatales, the genus is intermixed with another ecologically distinct group of Ophiostomatales that are collectively called ambrosia fungi. Because the ambrosia fungi represent different genera, their inclusion in phylogenies made Grosmannia paraphyletic. Therefore the main goals of this thesis were to clarify the phyogenetic status of the genus Grosmannia within Ophiostomatales, to define species boundaries in G. clavigera and its evolutionary relationship with other closely related pine-infesting species, and finally to find genes or genomic regions that might be involved in species divergence and host specificity.    35 In an evolutionary sense, plant-fungal interactions are known to be older than interactions between fungi and insects. Studies of beetle-associated microflora have generally focused on reporting the fungal associates of different bark beetle species. However, the host tree that directly nourishes the fungus may have a more important role in the speciation and diversity of ophiostomatoid fungi than does the beetle that vectors the fungus. Given this likelihood, I hypothesize that G. clavigera is a species complex composed of more than one phylogenetic species, each being associated with a distinct host-tree species. A phylogenetic species may be associated with MPB or JPB beetles, depending on the tree on which the beetle feeds. I predict that adaptation to the specific chemistries of host tree will emerge as an important feature in the evolutionary divergence of these fungi, and finally that genes potentially involved in host-pathogen interactions have diverged in response to selection in different host environments.           36 Chapter 2 Multigene phylogeny of ophiostomatoid fungi associated with bark and ambrosia beetles  2.1 Introduction  Bark and ambrosia beetles are weevils (Coleoptera: Curculionidae) in the subfamilies Scolytinae and Platypodinae (Farrell et al. 2001; Marvaldi et al. 2002). They spend part of their life cycles in galleries that they mine under the bark (scolytid bark beetles) or in the wood of trees (scolytid and platypodid ambrosia beetles); and they vector diverse fungi that colonise the wood (Batra 1966). In coniferous forests, the most common fungi in beetle galleries are filamentous ascomycetes that are generally known as ophiostomatoids (Six 2003; Harrington 2005). In many countries, ophiostomatoid fungi include species that are involved in tree diseases, cause considerable value loss to the wood product industry, and are considered as quarantine pests (Wingfield et al. 1993; Alfaro et al. 2007; Fraedrich et al. 2008).  The current molecular classification of ophiostomatoids is largely based on nuclear rDNA (rRNA gene). It places the more than 140 species into morphologically similar, non-monophyletic teleomorph genera: Ceratocystiopsis Upadhyay and Kendrick, Ceratocystis Ellis and Halstead, Grosmannia Goid?nich emend. de Beer, Zipfel and Wingfield and Ophiostoma H. and P. Sydow, as well as into a number of anamorph genera (Hausner et al. 2000; Zipfel et al. 2006). Species of genus Ceratocystis are  37 sensitive to the antibiotic cycloheximide; they are related to the Microascales, and are characterised by the Thielaviopsis Went anamorph and by phialidic conidia that are produced by the ring wall-building of the collarette (Minter et al. 1983). In contrast, species that are tolerant to cycloheximide are characterised by a variety of anamorphs that form conidia by building apical walls (e.g., Hyalorhinocladiella, Leptographium, Pesotum and Sporothrix); they are placed into the genera Ceratocystiopsis, Grosmannia and Ophiostoma of the Ophiostomatales (Hausner et al. 1993; Spatafora and Blackwell 1994; Zipfel et al. 2006). Ceratocystis species have less specific relationships with their beetle vectors than do members of Ophiostomatales teleomorph genera and related anamorphs, which are always associated with scolytid bark beetles (Kirisits 2004).  Ophiostomatoid fungi that were originally called ?ambrosia? are now classified in the anamorph genera Ambrosiella von Arx and Hennebert, Raffaelea von Arx and Hennebert and Dryadomyces Gebhardt (Batra 1967; Gebhardt et al. 2005). ?Ambrosia? represents heterogeneous groups of fungi that include yeasts and filamentous fungi (Batra 1963; Francke-Grosmann 1967). Except for two basidiomycetes (Batra 1972; Hsiau and Harrington 2003), all are ascomycetes, most of which belong to the genera Ambrosiella and Raffaelea. ?Ambrosia? fungi typically form dense mats of hyphae or clusters of small conidiophores (sporodochia) with conidia that germinate into mass of highly vacuolated sprout cells in beetle galleries; this is often referred to as the ?ambrosia phase? (Batra 1967). The majority of ?ambrosia? fungi have a symbiotic relationship with platypodid or scolytid ambrosia beetles (Batra 1967; Kubono and Ito  38 2002; Gebhardt et al. 2004; Gebhardt et al. 2005; Harrington et al. 2008). These beetles bore galleries into the wood of host trees, and dependent on their fungal symbionts to exploit the nutrient-poor xylem (Batra 1966; Roeper 1995). Many of the beetles have developed mycangia or similar structures that may support transferring fungi to new hosts (Batra 1966; Six 2003). In contrast to ambrosia beetles, scolytid bark beetles feed on the phloem of trees. While bark beetles seem to be less dependent on their fungal associates for nutrition, some may supplement their diets by consuming the fungal associates that they carry on their exoskeletons, and in their guts or mycangia, after the fungi have grown in the beetle galleries (Harrington 2005; Six 2003). Ambrosiella species are considered to be typical symbionts of ambrosia beetles; however, a number of Ambrosiella fungi are also reported from certain bark beetles (Batra 1967; Krokene and Solheim 1996; Rollins et al. 2001; Kirisits 2004). Confusingly, then, the term ?ambrosia? can refer to specific fungal associates and to a particular fungal morphological form (ambrosia phase) in beetle galleries that supports the beetle and its progeny developments (Hartig 1844; Batra 1967).  The taxonomy of ?ambrosia? fungi has been re-evaluated, because their classification was originally established using morphological characteristics that are poorly defined in artificial media, and because most are known only by their asexual state (Batra 1967; Gebhardt et al. 2005). For species that lack sexual structures, defining species using only the morphology of asexual phase can be problematic, because many species develop a combination of anamorphs or reduced and non-distinctive asexual structures (Tsuneda and Currah 2006). The genera Ambrosiella and  39 Raffaelea were differentiated based on the morphology of conidiogenous cells (von Arx and Hennebert 1965); Raffaelea have a series of cicatricial conidial scars, while Ambrosiella does not. However, applying electron microscopy to a number of Raffaelea and Ambrosiella, as well as to other asexual ophiostomatoids, has begun to reveal details of conidiogenesis that are not visible by light microscopy (Gebhardt et al. 2005; Gebhardt and Oberwinkler 2005).  Molecular phylogenies are clarifying the taxonomic status of most ?ambrosia? fungi among ascomycetes, and specifically in the ophiostomatoids (Cassar and Blackwell 1996; Jones and Blackwell 1998; Farrell et al. 2001; Rollins et al. 2001; Gebhardt et al. 2005). Nuclear small subunit (nSSU) rDNA phylogenies indicate that both Ambrosiella and Raffaelea are polyphyletic, suggesting that similar morphological characteristics and an intimate association with beetles have originated more than once in these genera (Cassar and Blackwell 1996; Farrell et al. 2001). Supporting the initial phylogenies, Gebhardt et al. (2005) showed that species of Ambrosiella in the order Microascales are also morphologically distinct from those in the order Ophiostomatales, suggesting that the taxonomic status of genus Ambrosiella should be re-evaluated. Despite the close relationship of genera Ambrosiella and Raffaelea with the ophiostomatoid fungi, they were not included in comprehensive ophiostomatoid fungi phylogenies (Hausner et al. 2000; Zipfel et al. 2006).   In the work described here, we address where the Ambrosiella and Raffaelea species should be placed within ophiostomatoids and their relationship to  40 Ophiostomatales genera. We report a multigene phylogenetic analysis of filamentous ?ambrosia? fungi that includes twenty-five species from the genera Ambrosiella, Raffaelea and Dryadomyces and thirty species from other ophiostomatoid clades. Our results indicate the limitations of using classical morphological traits and molecular analyses based on single genes to address the taxonomy of these fungi. Finally, we discuss whether the ecological characteristics of the beetle vectors (i.e., bark vs. wood), which originally contributed to the classification of ?ambrosia? fungi, appear to be phylogenetically significant.  2.2 Materials and methods  2.2.1 Taxon sampling  Twenty-two strains from the genera Ambrosiella, Raffaelea and Dryadomyces were requested from culture collections (Table 2.1). We also included three undescribed Ambrosiella species (sp. 1, sp. 2 and sp. 3) isolated from spruce-colonising bark beetles in Canada and Europe (Krokene and Solheim 1996; Massoumi Alamouti et al. 2007), as well as an undescribed species from the mycangia of ambrosia beetles Trypodendron rufitarsus and T. lineatum collected from lodgepole pines infested by the mountain pine beetle (MPB), Dendroctonus ponderosae in BC (Kuhnholz 2004). All the fungal strains are maintained at the Breuil Culture Collection (University of British Columbia, BC).   41 2.2.2 Morphological investigation  Fungal fruiting structures produced from one- to four-week-old cultures grown on Difco malt extract agar (MEA; 20g Difco malt extract, 10g Difco agar and 1L distilled water), potato dextrose agar (PDA) and MEA enriched with 1% Difco yeast extract (YEMEA), were mounted in water and observed using a Zeiss Axioplan compound light microscope. For scanning electron microscopy (SEM), small wood blocks (5x2x5 mm) bearing fungal structures were fixed using the method described by Lee et al. (2003). After fixation, samples were dried with a Blazers CPD 020 critical point drier. They were coated twice with gold palladium using a Nanotech Semprep II sputter coater and examined using a Hitachi S4700 scanning electron microscope.  2.2.3 DNA extraction, PCR amplification and sequencing   DNA was extracted from mycelia grown on Oxoid MEA [33 g malt extract agar (Oxoid CM59), 10 g agar ?tech. No.3? and 1L distilled water] plates overlaid with cellophane (gel dry grade, BioRad) following the method described by M?ller et al. (1992). The nSSU was amplified and sequenced with primers NS1 and NS4 (White et al. 1990), and the nuclear large subunit (nLSU) region was amplified and sequenced with ITS3 or NL1/LR3 or LROR (Vilgalys and Hester 1990; O?Donnell 1992). The partial !-tubulin gene (!-tubulin) was amplified and sequenced using the primer set BT2E/BT12 (Kim et al. 2004). PCR amplification was performed as described by Kim et  42 al. (2004). PCR products were purified with a Qiaquick PCR Purification Kit (Qiagen, Ont, Canada). Sequencing was performed on an ABI 3700 automated sequencer (Perkin-Elmer, Foster City, CA) at the DNA synthesis and Sequencing Facility, Macrogen (Seoul, Korea). GenBank accession numbers of the new sequences obtained are shown in Table 2.1.  2.2.4 Phylogenetic analysis  Sequences from the representatives related to ophiostomatoid fungi in the Ophiostomatales and Microascales, as well as those representing Xylariales and Hypocreales were included in the analysis (Table 2.1). Sequences were aligned using MAFFT (Katoh et al. 2002) and then manually adjusted with PHYDIT version 3.2 (http://plasza.snu.ac.kr/~jchun/phydit). The flanking regions were excluded from the analysis because sequence length varied with species. Phylogenetic analysis were conducted for the three loci (nSSU, nLSU and !-tubulin) under both maximum parsimony (MP) methods of PAUP*4.0b10 (Swofford 2003) and Bayesian inference of MrBayes v. 3.1.2 (Ronquist and Huelsenbeck 2003). Concordance of the three different gene datasets was evaluated with the partition homogeneity test (PHT) implemented with PAUP*4.0b10, using 1000 replicates and the heuristic general search option  (William and Ballard 1996; Swofford 2003). Taphrina populina and Penicillium expansum were assigned as the outgroup taxa (Jacobs et al. 2003; Gebhardt et al. 2005).  43  For parsimony analysis, all characters were equally weighted and unordered. Separate analyses were conducted with gaps treated as missing data and as a fifth character state (Swofford 2003). The MP trees (MPTs) were identified by heuristic searches with 100 random stepwise addition replicates and tree-bisection-reconnection branch-swapping algorithms. Statistical support for the branches was assessed by bootstrap analysis (BS) using 1000 MP heuristic searches with ten random sequence addition replicates for each bootstrap replicate. Bayesian inference of phylogeny was calculated based on a Markov chain Monte Carlo analysis with the general time reversible (GTR+I+G) substitution model as determined by AIC criteria of Modeltest (Posada and Crandall 1998). The proportion of alignment sites was assumed to be invariable with gamma-distributed substitution rates of the remaining sites. Four simultaneous Markov chains were run from random starting trees for 1 000 000 generations and sampled every 100 generations (generating 10 001 trees). The first 5000 trees were discarded as burn-in, and inferences of posterior probability (PP) were calculated from 5001 trees.         44 2.3   Results  2.3.1 Sequence analysis  For multigene phylogenetic analysis, we generated 59 rDNA and 35 !-tubulin new sequences on the genera Ambrosiella, Raffaelea, Dryadomyces, Ophiostoma and Ceratocystis and retrieved 41 sequences of other ophiostomatoid taxa from GenBank (Table 2.1). We were able to amplify the target loci in all species in the analysis, except for the nLSU locus in R. sulcati and the !-tubulin locus in A. gnathotrichi and R. arxii. From ophiostomatoid taxa, no significant length variations were observed in nSSU and nLSU amplicons, whose lengths varied from 831?833 and 506?539 nucleotides, respectively. However, !-tubulin sequences varied from 550 to1095 nucleotides. This region contained four exons and three introns. Sequences of the four exons were of equal length for all ophiostomatoid taxa in the analysis, whereas sequences of the three introns varied highly in both nucleotide composition and length. Some taxa lacked either one or two introns, which accounted for the large difference in !-tubulin sequence lengths.   The aligned dataset consisted of 837 nucleotides from nSSU, 592 nucleotides from nLSU and 1258 nucleotides from !-tubulin loci. We excluded no nucleotides from nLSU and nSSU loci. However, 735 intron positions were excluded from the !-tubulin locus  45 because the large differences in length and composition of intron sequences across the ophiostomatoid orders and genera made the regions unalignable (Swofford et al. 1996).  We submitted the sequences to BLAST to assess potential misidentifications. The comparisons confirmed the species identity of all Ambrosiella and Raffaelea fungi in the analysis expect for A. ips and A. sulcati. Ambrosiella ips showed a high level of sequence identity (rDNA + !-tubulin: 99.7%) with that of O. montium, suggesting a potential misidentification and the possibility that A. ips and O. montium might represent a single species. Ambrosiella sulcati showed high sequence identity with that of R. canadensis (rDNA + !-tubulin: 99.6%), indicating that these two taxa may represent a single species. Since the closest match of R. castellanii was a Dothideomycetes, which is unrelated to ophiostomatoid fungi, this species was not included in the final phylogenetic analysis.  2.3.2 Phylogenetic analysis  MPTs from conserved individual loci (nSSU, nLSU and !-tubulin exons) showed weak resolution for the topology of deeper nodes and terminal branches. Although the partition homogeneity test (P-value< 0.01) did not indicate that the rDNA and !-tubulin datasets were concordant, MPT topologies from individual rDNA loci was not in conflict to the combined nSSU+nLSU+!-tubulin dataset, which had better resolution and higher support values. The concatenated matrix (nSSU, nLSU and !-tubulin) included sixty- 46 seven taxa from different ophiostomatoid genera (figure 2.1, Table 2.1) and 1952 aligned sites, of which 719 sites were variable and 534 sites were parsimony informative.  Under the first gap treatment (i.e., gaps as missing data), the parsimony analysis of the concatenated dataset resulted in nine MPTs with a length of 2703 steps (CI=0.39, RI=0.71). Gaps as a fifth character state resulted in eleven MPTs with the same length and topologies; thus, for the remainder of the analysis gaps were treated as missing data. The consensus phylogeny inferred from the Bayesian analysis revealed similar topology within and between groupings but with higher supporting relationships than those from the MP analysis (figure 2.1). The most visible difference was the placement of the genus Ceratocystiopsis, which appeared either as part of the Ophiostoma clade or as a basal group to the Ophiostoma and Grosmannia clades. However, neither method supported the placement of this genus strongly.  Analysis divided the ingroup taxa into four major clades, each receiving 79% or more bootstrap support and 100% posterior probabilities (figure 2.1A?D). These clades corresponded to the three teleomorph genera Ophiostoma, Grosmannia, and Ceratocystiopsis, recently re-instated by Zipfel et al. (2006) in the Ophiostomatales, as well as to the genus Ceratocystis of the Microascales. The twenty-five Ambrosiella, Dryadomyces and Raffaelea species in the analysis were divided into at least six well-resolved (>51% BS; 100% PP) groups which nested within the clades of Ophiostoma (A), Grosmannia (B) and Ceratocystis (D) (figure 2.1).  47  Within clade A, Ambrosiella ips grouped strongly (100% BS) with O. montium in the Ophiostoma ips complex (group 1) (Zhou et al. 2004). Other Ambrosiella taxa isolated from scolytid bark beetles formed a single, monophyletic group (group 2) with 99% BS and 100% PP (figure 2.1). This group contained A. tingens, A. macrospora and three undescribed Ambrosiella taxa, and was well separated from other Ambrosiella species that have been isolated from ambrosia beetles. The bark-beetle associated group showed a sister relationship to a group containing members of the O. piceae complex (Harrington et al. 2001) and both groups are also sibling of the O. ips complex (figure 2.1A).  Four Ambrosiella species clustered with the representatives of Raffaelea and Dryadomyces in clade B, which also included the genus Grosmannia. All ?ambrosia? fungi in this clade had a close association with platypodid and scolytid ambrosia beetles but their monopoly received a poor bootstrap support (62%) and a low posterior probability of 81%. Instead they were subdivided into two distinct well-resolved groups (groups 3, 4) that were supported (79% BS; 100% PP) as sibling of the genus Grosmannia. Group 3, which received strong supports (78% BS; 100% PP), encompassed all Raffaelea taxa, except R. lauricola and R. montetyi. The group included: R. albimanens, R. ambrosiae, R. arxii, R. canadensis, R. santoroi, R. sulcati, R. tritirachium, as well as two Ambrosiella species. Ambrosiella sulcati clustered with R. canadensis, with strong support (100% BS). While A. gnathotrichi was closely related to  48 R. arxii. These two taxa formed a monophyletic relationship with the R. canadensis-clade without bootstrap support and a posterior probability of 95%.   Representatives of Raffaelea concentrated in clade B, but the genus is not monophyletic. However, the relationships among various taxa (R. albimanens, R. santoroi, R. sulcati, R. tritirachium) were well resolved. The only exception to this was the unstable positioning of R. ambrosiae, the type species of Raffaelea, which depending on the locus tested, formed a monophyletic relationship with the R. arxii and R. canadensis group, or a basal taxon in the group 3.  Group 4, which was supported with high posterior probability (100%), contained representatives from Ambrosiella, Raffaelea and Dryadomyces but none of these constituted a monophyletic cluster. Ambrosiella brunnea and A. sulphurea mixed with D. amasea and with R. lauricola and R. montetyi. Group 4 also included one unidentified species (TR25) isolated from the mycangia of ambrosia beetle T. rufitarsus from lodgepole pines infested by the MPB in BC (Table 2.2). This fungus appeared as sister of A. brunnea and both species formed a well-supported (82% BS; 100% PP) monophyletic clade with the recently described species R. lauricola (Harrington et al. 2008).  Three species of Ambrosiella belonged to the Microascales (clade D) but they did not form a single, monophyletic group (figure 2.1). Instead they formed two distinct  49 groups (groups 5, 6) that form a close association with scolytid ambrosia beetles but interspersed with Ceratocystis fungi bearing a loose relationship with different insects. First group included the type species A. xylebori as well as A. hartigii (group 5) while the other included two strains of A. ferruginea (group 6).  2.3.3 Morphological investigation  Table 2.2 summarised morphological features that have been used in the literature to describe the genera Ambrosiella, Raffaelea and Dryadomyces: conidiomatal types (hyphal/singly, sporodochial, synnematous) and conidial proliferation (annellidic, phialidic, and sympodial).  Representative strains of the Ophiostoma-related Ambrosiella tended to sporulate better on the PDA and YEMEA media therefore morphological observations were made on these cultures. Similar to other Ambrosiella, the undescribed Ambrosiella spp. (Ambrosiella sp. 1 and Ambrosiella sp. 2) isolated from bark beetles in Canada (Massoumi Alamouti et al. 2007) and those from European bark beetles (Ambrosiella sp. 3) (Krokene and Solheim 1996) produced simple, mononematous conidiophores (figures 2.2B, 2.3); these were arranged in a discrete sporodochium-like structure (figure 2.2). Solitary conidiophores were also observed in the younger cultures of these species. Observations by light microscopy of these three species and the A. tingens type culture (CBS 366.53) revealed a non-phialidic conidiogenesis (e.g. figure 2.3D). In  50 contrast, SEM observations revealed both annellidic (figures 2.2B, 2.3A-B) and sympodial (figure 2.3C) conidiogenesis in Ambrosiella sp. 1 as well as in Ambrosiella sp. 2; however, conidial development seemed to occur more frequently through annellidic percurrent proliferation than sympodial proliferation. Because A. macrospora, A. tingens and the species from Europe produced few spores on artificial media and wood blocks, we were unable to determine with certainty whether their conidiogenesis was sympodial or annellidic. Ambrosiella ips (CBS 435.34) also failed to produce any fruiting structures on media and therefore its morphological characters could not be compared to those of the genetically identical species O. montium (CBS 151.78).  2.4 Discussion  We established a comprehensive phylogeny that clarifies the relationships between most filamentous ?ambrosia? fungi isolated from platypodid and scolytid beetles and their relationships with the ophiostomatoid fungi. Our results are consistent with studies that described the polyphyletic status of the genera Ambrosiella and Raffaelea (Cassar and Blackwell 1996; Farrell et al. 2001; Rollins et al. 2001; Gebhardt et al. 2005). These earlier phylogenies used mainly nSSU rDNA sequences to characterise members of Ambrosiella and/or Raffaelea at higher taxonomic levels, and sets of species that that did not adequately represent the morphological and ecological diversity of ophiostomatoid fungi. In the work described here, we addressed both limitations. We generated a new multigene dataset, and we characterised a diverse set of fungi that included representatives from the genera Ambrosiella, Raffaelea and Dryadomyces, as  51 well as from taxa that we selected from currently accepted ophiostomatoid teleomorph and anamorph genera. Our analysis indicated that these fungi evolved from three major teleomorph groups in two ascomycete orders: Ophiostomatales and Microascales. We will not discuss the Ambrosiella species within Microascales that include the type species A. xylebori because this aspect has been thoroughly studied by Paulin-Mahady et al. (2002) and their results agree with ours. In the subsequent discussion we will focus on the Ambrosiella species that belong to the Ophiostomatales.  von Arx and Hennebert (1965) introduced the genus Ambrosiella to describe A. xylebori, the most frequent associate of the ambrosia beetle Xylosandrus compactus. Following the original description of the type species, nine additional species were assigned to the genus (Batra 1967). These species share certain morphological features: simple conidiophores, sporodochial arrangements of conidiophores and single terminal conidia (Batra 1967). Cassar and Blackwell (1996) showed that the genus Ambrosiella was not monophyletic within Ophiostomatales based on nSSU rDNA. Their phylogenies recognised two possible Ambrosiella groups that were closely related to either Leptographium-forming species or Ophiostoma species characterised by their Pesotum (e.g., O. piceae) and/or Hyalorhinocladiella (e.g., O. bicolour) anamorphs. Results from Farrell et al. (2001), Rollins et al. (2001) and Gebhardt et al. (2005) supported these groupings, but resolved neither the monophyly of different Ambrosiella groups nor their relationships with the closely related genera Ophiostoma and Raffaelea. Our multigene phylogeny clarified the relationships among these fungi and recognised at least four groups of Ambrosiella within the Ophiostomatales.  52  2.4.1 Ambrosiella associates of bark beetles are related to the teleomorph genus Ophiostoma  Our results supported a novel clade that consisted of five bark beetle associates: A. macrospora, A. tingens and the three undescribed Ambrosiella species from Canadian and European bark beetles (Krokene and Solheim 1996; Massoumi Alamouti et al. 2007). Ambrosiella macrospora and A. tingens were originally described in the genus Trichosporium Nannfeldt as T. tingens var. macrosporum and T. tingens (Lagerberg et al. 1927; Francke-Grosmann 1952). Batra (1967) reclassified these two bark beetle associates into Ambrosiella, while indicating that this genus should describe fungal associates of platypodid and scolytid ambrosia beetles. nSSU rDNA phylogenies subsequently showed that these two species were more closely related to Ophiostoma than to Ambrosiella (Cassar and Blackwell 1996; Rollins et al. 2001); however, limitations from nSSU rDNA sequences and taxon sampling prevented these studies from characterising the phylogenetic relationships in detail. In contrast, our multigene analysis resolved the Ambrosiella associates of bark beetles as a distinct group within Ophiostoma. We also showed that the bark-beetle associated group is a sibling to the O. piceae complex, members of which commonly inhabit sapwood and bark beetle tunnels in temperate forests (Harrington et al. 2001). Members of this complex are distinguished by their teleomorph fruiting bodies and Pesotum anamorph, which include both synnematous and Sporothrix conidiophore arrangements (Harrington et al. 2001, Seifert et al. 1993). Ambrosiella species produce conidiophores without denticles (Batra  53 1967; Gebhardt et al. 2005), which differentiate them from Sporothrix. Ambrosiella associates of bark beetles, including the undescribed species, do not produce synnemata; instead, their conidiophores are arranged in distinct sporodochial-like structures that resemble those of other Ambrosiella and Raffaelea species. Our analysis also strongly suggested that the bark beetle-associated Ambrosiella and the O. piceae complex form a monophyletic clade with members of the O. ips complex, which are distinguished by their pillow-shaped ascospores and continuum of anamorphs including Hyalorhinocladiella and Pesotum (Zipfel et al. 2006). This monophyletic clade is a sister to another Ophiostoma group that include a number of species (e.g., O. abietinum and O. stenoceras) foundo in a diverse range of ecological niches and identified by their Sporothrix anamorph.   Our analysis grouped A. ips with O. montium within the O. ips complex. Originally A. ips was described in the Tuberculariella von H?hnel (Leach et al. 1934), and then was transferred into the genus Ambrosiella (Batra 1967). However, our molecular results, as well as morphological and ecological descriptions from the literature suggested that these two species might represent a single taxon. We showed that the two species shared a high level of sequence identity (99.7%). We were unable to compare the morphology of A. ips and O. montium because, in our hands, the only A. ips strain available from the CBS culture collection and reported in the literature did not sporulate; however, the sporodochium-like structures illustrated in the description of A. ips (Leach et al. 1934) are similar to Graphilbum reported for O. montium and O. ips (Upadhyay 1981; Hutchison and Reid 1988). Because both A. ips and O. montium have  54 been isolated from bark beetles (I. pini and MPB, respectively) that infest the same pine-host trees in North America (Leach et al. 1934; Lee et al. 2006), it is possible that galleries of these beetles have overlapped, resulting in fungal associates being mixed and A. ips being misidentified.   2.4.2 Ambrosiella associates of ambrosia beetles are related to the teleomorph genus Grosmannia  In our analysis, the remaining Ophiostomatales Ambrosiella grouped with members of genera Raffaelea and Dryadomyces and formed a sister relationship with members of the genus Grosmannia. All members of this group (Raffaelea, Dryadomyces and Ambrosiella) are closely associated with the platypodid and scolytid ambrosia beetles (von Arx and Hennebert 1965; Guerrero 1966; Batra 1967; Funk 1970; Scott and Du Toit 1970). Our multigene phylogenies suggested that these ambrosia-beetle associates are monophyletic but with weak statistical support. Note that while we included most species described from ambrosia beetles, relatively few such associates have been fully characterised, and we were unable to further test the monophyly with sequence data for more isolates. The placement of ambrosia-beetle associates within the Ophiostomatales has been problematic because earlier phylogenetic studies consistently grouped them with species like O. piceaperdum and O. serpens of the well-defined genus Grosmannia (Rollins et al. 2001; Gebhardt et al. 2005; Hulcr et al. 2007). Zipfel et al. (2006) re-instated this genus to accommodate the most common fungal associates of bark beetles that are distinguished by their Leptographium anamorph (Jacobs and Wingfield  55 2001); however, when they tested the monophyly of Grosmannia they did not consider the close relatives of this genus, namely, members of genera Ambrosiella and Raffaelea. While our results confirmed that Grosmannia is monophyletic, we also included the Grosmannia-related associates of ambrosia beetles. Our analysis placed these into two distinct groups and provided the first robust indication that the ambrosia-beetle associates are close but independent relatives of Leptographium-forming species commonly isolated from bark beetles.  The first of the two groups included seven of the nine tested species of genus Raffaelea, as well as A. sulcati and A. gnathotrichi. The exceptions, R. lauricola and R. montetyi, were placed in the second group. The Raffaelea genus was established by von Arx and Hennebert (1965) to describe R. ambrosiae, which is frequently associated with the pinhole borer Platypus cylindricus in North America and Europe. Because the group members have similar ecological and morphological features and formed a strongly supported monophyletic clade, we suggested that they should be all recognised as species of genus Raffaelea s. str. (von Arx and Hennebert 1965; Guerrero 1966; Batra 1967; Funk 1970; Scott and du Toit 1970). We summarised evidence for this as follows. Ambrosiella sulcati and R. canadensis were respectively isolated from the ambrosia beetles Gnathotrichus retusus and Platypus wilsonii when these two beetles inhabited the same host, Pseudotsuga menziesii (Douglas fir) (Batra 1967; Funk 1970). Differentiating Raffaelea and Ambrosiella morphologically by assessing whether cicatricial conidial scars are present or absent using light microscopy is difficult, and often depends on subtle interpretations by researchers (von Arx and Hennebert 1965;  56 Batra 1967; Funk 1970). However, our results showed that these two species have high rDNA and !-tubulin sequence identify (99.6%) and formed a conspecific group within Raffaelea species, and so indicated that R. canadensis and A. sulcati represent a single taxon and that A. sulcati should be transferred into Raffaelea. Ambrosiella gnathotrichi is a frequent associate of the conifer-infesting species G. retusus in North America (Batra 1967). While R. arxii forms a close association with the Xyleborus torquatus on the Cussonia umbellif in South Africa (Scott and du Toit 1970). Gebhardt et al. (2005) suggested that A. gnathotrichi?s conidial ontogeny differs from that of the Ambrosiella type species and they showed that A. gnathotrichi form a sister taxon relationship with R. arxii. Our multigene analysis provided a higher resolution for the species-level phylogeny and showed that these two fungi are closely related species. Although, our analyses suggest that A. gnathotrichi be assigned to the genus Raffaelea, a more thorough morphological examination of the type material is needed, particularly of conidial ontogeny.   The second group included the remaining species from all three genera: Ambrosiella, Raffaelea and Dryadomyces. The group members were A. brunnea, A. sulphurea, R. lauricola, R. montetyi and D. amasae, and an unidentified species isolated from T. rufitarsus colonising MPB-attacked lodgepole pine in BC (Kuhnholz 2004). Our phylogenetic analysis consistently resolved these species as sister of other ambrosia-beetle associates in the Raffaelea clade, with weak statistical support. We will clarify briefly the group members. Ambrosiella brunnea and A. sulphurea have been isolated from North American and European hardwood species (Quercus and Acer) that  57 were infested with ambrosia beetles in the genus Monarthrum and with Xyleborus saxesenii, respectively (Verrall 1943; Batra 1967). Gebhardt et al. (2005) showed that these two Ambrosiella species produce a non-phialidic conidiogenesis, and included them in their Raffaelea phylogenetic clade. In contrast, our analysis segregated them from both Ambrosiella and Raffaelea genera. Our analysis showed that the Ambrosiella associate of Monarthrum spp. (A. brunnea), and the unidentified species isolated from T. rufitarsus in BC are sister taxa that form a distinct, well-supported monophyletic group with the new vascular wilt pathogen R. lauricola (Fraedrich et al. 2008). This pathogen is associated with the exotic ambrosia beetle Xyleborus glabratus, and causes substantial mortality of redbay and other Lauraceae in the USA (Fraedrich et al. 2008; Harrington et al. 2008). nSSU and nLSU sequences from Fraedrich et al. (2008) respectively suggested that the pathogen is related to A. brunnea and Leptographium spp. Our phylogenetic results were consistent with this pathogen being a distinct species; and included it with other ambrosia-beetle associates of genera Ambrosiella, Raffaelea and Dryadomyces that formed a sister group relationship with Leptographium -forming species. Harrington et al. (2008) suggested that this fungus most appropriately fit the genus Raffaelea; however, they did not provide a detailed morphology of the conidiogenous cells. Further, our analyses placed this pathogen in a clade that was clearly separated from other ambrosia-beetle associates in the genus Raffaelea. Therefore, the taxonomic status of the vascular wilt pathogen in the USA is not clear and needs further study. Our multigene phylogenies placed A. sulphurea into a distinct subclade that formed a highly supported sister relationship with R. montetyi. Previous nSSU phylogenies had also suggested that these two species were closely related  58 (Gebhardt et al. 2004, 2005; Hulcr et al. 2007). Both A. sulphurea and R. montetyi are frequently associated with ambrosia beetles in the genus Xyleborus that inhabit the oak trees (Gebhardt et al. 2004). Although, R. montetyi has been shown to produce conidia by annellidic percurrent proliferation, resolving the taxonomic placement of R. montetyi and that of A. sulphurea will require additional morphological studies. Finally, the second group comprises a monotypic genus, Dryadomyces, which accommodates the single species D. amasae. The genus was introduced by Gebhardt et al. (2005) to describe fungi frequently isolated from ambrosia beetle Amasa concitatus infesting hardwood timbers in Taiwan. nSSU results indicated that these fungi were phylogenetically related to species of genera Raffaelea and Ambrosiella but they were included into the new genus Dryadomyces based on their unique conidiogenous cells (Gebhardt et al. 2005). Later, Harrington et al. (2008) amended the genus indicating that until the taxonomy of the genus Ophiostoma was better resolved, Raffaelea should include all ambrosia beetle symbionts with affinities to Ophiostoma. Our multigene analysis provided a better resolution for the phylogenetic status of the genus Ophiostoma and consistent with the morphological observations of Gebhardt et al. (2005), recognised D. amasae as a distinct monotypic lineage that formed a highly supported monophyletic relationship with R. montetyi and A. sulphurea.  2.4.3 Morphological features  Morphological characters used to define Ambrosiella are less informative in phylogeny because the genus is polyphyletic. The morpho-taxonomy of Ambrosiella and  59 Raffaelea has been difficult and unstable. The shape (reduced conidiophores) and arrangement of conidiophores (sporodochia), as well as the mode of conidiogenesis are the key morphological characteristics that traditionally used to differentiate the genus Ambrosiella from the closely related genus Raffaelea (von Arx and Hennebert 1965; Batra 1967). However, complex or simple conidiophores did not correlate with the generic or sub-generic classification. Also the mode of conidiogenesis is difficult to observe under light microscopy (Tsuneda and Currah 2006). Ambrosiella and Raffaelea genera were reported as sympodial (von Arx and Hennebert 1965). But, Gebhardt et al. (2005) showed the presence of phialidic conidia for A. xylebori, and in other two Ceratocystis-related species: A. hartigii and A. ferruginea. Recently Raffaelea species having annellidic conidiogenesis and sympodial proliferations in D. amasae were also illustrated (Gebhardt et al. 2004; Gebhardt and Oberwinkler 2005).  Our phylogenetic results also indicated that the characters used to define anamorphs are convergent within the Ophiostomatales. The conidial proliferation in the unidentified Ambrosiella species, as well as A. tingens was non-phialidic, which clearly distinguished them from the type species of Ambrosiella. In addition, we observed that the conidiophores of these strains lacked denticles; this differentiated them from anamorph genera Sporothrix and Dryadomyces, both of which have conidia formed sympodially on denticles arising from undifferentiated hyphae (Gebhardt et al. 2005; Harrington et al. 2001). In our SEM micrographs, we found that the non-phialidic conidiogenesis observed for Ambrosiella sp. 1 (figure 2.3A) and Ambrosiella sp. 2 (figure 2.3B) was occurring through annellidic proliferation, and consequently was  60 identical to that found for Ophiostomatales anamorph genera Raffaelea, Hyalorhinocladiella, Leptographium and Pesotum (Seifert and Okada 1993; Benade et al. 1995; Tsuneda and Currah 2006). These two undescribed Ambrosiella also formed apical sympodial conidiogenesis but less frequently. Although, additional work is necessary to define the true mode of conidiogenesis for all Ambrosiella species related to the Ophiostoma, it is important to note that the conidium ontogeny, an early important taxonomic character, is now being challenged, because conidial fungi often develop more than one pattern of conidiogenesis and can be assigned to different anamorphic genera (Tsuneda and Currah 2006).  While Ambrosiella sp. 1 and Ambrosiella sp. 2 with their annellidic conidiogenesis and the morphology of their conidiophores are most similar to species of the genus Raffaelea, they were clearly distinguished from the Raffaelea group by our multigene phylogeny. Consistent with our phylogenetic classification, members of Raffaelea clade also colonise different ecological niches. The morphological characteristics of Ambrosiella sp. 1 and Ambrosiella sp. 2 also resemble those of Hyalorhinocladiella anamorph. This anamorphic state is not clearly delimited to a genus and is present in anamorphs of Ceratocystiopsis and Ophiostoma (e.g., O. ips complex) (Upadhyay and Kendrick 1974; Benade et al. 1996). Currently, species of genera Ambrosiella and Raffaelea are differentiated from other ophiostomatoid genera including Hyalorhinocladiella by the formation of sporodochia. However, the production of sporodochia is variable and is often associated with the growth of the fungus in its  61 natural habitat (beetle gallery); as well, the importance of this structure for segregating anamorphic genera within the ophiostomatoid fungi has not been clarified.   In conclusion, we clarified the phylogenetic classification of Ambrosiella species isolated from ambrosia and bark beetles and that of the Raffaelea and Dryadomyces associates of ambrosia beetles, as well as the relationships between these species and ophiostomatoid relatives. We found that species of genus Ambrosiella are distributed in a number of distinct phylogenetic groups that each might be reassigned to different genera, and that the genus Raffaelea should be revised. While no morphological characteristics unambiguously supported the monophyletic groups that we report from our molecular data for the genera Ambrosiella and Raffaelea, these groups are clearly associated with the feeding behaviour of their beetle vectors. Specifically, Ambrosiella associates of scolytid bark beetles formed a monophyletic group in the genus Ophiostoma, while species associated with scolytid and platypodid ambrosia beetles form separate lineages that have a monophyletic relationship with the genus Grosmannia. Generating additional support for the monophyly presented will require characterising a range of morphological characters and/or ecological traits on an expanded collection of freshly isolated fungi from ambrosia and bark beetles.   62 2.5 Tables and figures  Table 2.1 Fungal species used in this study  GenBank accession no. b Species Source a nSSU rDNA  nLSU rDNA  !-tubulin  Ambrosiella sp. 1 UAMH10632 EU984247  (DQ268582)  (DQ268618)  UAMH10633 EU984248  (DQ268583)  (DQ268619) Ambrosiella sp. 2 UAMH10634 EU984249  (DQ268584)  (DQ268620)  UAMH10635 EU984250  (DQ268585)  (DQ268621) Ambrosiella sp. 3 NISK-1994-166-39A EU984252  EU984282  EU977458  NISK-1994-176-B4 EU984253  EU984283  EU977459 A. brunnea Batra CBS 378.68 (AY858654)  EU984284  EU977460 A. ferruginea (Mathiesen-K??rik) Batra CBS 408.68 EU984254  EU984285  EU977461  JB13 CB EU984255  EU984286  EU977462 A. gnathotrichi Batra  CBS 379.68 (AY858655)  EU984287  N/A A. hartigii Batra CBS 404.82 EU984256  EU984288  EU977463 A. ips (Leach, Orr and Christensen) Batra CBS 435.34 AY858657  EU984289  EU977464 A. macrospora (Francke-Grosmann) Batra CBS 367.53 EU984257  EU984290  EU977465 A. sulcati Funk CBS 805.70 (AY858658)  EU984291  EU977466 A. sulphurea Batra CBS 380.68 (AY497509)  EU984292  EU977467 A. tingens (Lagerberg and Melin) Batra CBS 366.53 EU984258  EU984293  EU977468 A. xylebori Brader ex von Arx and Hennebert CBS 110.61 (AY858659)  EU984294  EU977469 Dryadomyces amasae Gebhardt CBS 116694 (AY858661)  EU984295  EU977470 Raffaelea albimanens Scott and du Toit CBS 271.70 EU984259  EU984296  EU977471 R. ambrosiae von Arx and Hennebert CBS 185.64 (AY497518)  EU984297  EU977472 R. arxii Scott and Toit CBS 273.70 (AY497519)  EU984298  N/A R. canadensis Batra CBS 168.66 (AY858665)  EU984299  EU977473   63 GenBank accession no. b Species Source a nSSU rDNA  nLSU rDNA  !-tubulin  R. castellanii (Pinoy) de Hoog MUCL 15755 EU984260  EU984300  EU977474 R. lauricola Harrington, Fraedrich and Aghayeva   (EU123076)  (EU123077)  N/A R. montetyi Morelet CBS 451.94 (AY497520)  EU984301  EU977475 R. santoroi Guerrero CBS 399.67 EU984261  EU984302  EU977476 R. sulcati Funk CBS 806.70 (AY858666)  N/A  EU977477 R. tritirachium Batra CBS 726.69 EU984262  EU984303  EU977478 Unidentified species TR25 CB EU984251  EU984281  EU977457 Ceratocystiopsis manitobensis (Reid and Hausner) Zipfel, Beer and Wingfield UM 237 EU984266  (DQ268607)  (DQ268638) Cop. minuta (Siemaszko) Upadhyay and Kendrick CBS 463.77 EU984267  (DQ268615)  EU977481 Cop. minuta-bicolor (Davidson) Upadhyay CBS 635.66 EU984268  (DQ268616)  EU977482 Ceratocystis adiposa (Butler) Moreau, CBS 600.74 EU984263  EU984304  EU977479 C. coerulescens (M?nch) Bakshi  CL 13-12 CB EU984264  (AY214000)  (AY140945) C. moniliformis (Hedgc.) Moreau CBS 155.62 EU984265  EU984305  EU977480 Grosmannia abiocarpa (Davidson) Zipfel, Beer and Wingfield MUCL 18351 EU984269  (AJ538339)  (DQ097857) G. clavigera (Rob.-Jeffr. and Davidson) Zipfel, Beer and Wingfield ATCC 18086 EU984270  (AY544613)  (AY263194) G. cucullata (Solheim) Zipfel, Beer and Wingfield CBS 218.83 (AY497513)  (AJ538335)  EU977483 G. penicillata (Grosmann) Goid. Zipfel, Beer and Wingfield  (AY858662)  (DQ097851)  (DQ097861) G. piceaperda (Rumbold) Goid. Zipfel, Beer and Wingfield  (AY497514)  (AY707209)  (AY707195) G. serpens (Goid.) Zipfel, Beer and Wingfield  (AY497516)  (DQ294394)  (AY707188) Leptographium abietinum (Peck) Wingfield DAOM 60343 EU984271  (DQ097852)  (AY263182) L. fruticetum M. Alamouti, Kim and Breuil  DAOM 234390 EU984272  (DQ097848)  (DQ097855) L. longiclavatum Lee, Kim and Breuil  DAOM 23419 EU984273  (AY816686)  (AY288934) L. lundbergii Lagerberg and Melin UAMH 9584 EU984274  (AY544603)  (AY263184) L. terebrantis Barras and Perry UAMH 9722 EU984275  (AY544606)  (AY263192) Ophiostoma abietinum Marm. and Butin CMW 1468 EU984276  (DQ294356)  EU977484 O. bicolor Davidson and Wells  (AY497512)  (DQ268604)  (DQ268635) O. canum (M?nch) Sydow and P. Sydow AU 30 CB EU984277  (AJ538342)  EU977485 O. floccosum Mathiesen-K??rik  (AF139810)  (AJ538343)  (AY789142) O. ips (Rumbold) Nannfeldt  (AY172021)  (AY172022)  (AY789146)   64 GenBank accession no. b Species Source a nSSU rDNA  nLSU rDNA  !-tubulin  O. montium (Rumbold) Hunt CBS 151.78 EU984278  (AY194947)  (AY194963) O. novo-ulmi Brasier NAN-MH75 CB N/A  (DQ294375)  EU977486 O. piceae (M?nch) Sydow and P. Sydow  (AB007663)  (AJ538341)  (AY305698) O. pulvinisporum Zhou and Wingfield CMW 9020 N/A  (DQ294380)  EU977487 O. setosum Uzunovic, Seifert, Kim and Breuil  N/A  (AF128929)  (AY305703) O. stenoceras (Robak) Nannfeldt C80 (M85054)  (DQ294350)  EU977488 O. quercus (Georgev.) Nannfeldt  (AF234835)  (DQ294376)  (AY789157) O. ulmi (Buisman) Nannfeldt W9 CB (M83261)  (DQ368627)  EU977489 Claviceps sp.  (U32401)  (U17402)  (AF263569) Daldinia sp.  (U32402)  (U47828)  (AY951701) Epichloe typhina (Pers.) Tul. and C. Tul.  (AB105953)  (U17396)  (X52616) Microascus cirrosus Curzi  CBS 217.31 EU984279  (AF275539)  EU977490 Penicillium expansum Link  (DQ912698)  (AF003359)  (AY674400) Petriella setifera (Schmidt) Curzi CBS 385.87 EU984280  (DQ470969)  EU977491 Taphrina populina (Fr.) Fr.   (D14165)  (AF492053)  (AF170968) Xylaria sp.   (U32417)  (AY327481)  (AY951763) a Source of isolates sequenced in this study: ATCC, American Type Culture Collection, Manassas, USA; C, Iowa State University, Dept. of Plant Pathology, USA; CBS, Centraalbureau voor Schimmelcultures, Utrecht, the Netherlands; CMW, Culture Collection Forestry and Agricultural Biotechnology Institute (FABI), University of Pretoria, South Africa; DAOM, Canadian Collection National Fungus Herbarium and Culture Collection, Ottawa, Canada; NISK, Norwegian Forest Research Institute, Austria; MUCL, (Agro) Industrial Fungi and Yeasts Collection, Belgium; UAMH, University of Alberta Microfungus Collection and Herbarium, Edmonton, Canada; UM, University of Manitoba, Dept. of Botany, Winnipeg, Canada; CB, Colette Breuil?s Culture Collection, University of British Columbia, Canada.  b Accession numbers of sequences newly produced, updated (bold) or downloaded from GenBank (parentheses); N/A, not available.    65 Table 2.2 Morphological and ecological characters reported for the genera Ambrosiella, Raffaelea and Dryadomyces in the literature or in this study  Morphological characteristic Ecological characteristic Fungal species Conidiophore arrangement  Conidiogenesis Bark beetle ex host tree (geographic origin) Reference Ambrosiella sp. 1  (UAMH10632-10633) Distinct sporodochia-like and solitary a  Annellidic & sympodial a  Ips spp., Dryocoetes affaber, Polygraphus rufipennis ex Picea spp. (Canada) Beaulieu e, Harrison e, Massoumi Alamouti et al. 2007 Ambrosiella sp. 2 (UAMH10634-10635) Distinct sporodochia-like and solitary a  Annellidic & sympodial a  Ips spp., Dryocoetes affaber, Polygraphus rufipennis ex Picea spp. (western Canada) Massoumi Alamouti et al. 2007 Ambrosiella sp. 3 (NISK-94-166/39A, 1994-176-B4)  Distinct sporodochia-like and solitary (data not shown) a  Non-phialidic (data not shown) a Hylurgops palliatus, Polygraphus poligraphus ex Picea abies (Norway) Krokene & Solheim 1996  A. brunnea Distinct sporodochia  Sympodial b Monarthrum spp. ex Acer spp., Quercus spp. (western Canada, USA) Batra (1967) d, Funk 1965, Verrall 1943 A. ferruginea Effused sporodochia  Phialidic Trypodendron spp., Xyloterus signatus ex Betula sp., Fagus sylvatica, Larix spp., Picea spp., Pinus spp., Populus sp., Quercus sp. (Canada, Europe, USA) Batra (1967), Gebhardt et al. (2005), Kuhnholz 2004, Mathiesen-K??rik 1953 [A. gnathotrichi] f Indistinct sporodochia (fascicles in younger part of colonies)  Non-phialidic Gnathotrichus retusus ex Picea engelmannii, Pinus ponderosa (Colorado) Batra (1967), Gebhardt et al. (2005) A. hartigii Distinct sporodochia   Phialidic Xyleborus dispar, Xylosandrus germanus ex Malus sylvestris  (Asia, Europe, USA) Batra (1967), Gebhardt et al. (2005), Hartig 1844, Kajimura & Hijii 1992  [A. ips] Solitary (sometimes indistinct sporodochia-like in old cultures)  Sympodial b Ips spp., I. sexdentatus ex Pinus spp. (western USA, Europe)   Batra (1967), Leach et al. (1934) A. macrospora Effused sporodochia and solitary  Sympodial b Ips acuminatus ex Pinus sylvestris, Pinus spp. (Europe) Batra (1967), Francke-Grosmann 1952 [A. sulcati] Distinct sporodochia  Sympodial b Gnathotrichus retusus ex Pseudotsuga menziesii (Canada) Funk (1970) A. sulphurea Distinct sporodochia   Non-phialidic Xyleborinus saxesenii ex Populus spp., Quercus spp. (USA, Germany) Batra (1967), Gebhardt et al. (2005)   66 Morphological characteristic Ecological characteristic Fungal species Conidiophore arrangement  Conidiogenesis Bark beetle ex host tree (geographic origin) Reference A. tingens Solitary or distinct sporodochia  Non-phialidic a Tomicus minor, T. piniperda, Ips sexdentatus, ex Pinus spp. (Europe) Batra (1967), Francke-Grosmann 1952, Mathiesen-K??rik 1953  A. xylebori Confluent sporodochia (fascicle in younger part of colonies)  Phialidic Xylosandrus compactus, X. crassiusculus, Corthylus  columbianus, ex Coffea canephora, Acer rubrum and Ulmus sp. (Africa, Ceylon, India, eastern USA, Taiwan)  Batra (1967), Brader 1964, Gebhardt et al. (2005), von Arx & Hennebert (1965) Dryadomyces amasae Confluent sporodochia  Sympodial Amasa spp. (Taiwan ) Gebhardt et al. (2005) Raffaelea albimanens Distinct sporodochia, solitary and indeterminate synnemata  Annellidic  Platypus externedentatus ex Ficus sycamorus (South Africa) Gebhardt & Oberwinkler (2005), Scott & du Toit (1970),  R. ambrosiae Distinct sporodochia, solitary or loose fascicles  Annellidic Platypus spp. (i.e. P. wilsonii, P. cylindrus) ex Quercus spp. (British Columbia, England, USA) Batra (1967), Gebhardt & Oberwinkler (2005), von Arx & Hennebert (1965) R. arxii Confluent sporodochia, solitary and indeterminate synnemata  Annellidic Xyleborus torquatus ex Cussonia umbellif (South Africa) Gebhardt & Oberwinkler (2005), Scott & du Toit (1970)  R. canadensis Solitary and sporodochia  Sympodial b Platypus wilsonii ex Pseudotsuga menziesii (British Columbia, Oregon)  Batra (1967)  R. lauricola  Solitary and sporodochia  Sympodial b Xyleborus glabratus ex Persea borbonia, other members of Lauraceae (USA) Fraedrich et al. 2008, Harrington et al. (2008) R. montetyi Solitary or fascicles in the beetle galleries  Annellidic Platypus cylindrus, Xyleborus monographus, X. dryographusex Quercus spp. (Europe, Portugal) Gebhardt et al. (2004), Morelet (1998) R. santori N/A c  Sympodial b Platypus spp. (i.e. P. sulcatus) ex N/A (Argentine) Guerrero (1996)  R. sulcati Effused sporodochia (dense fascicles)  Sympodial b Gnathotrichus sulcatus ex Pseudotsuga menziesii (Canada) Funk (1970)  R. quercivora Distinct sporodochia or solitary  Sympodial b Platypus quercivorus ex Quercus spp. (Japan)  Kubono & Ito (2002) R. tritirachium Confluent sporodochia (fascicles)  Sympodial b Monarthrum mali ex Quercus spp. (Pennsylvania)  Batra (1967) Unidentified species (TR25)  N/A  N/A Trypodendron rufitarsus ex Pinus contorta (western Canada) Kuhnholz 2004 a Reported in this paper .    67 b Sympodial proliferations have been reported by earlier studies through light microscopy, and the refore the true mode of conidiogenesis in these fungi needs further investigation.  c N/A, no description available.  d References in which the publication year is being shown in parentheses are dealing with the conidiomatal types or conidiogenesis.  e Unpublished data from spruce-beetle-fungal survey in eastern Canada by Dr. Ken Harrison (Natural Resources Canada, Canadian Forest Service-Atlantic) and Marie-Eve Beaulieu in Centre d??tude de la for?t, Universit? Laval Research. f Fungi of doubtful genus identity based on the multigene dataset are shown in brackets.   68          69 Figure 2.1 Multigene phylogeny of ophiostomatoid fungi   The tree is one of the nine MPTs of the combined nSSU+5.8S+nLSU-rDNA+!-tubulin dataset from Ambrosiella, Raffaelea and Dryadomyces taxa with the selected members of different ophiostomatoid teleomorph and anamorph genera including the Grosmannia. Bootstrap percentage values (! 50%) generated from 1000 replicates from maximum parsimony and posterior probabilities (! 80%) from Bayesian analysis are shown on the branches. Posterior probabilities of 100% are shown by *. Thickened black branches represent the major ambrosia clades of genera Ambrosiella, Raffaelea and Dryadomyces produced from the combined dataset. The taxa given in bold represent all the ambrosia fungi included in the analysis. The beetle association (ambrosia and bark beetles) is also mapped onto the phylogenetic tree. Taphrina populina and Penicillium expansum from orders Taphrinales and Eurotiales, respectively, were used as outgroup taxa to root the phylogenetic tree.           70  Figure 2.2 Sporodochium-like formation  Scanning electron micrograph of Ophiostoma-related Ambrosiella (UAMH 10635). (A) Low magnification of sporodochium-like structures developed from an interwoven mat of hyphae. (B) Close-up of the conidiophores and annellidic conidiogenous cells comprising a portion of the sporodochium-like structures indicated in figure 2.2A (arrow). Bar = (A) 300 ?m, (B) 2.5 !m.          71  Figure 2.3 Annellidic conidiogenous cells   Scanning electron (A?C) and light micrographs (D) of Ophiostoma-related Ambrosiella spp. (A, B) Annellidic conidiogenous cells (arrowed) of Ambrosiella sp. 1 UAMH 10632 and Ambrosiella sp. 2 UAMH 10635, developing from mononematous conidiophores tapering toward the apex. (C) Conidiogenous cell of Ambrosiella sp. 2 UAMH 10635 showing apical conidia produced through sympodial proliferation (arrowed). (D) A. tingens CBS 366.53, producing conidia through the non-phialidic conidiogenesis. Bars = 2.5 ?m.   7 2  Chapter 3 Gene genealogies reveal cryptic species and host preferences for the pine fungal pathogen Grosmannia clavigera  3.1 Introduction  Because of global trade, and environmental and climate changes, phytophagous insects and insect-vectored fungi that are pathogenic to trees have the potential to undergo rapid population expansion and cause substantial ecological changes (Anderson et al. 2004) . A key aspect of estimating risks to ecosystems due to the spread of native or introduced pathogenic species involves defining species boundaries and genetic diversity. A growing number of fungal pathogens that were originally reported as dispersed generalists are now described as collections of populations or sister species adapted to new hosts or environments (Burnett 2003; Giraud et al. 2008) . However, like many other organisms, it is difficult to define species boundaries in fungi. Molecular approaches, such as phylogenetic species recognition by genealogical concordance (Taylor et al. 2000; Dettman et al. 2003) , can be more effective than traditional concepts. While it can be challenging to identify a genetic threshold that defines a species boundary, it is becoming increasingly practical to generate genomic sequence data for delimiting species with many independent gene genealogies (Knowles and Carstens 2007) .  Native bark beetles and their fungal associates, which evolve within coniferous trees, are among the most damaging forest pests in North America (Harrington 2005).    73 The current Dendroctonus ponderosae (mountain pine beetle: MPB) outbreak is the largest epidemic in recorded history. It has affected more than 18 million hectares of Pinus contorta forest in western Canada (www.for.gov.bc.ca/hfp), leading to major impacts on ecosystem dynamics and associated economic (Kurz et al. 2008). MPB normally remains at low population levels within pine forests for many decades, but can rapidly erupt into large-scaled outbreaks, killing large areas of susceptible host trees (Raffa 1988). Climate change and large areas of susceptible host trees likely contribute to the epidemic expanding northward and into high-elevation pine forests, beyond the MPB?s recorded historical range (Safranyik et al. 2010).  Population studies of both beetles and fungal associates (i.e. Grosmannia clavigera) confirmed population expansion in the northern part of the beetle/fungal species range, where outbreak activity is currently increasing (Mock et al. 2007; Lee et al. 2007; Roe et al. 2011). Further, if conditions continue to be suitable for MPB in its current geographic range, there is a risk that the outbreak will expand eastward into the boreal forests via P. banksiana (Logan and Powell 2001; Bentz et al. 2010; Safranyik et al. 2010).   One of the most common characteristics of bark beetles is their association with the wood-colonizing filamentous ascomycetes grouped as ophiostomatoid fungi (Six and Wingfield 2011). Grosmannia clavigera is an ophiostomatoid tree pathogen that forms a symbiotic association with MPB (Robinson and Davidson 1968) and its sister species D. jeffreyi (jeffrey pine beetle: JPB; Six and Paine 1997). While MPB and JPB have only subtle phenotypic and genetic differences, they inhabit different host trees. JPB is highly specialized, infests only P. jeffreyi, and has no history of large-scaled epidemics,   74 whereas MPB inhabits its primary host P. contorta and 22 other pine species, but not P. jeffreyi (Wood 1982; Safranyik et al. 2010). These bark beetles carry similar mycoflora and their geographic distributions overlap in some regions of the USA (Wood 1982; Six and Paine 1997; Kelley and Farrell 1998). G. clavigera is one of the most important fungal associates of MPB and JPB (Robinson-Jeffrey and Davidson 1968; Six and Paine 1997; Lee et al. 2006a; Rice, Markus N. Thormann, et al. 2007), and a central component of current MPB epidemics. Vectored fungi benefit from the association because the beetles carry them through the tree bark into a new host?s tissues (Six and Wingfield 2011). The benefits reported for the beetle and its progeny include the fungi providing a suitable substrate for brood development, participating in weakening tree defenses, and serving as a source of nutrients (Raffa and Berryman 1983; Harrington 2005; Bleiker and Six 2009; Lieutier et al. 2009; DiGuistini et al. 2011). While both beetle and fungi contribute to tree death, G. clavigera can kill trees without the beetle when manually inoculated into the host at a certain density (Solheim and Krokene 1998; Lee et al. 2006b). Mountain pine beetles have also been reported to attack other coniferous genera in epidemic regions; however, the beetles were only colonizing and reproducing inside Picea trees that have been baited with pheromone (Huber et al. 2009; Safranyik et al. 2010).  While species identification is important to understand the ecology and biology of organisms, boundaries between closely related species often lack clear limits and diagnostic characteristics. In G. clavigera, the sexual state (teleomorph) has been rarely found in nature and is not produced in the laboratory (Lee et al. 2003). Therefore, the   75 fungus is predominantly haploid through its life cycle and is known to mainly reproduce asexually (Lee et al. 2007; Six and Paine 1999). Because teleomorphs are rare, morphological identification relies on asexual structures, which occur in a variety of forms including the anamorph Leptographium (Jacobs and Wingfield 2001; Six et al. 2003). Conidiophores and conidia in G. clavigera show a great variation of shapes and sizes that can become confusingly indistinguishable from those of other Leptographium-forming species (Tsuneda and Hiratsuka 1984; Six et al. 2003). Early molecular studies have not been successful in separating G. clavigera from morphologically similar species, due to the lack of diagnostic DNA substitutions in the loci that have been commonly used for systematics (Zambino and Harrington 1992; Six et al. 2003). Multigene phylogenies, using ribosomal DNA and housekeeping genes (e.g. actin, elongation factor 1, alpha and beta tubulin), have provided better resolution for separating G. clavigera from its most closely related morphological species (e.g. MPB-associate L. longiclavatum), with the exception of a generalist fungus L. terebrantis (Six et al. 2003; Lim et al. 2004; Roe et al. 2010).  Based on morphological and phylogenetic species criteria, G. clavigera has been found colonizing different species of pine infested by MPBs and JPBs across western North America (Six et al. 2003; Lim et al. 2004). Using AFLP and microsatelitte makers, as well as multilocus sequencing, G. clavigera?s population structure has been mainly studied for those associated with the MPB epidemics in BC, Alberta and the USA (Lee et al. 2007; Roe et al. 2010; Tsui et al. 2012). All these studies have shown low genetic divergence and low nucleotide diversity between and within different epidemic   76 populations with a northern to southern pattern of differentiation. AFLP markers have also suggested the presence of two genetically distinct groups within G. clavigera associated with MPB epidemics infesting P. contorta in the Rocky Mountain of Alberta (i.e. Banff) and the northern USA (Lee et al. 2007); however, these groups have not been supported by microsatellite markers (Tsui et al. 2012). Similarly, using multilocus sequencing, Roe et al. (2011) reported no evidence of the two AFLP groups; but in this work the sampling regions focused on BC and Alberta epidemics and did not include the region where the second AFLP group had been reported.  Beetle-associated plant pathogens like G. clavigera depend on beetle vectors and host trees to complete their life cycles (Harrington 2005). Therefore, detecting genetic isolation in relation to the degree of host specialization or evolution of symbiosis is relevant to this group of fungi. As they grow, G. clavigera develop tight ecological and/or biological association with different species of pines, as well as with the two distinct sibling beetle species MPB and JPB. Six and Paine (1998) suggested that G. clavigera isolated from MPB or JPB had different tolerance to host defense metabolites. These beetles segregate in different ecological niches; as such we hypothesize that this segregation may have resulted in genetic divergence of their fungal associate G. clavigera producing a complex of distinct phylogenetic species adapted to different host tree species.   Defining species boundaries is essential for understanding the potential adaptive variations and the ecological and/or biological traits that may impact the evolution of   77 beetle-associated fungi. Theoretical models that incorporate adaptation and divergence among pathogens are applicable to risk assessment and to developing control measures, and detailed genetic information on evolving species should improve such models (Giraud et al. 2010) . Currently, information on genetic variation in G. clavigera is limited to few protein coding genes and non-coding markers (Six and Paine 1999; Lee et al. 2007; Tsui et al. 2009; Roe et al. 2011). Here, we screened nucleotide polymorphisms in 67 loci and applied ?p hylogenetic species recognition by genealogical concordance? (Taylor et al. 2000; Dettman et al. 2003)  using a subset of 15 protein-coding genes to assess whether genetically isolated lineages occur within G. clavigera, and whether host beetle and/or tree specialization may influence the evolution of these fungi. We combined the sequence data from the 15 loci to clarify how the phylogenetic species were related to each other. We show evidence of recombination in these apparently asexual fungi. Finally, we conclude that G. clavigera consists of Gc and Gs lineages, and discuss the ecology and biology of these fungi.   3.2 Materials and methods  3.2.1 Samples  We examined 166 iso lates of G. clavigera and eight additional isolates of its four closely related species G. aurea, L. longiclavatum, L. terebrantis and L. wingfieldii (Table 3.1 and Appendix A.1).  The G. clavigera isolates were collected from the two   78 beetle associates MPB and JPB and their host trees at different sites in Canada and the USA. In California, where both MPB and JPB are present (i.e. sympatric regions), we collected 30 and 25 isolates from P. jeffreyi and P. contorta, respectively, as well as a few isolates from P. ponderosa. We also included G. clavigera from locations where only MPB is present (i.e. allopatric regions). Sixty-seven isolates were from P. contorta in Canada and the USA, 29 isolates from P. ponderosae in South Dakota and British Columbia (BC), and a limited number of isolates from other MPB-host species, including P. albicaulis in BC and P. strobiformis in Arizona. Our fungal collection provides a comprehensive coverage of the beetles? geographic distribution. This included samples from current MPB epidemics in Canada, Idaho and Montana as well as from previous outbreaks in the 1960s and 1980s. It also included samples from small, geographically isolated outbreak populations in South Dakota, California and Arizona. In the work described here, we refer to such populations as ?localized?. Figure 3.2a shows the sampling locations.   3.2.2 Isolation   Fungal isolations from beetle exoskeletons or mycangia, or from galleries in infested trees were carried out following the methods described by Six and Paine (1997) and Massoumi Alamouti et al . (2007). Identification and molecular analyses were done from single-spore isolates. All cultures are maintained at the Breuil culture collection (University of British Columbia, Canada). Morphological features were determined from colonies grown on 2% MEA (20 g Difco malt extract, 10 g Difco agar and 1 L distilled   79 water) or from fungi taken from beetle galleries. The G. clavigera reproductive structures were examined and compared to those described by Robinson and Davidson (1968), using light microscopy.     3.2.3 Polymorphism detection   We identified polymorphic loci using two approaches. The first approach involved sequencing 28 candidate genes that were available from the G. clavigera genome sequence and EST-supported gene predictions (DiGuistini et al. 2007; 2009; 2011), followed by polymorphism discovery and verification. The second approach involved sequencing 39 putative polymorphic loci identified from an expressed sequence tag (EST) library obtained by pooling mRNA from eight G. clavigera isolates (DiGuistini et al. 2009) that were characterized as distinct haplotypes (i.e. unique sequence type) using the first approach. The target loci were identified from the genomic resource using CLCbio Genomics Workbench (CLC) 3.7.1 (Aarhus, Denmark). To discover polymorphisms, we sequenced the 67 loci across nine G. clavigera isolates (Table 3.1) chosen from distinct ecological and geographical sources. The sequences were aligned and analyzed for polymorphisms using CLC. We validated the novel polymorphisms in 15 genes (Table 3.2) selected for further characterization in an additional 53 G. clavigera isolates and eight isolates of four closely related species (Table 3.1). Genomic locations and gene descriptions of the 67 G. clavigera loci screened for polymorphisms are listed in Appendix A.2 and concatenated alignment of these datasets are deposited in TreeBASE (TB2: www.treebase.org).   80  3 . 2 . 4  DNA extraction , primer design and sequencing   We followed DNA extraction method by M?ller (et al. 1992) for mycelia grown on 2% MEA (33 g Oxoid malt extract agar, 10 g Technical agar No.3, and 1 L distilled water) plates overlaid with cellophane (gel dry grade, BioRad). Primer pairs were designed with optimal melting temperatures of 58?62?C, using CLC (Appendix A.2). PCR amplifications were performed following standard methods (Lim et al. 2004). Amplicons were purified and sequenced at the Sequencing and Genotyping Platform, CHUL Research Center (Qu?bec, Canada). Sequence data were collected from one strand, except for new haplotypes, which were all confirmed by sequencing both strands. All sequences are available at GenBank (accession nos. HQ633073?HQ634118).   3 . 2 . 5  Sequence alignments and analyses   Sequences were edited and aligned using Geneious 5.1 (Biomatters Ltd, New Zealand). Coding, intronic and untranslated (UTR) regions were determined based on alignment of DNA sequences to the G. clavigera genome sequence and gene prediction models. Genetic diversity indices and divergence analyses were assessed using DnaSP 5.10 (Librado and Rozas 2009). Net nucleotide divergence (Dxy) (Nei 1987) was   81 calculated with the Ta mura-Nei gamma correction model using Mega 4.0 (Tamura et al. 2007).  3.2.6  Gene trees and concatenated data phylogeny   Phylogenetic analyses were conducted using maximum parsimony (MP) and Bayesian inference of each of the 15 genes, as well as the combined dataset of these genes. T he best-fit model of sequence evolution for each gene was determined using the Akaike information criterion (AIC) implemented in JModelTest 0.1.1 (Posada 2008). MP trees were identified using PAUP* 4.0b10 (Swofford 2003) by heuristic searches and 100 random sequence additions. G aps were treated as missing data and no weighting was introduced in single gene analysis. Confidence was examined using bootstrapping (BS) with 1000 replicates and the heuristic option (Felsenstein 1985). Bayesian analyses were run using MrBayes 3.2 (Ronquist and Huelsenbeck 2003) , under the best-fit substitution model. Each run consisted of four incrementally heated Markov chains , with default heating values . The chains were initiated from a random tree, and were run for 2 million generations with sampling every 1000 generations. Posterior probabilities (PP) were inferred with a 50% majority -rule consensus tree sampled after the likelihood scores had converged . The 15 nuclear genes were concatenated to conduct partitioned maximum likelihood (ML) analysis (with 1000 nonparametric replicates bootstrap) using RAxML -VI-HPC 7.0.4 (Stamatakis 2006) and partitioned Bayesian analysis. The partitioned -ML and -Bayesian analysis utilized the substitution models selected by the AIC in JModelTest for each gene locus. The   82 combined dataset was also analyzed with weighted parsimony, with the weighting inversely proportional to the number of parsimony informative characters at each locus. Weighting allowed each locus to contribute equally to the combined data tree. All phylogenetic trees were rooted using G. aurea as outgroup (Massoumi Alamouti et al. 2007). Monophylies supported by both BS ! 70% and PP ! 95% were considered as significant.   Constraints on topologies were applied in PAUP* and the Wilcoxon signed-rank (WSR; Templeton 1983) test was employed to assess significant differences among topologies. For this test, up to 100 MPTs recovered were used as constraint topologies. When testing the constraint of lineage-specific monophyly, the lack of significance in the WSR tests indicates that nonmonophyly could be the result of insufficient phylogenetic signal.  3.2.7 Network approaches and evidence for recombination in G. clavigera   For each of the 15 gene datasets, we generated parsimony networks of G. clavigera haplotypes, which is described in the supporting information Method S1. We applied three approaches to detect the presence or absence of recombination in G. clavigera. First, we applied the index of association (IA) to estimate the extent of clonality in G. clavigera, using the program Multilocus 1.3b (Agapow and Burt 2001). IA determines to what extent individuals that are the same at one locus are more likely   83 than random to be the same at other loci. We used 10,000 randomizations on the subset of polymorphic sites that showed the most balanced distribution of alleles in each gene (i.e. excluding the uninformative sites). The test assumes an infinite amount of recombination so significant departure (p<0.05) from simulated recombined datasets suggest the presence of clonality (Smith et al. 1993). Second, we used the NeighborNet algorithm (Bryant and Moulton 2004) for decomposition analysis with SplitsTree 4.10 (Huson and Bryant 2006) to visualize the incongruence generated by recombination from the pairwise distance matrix of the G. clavigera concatenated sequence dataset estimated under the GTR model. Third, we estimated the pairwise homoplasy index (PHI; Bruen et al. 2006) in SplitsTree. Using a 100 bp window, compatibility among sites was calculated and, assuming no recombination, significance was determined with a permutation test.   3.3 Results   3.3.1 Polymorphism discovery  Sixty-seven loci, which represented 50 nuclear protein-coding genes with predicted functions, were sequenced and screened for polymorphisms (Appendix A.2). Some genes were constitutively expressed, e.g. housekeeping genes; others were differentially expressed in specific growth conditions, and were potentially involved in growth, metabolic processes or host tree pathogenicity (DiGuistini et al. 2007; 2009;   84 2011). A number of genes lacked significant homology with proteins or domains of known functions. We obtained ~50 kb of high quality sequence data for these genes in nine G. clavigera isolates (TB2:S11053) and identified 128 polymorphic sites (i.e. substitutions) across 33 genes. The majority of variations (63%), i.e. 81 single nucleotide polymorphisms (SNP) in 31 genes, separated the seven isolates representing the MPB associates at the epidemic sites from the two other isolates: G. clavigera holotype (ATCC 18086) and JPB associate (DLS1575). A subset of 18 informative (i.e. shared by two or more isolates) SNP (14%) in 12 genes, were exclusive polymorphisms that segregated only within the seven epidemic isolates. The rest of polymorphisms were substitutions that were unique to one isolate (i.e. singletons).   3.3.2 Polymorphism validation  For further analysis, we selected 15 genes (Table 3.2) that showed different levels and patterns of variation in the polymorphism-discovery panel and sequenced them in an additional 53 G. clavigera isolates (Table 3.1). These isolates were selected to represent the beetle associates MPB and JPB, and their respective primary host trees P. contorta and P. jeffreyi, as well as a few other MPB pine host species. Within G. clavigera isolates, we identified a total of 86/13,198 (0.65%) base substitutions and three indels in the concatenated 15-gene dataset. No site had more than two alleles (biallelic). The most polymorphic gene region was CFEM-II (!= 0.0039) and the least polymorphic was alpha-tubulin (!= 0.00073). Of the 86 polymorphic sites, 68 were informative and 18 were singletons. Eighteen of the changes were predicted in   85 noncoding locations (i.e. intronic and UTR), and, for the 68 that were in coding regions, 33 were synonymous and 35 were non -synonymous. The vast majority of variants were identified either as fixed SNPs (n=33) or as exclusive polymorphisms (n=49) that segregate only within one of the two potential G. clavigera lineages. The remaining four SNPs and one indels were the only shared polymorphisms found. The number of sites in the different classes of polymorphisms for each gene as well as for the concatenated dataset is shown in Tables 3.3 and Appendix A.3.   3.3.3 Single-gene phylogenies, phylogenetic species recognition and concatenated phylogeny  Using 15 gene phylogenies, we assessed G. clavigera species boundaries and phylogenetic relationships with related pine-infesting fungi: G. aurea, L. longiclavatum, L. terebrantis and L. wingfieldii. A summary of the phylogenetic data and model parameters inferred for each locus and the combined dataset are presented in Table 3.4. T he target genes were amplified in all species, except for TRPG and MPEP in the outgroup taxon G. aurea and anonymous-I in L. wingfieldii. MP and Bayesian consensus trees inferred similar topologies that are only shown for MP trees (figure 3.1). MP analyses yielded one to five trees for each locus, which mainly diff ered in the branching orders of two close relatives L. terebrantis and L. wingfieldii. The majority of gene trees (10/15) resolved the pathogen G. clavigera into two distinct clades. We referred to these clades as Gs with 40 isolates and Gc with 22 isolate s (figure 3.1). The Gs clade contained all isolates from epidemic MPBs, as well as those from localized   86 populations except for those collected from P. ponderosa trees. This clade was significantly (BS ! 70% and PP ! 0.95) supported by five loci (TRPG, MPEP, P450-I, LAH, anonymous -II). The Gc clade encompassed all JPB associates, as well as isolates from MPB that were infesting P. ponderosa trees in sympatric regions (California) including P. jeffreyi and P. contorta, and in allopatric regions of South Dakota. The G. clavigera holotype (ATCC18086; Robinson and Davidson, 1968) was also placed within Gc. This  clade was significantly supported by the same subset of loci that supported the Gs clade. Clades in  gene trees that did not agree with this partitioning were either not fully resolved (figure 3.1 40SRP, P450 -II) and/or not significantly supported (figure 3.1 CFEM-II: PP " 0.95 and/or BS " 70%). While one additional group showed a high level of support (figure 3.1: BS=100% and PP=1.0) in the TRPG  and another in the CFEM-I (figure 3.1: BS=75% and PP=0.95) phylogenies, we considered neither clade to be an independent lineage, since their partitions contradicted each other and neither was supported in the concatenated phylogeny (figure 3.2b). In the constraint analysis forcing the monophyly of G s and Gc, WSR results were significant (p=0.04) for only CFEM -II, indicating that incongruence from the constraint phylogeny is only significant in 1/15 of the loci.   Of the 15 genes, TRPG , MPEP and anonymous-II showed the highest resolving power for species boundaries, supporting five monophyletic groups: Gs, Gc, L. terebrantis, L. wingfieldii and L. longiclavatum. While species-level clades were strongly supported by a number of single-gene phylogenies, relationships between species were difficult to resolve. For example, L. terebrantis showed a non-robust phylogenetic   87 placement among trees, and it was collapsed into polytomy in at least ten single-gene phylogenies. Although positioning of some ingroup taxa varied among gene trees, TRPG, MPEP, LAH and anonymous-II significantly supported a sister group relationship between Gs and Gc.  The concatenated matrix of 15 gene sequences resulted in 13,239 bp of aligned nucleotide positions, 402 variable sites, and 226 informative characters (Table 3.4). MP, partitioned ML and partitioned Bayesian analyses resulted in similar topologies that had only minor differences in the placement of terminal taxa (figure 3.2b, ML tree). The topology of concatenated phylogeny was consistent with the single-gene tree partitions resolving the G. clavigera isolates into two monophyletic clades, and with the sister-group relationship between Gs and Gc (MP and ML BS=100%, PP=1.0).  Finally, we challenged our phylogenetic results by testing whether the polymorphism distribution of G. clavigera into two groups was due to independent evolutionary histories or to random sorting of genetic variations. The probability of observing different groups that, by chance, do not share polymorphisms, were tested by random shuffling the 15 dataset across (nonpartitioned dataset) and within (partitioned dataset) the two phylogenetic species. For the randomization, the association of polymorphic sites within each gene was left intact (i.e. each gene was randomized as blocks). In 1000 such randomizations, we found no partition that would create groups with no shared polymorphisms. Among 62 G. clavigera isolates, the shortest trees acquired from the nonpartitioned, randomized dataset were significantly (p < 0.001)   88 longer (510 ? 570 steps) than trees obtained from the randomized dataset considering the Gs and Gc partitions (61 ? 185 steps).   Within Gs (n=40), we identified 36 distinct haplotypes that were characterized by 33 base substitutions across 12 polymorphic genes (See appendices A.3 and A.5). Of the 33/13,198 (0.25%) polymorphic sites, 23 were informative. The number of haplotypes ranged from 1 to 4 among the genes.  Gene/h aplotype diversity (H) ranged from 0.0 to 0.73 in CFEM -II. The diversity over the combined dataset showed a high value of 0.99; however, genetic differentiation within the isolates was low, resulting in minor nucleotide diversity (!=0.00068). Similar  haplotypes did not cluster based on either geographic locations or the pine host species (See appendix A.5). Within Gs, we found seven isolates representing three identical haplotypes. Two isolates with the same haplotype were from the same California locality and two isolates with the same haplotype were from Alberta. Identical haplotypes were also isolated from different localities, two from BC and one from Arizona. The probability of identical haplotypes (i.e. isolates sha ring the same sequence type at all 12  polymorphic loci) resulting from random mating and recombination was small (4.2 " 10? 3  ?  1.8 " 10? 6), suggesting that they represent epidemic clones from the asexual reproductions. In comparison, the Gc isolates (n=22 ) showed a similar pattern but with a slightly lower level of nucleotide diversity (Appendix A.3). They represented a collection of 22 unique haplotypes (i.e. H=1.0) that could be distinguished by a total of 24 base substitutions across 12 polymorphic gene s. Of the 24/13,198 (0.18%) polymorphic sites, 14 were informative. As was the case for the Gs group, CFEM -II showed the highest level of both haplotype and nucleotide diversity.   89 However, some genes that showed a high level of variation within Gs isolates (e.g. TRPG, LPL, PLT) showed almost no polymorphisms among Gc isolates. Based on these genes dataset, haplotypes did not correlate with the host beetle/tree species, except for one allele in MPEP that was only found for MPB/P. ponderosa associates; however, this partition was not statistically supported.    3.3.4 Evidence of recombination  For Gs, we evaluated IA for all isolates, as well as for the reduced-by-haplotype dataset in which we excluded identical haplotypes. When all isolates were included (n=40), the IA=1.6 was significantly different (p=0.007) from the values obtained for the simulated recombined dataset, leading us to reject the null hypothesis of recombination. However, the IA=1.3 for the unique haplotypes (n=36) was indistinguishable (p=0.09) from that expected for a recombinant population (figure 3.3b?Gs). Within Gc, the IA=0.5 (p=0.3) also suggested recombination, both when all 22 isolates were included or only those from JPB (figure 3.3b?Gc). Split decomposition analysis also provided evidence for network relationships, giving a graphical support for the evidence of recombination and/or lineage sorting within both Gs and Gc (figure 3.3a). Finally, PHI provided another significant evidence (p= 0.00006) of recombination history.       90 3.3.5  Ecological and morphological characteristics   To assess the host and distribution ranges of the G. clavigera lineages in more detail, we sequenced a single informative locus, PCAS (Table 3.2) , in an additional 104 isolates (Appendix A.1, figure 3.2a). Locus PCAS contains two fixed SNPs (2/685=0.29%) that differentiate the two G. clavigera lineages and possesses exclusive polymorphisms that are not shared between Gs and Gc. This locus has been tested against a large number of other closely related species, and has been used as target-specific PCR -primers to detect and differentiate microbial communities associated with the MPB (Khadempour et al. 2010).  We generated the data for P. contorta-associated isolates (n= 67) from BC, Alberta, Montana and Idaho, as well as isolates from P. ponderosa trees in BC (n=13). Consistent with results from 15-gene phylogenies, the deeper single-locus sampling showed that the Gc group was not present in the epidemic populations of MPB; instead, the Gc lineage largely represented isolates from P. ponderosa  (n=18) and P. jeffreyi (n=30) trees  attacked by the localized populations of respective beetle associates MPB and JPB  in South Dakota and California. In contrast, Gs  (n=117) occurred on MPB in epidemic populations of the beetle and its pine-host species in western Canada and the USA, as well as in localized populations infesting P. contorta in California and P. strobiformis in Arizona. Both Gs and Gc were found in MPB and JPB localized populations in California, where the two beetle associates live in sympatry on P. contorta, P. jeffreyi and P. ponderosa. In South Dakota, where the localized population of MPB infests P. ponderosa trees, we found Gc (n=15) but no evidence of Gs.    91  We compared the reproductive structures of five isolates representing the Gs group to those of the G. clavigera holotype, which was included in our analysis and represented the Gc group. The anamorph (conidia and conidiophore) and teleomorph (i.e. ascocarp and ascospores) morphologies of Gs (figure 3.4) representatives agreed with the formal G. clavigera holotype description by Robinson and Davidson (1968). The conidiophores and conidia sizes varied among isolates, but all measurements (Appendix A.4) agreed with the G. clavigera original descriptions, as well as with descriptions of P. contorta associates (Robinson and Davidson 1968; Six and Paine 1997; Lee et al. 2003).   3.4 Discussion  We generated the first comprehensive dataset of protein-coding gene variability in the bark-beetle symbiont and pine pathogen G. clavigera. We used this dataset to characterize patterns of DNA polymorphism and divergence within the pathogen and among four close relatives that also inhabit pine trees. In contrast to the current taxonomy, our results show species diversity and ecological complexity with respect to host species. Paine and Hanlon (1994) and Six and Paine (1998) showed that the G. clavigera isolates of JPB were more tolerant to host oleoresin than those of MPB, suggesting some potential physiological differences between these two types of isolates. Here we suggest that the genetic divergence and diversity in G. clavigera   92 isolates may have resulted from the fungus adapting to particular pine species and to extensive expansion of the epidemic.   Our phylogenetic analyses identified two distinct lineages in G. clavigera. While the combined dataset of nuclear ribosomal DNA and the protein-coding genes have improved the phylogenetic positioning of G. clavigera (Lim et al. 2004; Roe et al. 2010), these loci failed to distinguish the two lineages identified in this study. These results indicate that the sequences currently available for ophiostomatoid systematics are inadequate for phylogenetic species recognition and inferring evolutionary relationships in the genus Grosmannia. We demonstrated that sequencing more genomic regions is more effective for inferring species boundaries. Given this, care should be taken when interpreting ecological characteristics of this group of fungi. The literature suggests speculative evolutionary processes (Six et al. 2003; Lim et al. 2004; Roe et al. 2011) that rely on data that are insufficient for identifying species and on an imperfectly known phylogeny. Six et al. (2003) and Lim et al. (2004) suggested that G. clavigera is a recently diverged morphological variant of the generalist fungus L. terebrantis . Our results show that L. terebrantis is a distinct species separated from both G. clavigera lineages; we also found that some isolates assigned as L. terebrantis were genetically different from the L. tere brantis  holotype isolated from D.  terebrans (Six and Massoumi Alamouti unpublished data), suggesting that this fungus represents a complex of closely related species that need to be taxonomically and ecologically re-assessed. Ecological descriptions of this species and other close relatives will be more valuable if based on solid taxonomic foundations.   93  Below, we provide two main lines of evidence to show that G. clavigera lineages represent two distinct species: a) they are evolutionary independent, and b) they are ecologically distinguishable. Because a lineage can represent a species, a clone, or a divergent group within a population, we will discuss these two concepts and discuss the evidence of recombination and ecological significance in each species. Current concepts agree that species correspond to ?segments of separately evolving lineages? (de Queiroz 2007); however, different characteristics (e.g. morphological, reproductive and nucleotide divergence) are used to infer boundaries for species, clones, and divergent groups. Such characteristics do not arise at the same time during the process of speciation, and so each type of evidence can lead to different conclusions regarding species boundaries (Avise 2004).  3.4.1  Evolutionarily independent lineages   Phylogenetic species recognition by genealogical concordance (Taylor et al. 2000; Dettman et al. 2003) stipulates that when lineages are separated for long periods of time relative to population size, genealogies from the majority of loci should be congruent. This criterion considers a clade to be an independent evolutionary lineage and a phylogenetic species if it is present in the majority of single locus phylogenies (Dettman et al. 2003). Here, the concordance of ten genealogies define G. clavigera   94 lineages as two sibling phylogenetic species, and suggest genetic isolation?even when the lineages occur in the same geographic region, as in California.  Enforcing topological constraints for the monophyly of Gs and Gc did not result in significantly worse fit of the data to the tree (compared with the fit to an unconstrained tree) for 14 of the genomic regions. This means that for three regions, lack of reciprocal monophyly and/or lack of nodal support were the result of insufficient phylogenetic signal.   In general, the pattern of gene genealogies and the level of polymorphism depend on the timing of the speciation event, historical population sizes, modes of reproduction, extent of hybridization and natural selection (Avise 2004). For only one (CFEM-II) of the 15 regions, the fit of the data to the constrained tree was significantly worse (at the ! = 0.05 level). For CFEM-II, we found no evidence for significant departure from neutrality (data not shown), intragenic recombination, or paralogs in the G. clavigera genome?s predicted gene models (DiGuistini et al. 2009; 2011). Therefore, none of these mechanisms can explain the incongruent pattern. Introgression can occur when interspecific hybridization results in the transfer of genetic material from one species into another, which leads to paraphyly of recipient species; alternatively, incomplete lineage sorting or recombination before species divergence can result in incongruent genealogies if species divergence occurred too recently for ancestral polymorphisms to have sorted into reciprocal monophyly (Avise 2004).    95  Distinguishing between interspecific hybridization and lineages sorting is difficult, because both result in the same pattern of incongruence (Hey and Nielsen 2004). However, when population genomic datasets are available, one could expect to find an alternative topology of a set of independent gene trees more frequently for hybridization event than other equally possible topologies under stochastic nature of lineage sorting (Huson et al. 2005). While we could not estimate the divergence time of G. clavigera lineages with certainty, due to the lack of fossils and the great variance in fungal nucleotide substitution rates (Kasuga et al. 2002), two observations suggest that these fungi diverged recently. First, the low interspecific nucleotide divergence (0.0037? 5.7 x 10?4) and the unresolved species phylogeny suggest that not all loci have reached reciprocal monophyly. Second, when we compared ingroup and outgroup taxa of two or four species, a large number of ancestral polymorphisms appeared to predate divergence, consistent with the speciation event being so recent that ancestral polymorphisms were retained.  3.4.2  Evidence  of recombination   Because classical phylogenetic trees can give only a snapshot of the actual complex relationships that can be encountered when intraspecific details are considered, we describe G. clavigera population structure with modified phylogenies using split decomposition analysis. In this, network relationships account for   96 recombination within both Gs and Gc that, in agreement with our gene phylogenies, are separated into two phylogenetic groups. Within each group, IA values were not significantly different from artificially recombined datasets, and the occurrence of unique sequence types suggested a history of recombination within each phylogenetic species that had created different combinations of alleles. While these results can also be explained by convergent or parallel mutations, the very low sequence divergence and lack of multiple alleles observed for each polymorphic site, even when compared against other close relatives, indicate that the most likely explanations are recombination (i.e. current and/or historical) and incomplete lineage sorting.   We also observed direct evidence for clonal propagation in Gs with the occurrence of the same haplotype over a wide geographic area. In this species, applying the IA test for all isolates and for the reduced-by-haplotype dataset suggested the existence of epidemic clonality (Smith et al. 1993). Overall, the recombination component appears greater in Gc (100% unique haplotypes and lower IA); however, concordant with the fungal asexual reproductions in natural environments (Six and Paine 1997), IA greater than zero still suggests some deviation from complete panmixia.  These results agree with the genomic analysis of G. clavigera sensu lato ; both suggested that this fungus is a heterothallic sexual species (Tsui et al. 2009; DiGuistini et al. 2011). Consistent with this, G. clavigera ascocarps have been occasionally reported at epidemic sites in one-year old MPB galleries (Robinson and Davidson 1968; Lee et al. 2003). Although a sexual state has yet to be reported for G. clavigera   97 associated with JPB under either field or experimental conditions, molecular results suggest history of recombination in this fungus. However, sexual reproduction seems to occur in older galleries when competition and predation increases and when environmental variables change. The asexual state is abundant in the galleries and pupal chambers during the active life cycle of the two beetles; as well it is abundant on artificial media used for fungal isolations. Systematic investigations with more isolates from different phases of the JPB life cycle may also allow the discovery of the sexual reproductive mode in Gc.  3.4.3  Ecolog ically distinguishable   Evidence for host-specific differentiation between the two G. clavigera lineages is as follows. While we expected that the Gc and Gs would be specific to beetle vectors, our ecological data indicate that one lineage (Gc) occurs on MPBs colonizing P. ponderosa as well as on JPBs infesting P. jeffreyi, whereas the other (Gs) is exclusively associated with MPBs. Gc was only isolated from two geographically distinct and localized US populations, one of which was populated with P. contorta, P. jeffreyi and P. ponderosa and the other only with P. ponderosa. In contrast, Gs was associated with epidemic and localized populations of MPB inhabiting P. contorta, as well as other pine species in the epidemic regions, but not with P. jeffreyi and the localized P. ponderosa supporting the Gc clade. Further, our phylogenetic data showed that G. clavigera from the same host species in different geographic areas are genetically closer than those collected from different host species occurring in the same geographic region (e.g.   98 California). While our data in some geographic areas were limited, preventing us from assessing the role of geographical isolation in speciation, overall, the data indicate that both a beetle vector?s preference for a host tree species, and the geographic isolation of the host species, can contribute to progressive differentiation of the vectored fungal species.  G. clavigera lineages develop all phases of their life cycles on host trees and are dispersed by their respective beetle vectors via a specific association (Harrington 2005). Between beetle generations, these fungi are protected and maintained inside the specialized beetle structures called mycangia. Given this, the fates of the mutualistic fungus and beetle partners are linked, and mating is more likely to occur between fungi within the specific host tree. Such a degree of inherent isolation has been suggested to facilitate adaptive differentiation in a large number of fungal plant pathogens recognized as complexes of specialized sibling species (Giraud et al. 2006). The frequent asexual reproduction and sexual recombination in fungi can also promote ecological divergence by creating new combinations of alleles and rapid reproduction of those combinations that favor host adaptation (Giraud et al. 2010).  During the early phases of a massive attack by a beetle-fungal complex, healthy standing pine trees release constitutive or induce defense chemicals such as oleoresin. To survive in such hostile and toxic environments, beetle-fungal complexes must have mechanisms for modifying or metabolizing tree defense compounds (DiGuistini et al. 2009; 2011). While pine species have similar chemical defense systems, there are   99 quantitative and specific chemical differences among pine species and even between populations of the same host species (Forrest 1980; Gerson et al. 2009). For example, !-phellandrene is the most abundant monoterpene in Pinus contorta while heptane is the major volatile chemicals in Pinus Jeffreyi (Mirov and Hasbrouck 1976; Smith 2000). Heptane has been found at moderate concentration in the hybrid between P. Jeffreyi and P. ponderosae but has not been reported in P. ponderosae. However, tree chemical data are limited, especially for P. ponderosae, which needs to be systematically characterized across its range in western North America. Given this, specific association of the fungal pathogen with a host tree may also be maintained by the ability of pathogen to overcome and adapt to a tree?s chemical defense systems.  Concordant with our results showing a distinct phylogenetic separation between P. jeffreyi  (Gc) and P. contorta (Gs) associates, Six and Paine (1998) showed that G. clavigera from P. contorta exhibit a poor growth in P. jeffreyi. They also indicated that JPB associates were tolerant to a wider range of host chemicals. These differences might be due to the pathogen adapting to tree?s chemical defense compounds, e.g. ! phellandrene being at higher concentration in P. contorta than either P. jeffreyi and P. ponderosae. Further, molecular phylogeny of Pinus species is concordant with the monophyly of G. clavigera from localized populations of P. ponderosae and P. jeffreyi, and with the separation of the Gc from these two pine species from the Gs of P. contorta. P. jeffreyi and P. ponderosae are genetically and morphologically close relatives; they can hybridize and are classified in the Pinus subsection Ponderosa    (Gernandt et al. 2009), while P. contorta is phylogenetically distinct and is classified in   100 the subsection Contorta (Krupkin et al. 1996) . Similarly, MPB genetic divergence related to host trees have been also reported between localized beetle populations in mixed P. contorta / P. ponderosa forests in California, Colorado, Utah and Alberta (Stock and Amman 1980; Stock et al. 1984; Sturgeon and Mitton 1986b; Langor et al. 1990; Kelley et al. 2000). Hopkins (1909) described MPB as two species, D. ponderosae and D. monticola. Although these species were synonymized by Wood (1982), they attack and breed in different pine species (Stock et al. 1984). Genetic studies using allozyme and AFLP markers have reported contradictory results: host -dependent (e.g. P. contorta versus P. ponderosa) differentiation between localized beetle populations for allozymes and no host-dependent differences between MPB populations for AFLP (Mock et al. 2007). However, tree species and geographic areas vary between these studies, and it will be necessary to sample additional populations in the eastern and southern portion of the MPB range, and from different host trees including P. contorta, P. ponderosa and P. flexilis to resolve these contradictory results.  While MPB can attack and breed in different pine species, it is important to note that localized populations of MPB prefer one host pine species, even when that species is intermixed with other species that MPB could colonize (Wood 1982; Langor et al. 1990). A combination of events may contribute to the accumulation of host-adapted genes in MPB localized populations; for example, selective pressures on developing broods imposed by different tree species, host preferences by the beetle, differences among trees and allochronic separation of beetles? emergence from different hosts (Sturgeon and Mitton 1982; Borden 1984; Langor et al. 1990). Localized populations are   101 also characterized by temporary small outbreaks that are often initiated by secondary bark beetles attacking stressed trees (Smith et al. 2010); beetle populations in such regions may maintain a stable diversity of fungal species for extended periods of time. In contrast, epidemic populations of MPB often contain a high number of beetles relative to the preferred pine species in a given geographic range, and so attack other pine species (Wood 1963; Logan and Powell 2001; Bentz et al. 2010; Safranyik et al. 2010) . Furthermore, during extensive outbreaks, MPBs have been reported occasionally as attacking and reproducing in non-host pine such as Picea when faced with a shortage of host trees (Huber et al. 2009).  Consequently the spread of epidemics, which is affected by host tree?s susceptibility, availability and continuity on lar ge geographic regions (Safranyik et al. 2010), may dilute or replace older fungal populations that have become host adapted during the non-epidemic phases (Sturgeon and Mitton 1982; Langor et al. 1990). Such a population change was suggested by AFLP analysis of both MPB and G. clavigera populations (Lee et al. 2007; Mock et al. 2007; Roe et al. 2011). Lee et al. (2007) reported two genetically distinct groups of G. clavigera associated with P. contorta in the epidemic regions; the major group contains 166 i ndividuals from BC and the Rocky Mountains and the second group include nine individuals from the Rocky Mountains. They suggested that the latter might represent the original population of the Rocky Mountains that was mixed with the larger group that was introduced into the region by the eastward expansion of MPB epidemic. Although representative isolates were included in our dataset, we found no evidence of these two MPB-associated G. clavigera groups. While support of distinct lineages based on independen t gene   102 genealogies would indicate more ancient divergence among these fungi, microsatellite makers have also not supported such a distinction (Tsui et al. 2012).   Although the data from localized populations (i.e. California and South Dakota) suggested that P. ponderosa might not be a preferred host of the Gs lineage, this tree species was found hosting Gs in the epidemic regions (BC and Rocky mountains). This might be the result of the current rapid expansion of MPB and the pathogen (Gs) from their primary preferred host P. contorta to other pine species, including P. ponderosa. The holotype (ATCC 18086; Robinson and Davidson 1968 ) is the only remaining isolate from P. ponderosa-infested trees before the current epidemic in BC. It clusters genetically with other current localized P. ponderosa-associates, and not with Gs isolates from epidemic regions; this is consistent with the MPB rapidly expanding its population and geographic range in the epidemic. While no other historical isolates of G. clavigera are available, we would expect to find additional evidence for host tree preferences among G. clavigera lineages by sampling populations from different infested-tree species in the eastern and southern portion of the MPB range, i.e. in areas that have not been reached by the current epidemics. If fungal lineages are adapted to host species, then lineages should correlate with host species locations; however, this assumes, simplistically (Thompson 1994) , that ecological constraints or genetic structure of host beetles/trees and pathogen are the sam e in different geographic regions. But they are not; both the beetle and host trees vary genetically and phenotypically between different geographic regions (Krupkin et al. 1996; Richardson 2000; Mock et al. 2007; Gernandt et al. 2009). And there are significant chemical   103 differences between trees at different geographic locations and with environmental conditions that need to be further characterized (Mirov 1948; Latta et al. 2003).   While the nomenclatural name G. clav igera is tied to the species that is genetically and ecologically represented by the holotype (Robinson and Davidson 1968), we showed that the fungus consists of Gs and Gc lineages. These are distinct sibling species that should be recognized taxonomically. Gc should retain the name G. clavigera , while Gs should be described as a new species. In the future we can anticipate that Gc genetic variation will evolve slowly while Gs might go through further genetic variation, and we outline two scenarios. In BC we already observed a post-epidemic phase in which the MBP population is decreasing and we anticipate that this population will collapse in the near future due to the lack of mature P. contorta . In the first scenario, only a small number of Gs haplotypes survive the MPB collapse and are maintained through the endemic cycle of the beetle until young pine trees reach maturity. At that point, in a future outbreak, the population and the fungal genetic diversity will increase, as it did in the current epidemic, leading to an array of closely related new haplotypes. In a second scenario that is potentially a shorter term concern a subset of the current large population in Alberta succeeds in becoming established in a new host tree species, and, with its fungal symbionts, adapts to the new physical and chemical environment presented by this host. There is evidence that this may already be occurring, as it has recently been shown that the beetle can successfully reproduce in the wild, in hybrids between P. contorta  and P. banksiana.  Significantly, P. banksiana  occurs across the northern Canadian boreal forest. While P. banksiana  is more closely   104 related to P. contorta than to P. ponderosa  or P. jeffreyi, landscape and environmental conditions prevailing in the boreal forest would lead the symbiotic partners to evolve as the MPB spread across the boreal forest. Extending the work described above could characterize how Gc and Gs populations are evolving, and so help to assess threats related to the above scenarios. Even if MPB does not become established on P. banksiana , in the near future climate change will affect geographic distributions of trees and beetles, and populations of fungal associates will evolve with vectors and hosts. Similar work on other MPB host trees or other beetle systems could establish accurate species diversity and provide a foundation for understanding ecological interactions of the ophiostomatoid group that includes the most common fungal symbionts associated with bark beetles.   105 3.5 Tables and figures  Table 3.1 Fungal isolates used in this study  Fungal species Beetle associate Host tree Collection site (Map no. a) No. isolates b Source c ID d Collector (Date sampled) Grosmannia sp. (Gs clade) Dendroctonus ponderosae Pinus contorta Canada, BC, Riske Creek  1 UAMH 4585 B01 Whitney (1982)  . . Canada, BC, Terry Fox Creek  1 NOF 1280  B02 Hiratsuka & Maruyama (1987)  . . Canada, BC, Houston (1) 10 UAMH (11153) B03 Lee (2003)   . . Canada, BC, Tweedsmuir park  (2) 1 CB SLA11  B04 .  . . Canada, BC, Williams Lake (3)  2 CB W14  B05 .   . . Canada, BC, Kamloops (4)  2 UAMH (11150) B06 Lee (2003)   . . .  UAMH (11151) B07 Lee (2004)   . . Canada, BC, Kelowna (5)  2 CB KDW4  B08 M. Alamouti (2007)    .  UAMH (11152) B09 .  . . Canada , BC, Manning Park (6) 5 CB M6 B10 Lee (2003)   . . Canada, Alberta, Westcastle  1 UAMH 4818 A01 Tsuneda (1983)  . . Canada, Alberta, Carbondale  1 NOF 842  A02 .  . . Canada, Alberta, Blairmore  1 NOF 2893  A03 Unknown (1983)  . . Canada, Alberta, Banff (7) 15 CB B20 A04 Lee (2003)   . . .  UAMH (11154) A05 .  . . .  CB B6  A06 .  . . .  CB BW26  A07 .  . . .  CB B14  A08 .  . . .  UAMH (11155) A09 .  . . Canada, Alberta, Cypress Hills (8)  5 (2 trees) CB (CHMC3)  A10 M. Alamouti (2007)  . . .  CB CHDS C7  A11   . . .  CB CHEBC10  A12   . . USA, Montana, Hidden Valley (9)  10 CB HV14  M01 Six (2003)  . . .  CB HV30  M02 .   106 Fungal species Beetle associate Host tree Collection site (Map no. a) No. isolates b Source c ID d Collector (Date sampled)  . . USA, Idaho, Hell Roaring (10) 10 CB D1128  I01 Six (2002)  . . .  CB D1151 I02 .  . . USA, California, Sierra Nevada (11) 2 DLS 1061 C01 Six (1995)  . . .  DLS 1037 C02 .  . . . 23 (5 trees) CB 12G13 C03 M. Alamouti (2009)  . . .  CB 23G23 C04 .  . . .  CB 55B11 C05 .  . . .  CB 68B21 C06 .  . . .  CB 710G16 C07 .  D. ponderosae P. albicaulis Canada, BC, Nelson (12) 5 CB Pa?9 B11 Blaiker (2007)  . . .  CB Pa?6 B12 .  D. ponderosae P. strobiformis USA, Arizona, Pinale?o mountains (13) 7 CB GCA02 Z01 Six (2009)  . . .  CB GCA04 Z02 .  D. ponderosae P. ponderosae Canada, BC, Kamloops (4) 8 (4 trees) CB PY2?3b B13 M. Alamouti (2007)  . . .  CB PY8?8 B14 .  . . Canada, BC, Kelowna (5) 5 (5 trees) CB KGW5 B15 . G. clavigera (Gc clade) . . Canada, BC, Cache Creek 1 (ATCC 18086) B16 Robinson-J. (1968)  . . USA, South Dakota, Black Hills (14) 15 (5 trees) CB 15B29C1 D01 Blaiker (2009)  . . .  CB 34B94C6 D02 .  . . .  CB 24B166C8 D03 .  . . USA, California, Sierra Nevada 1 DLS 15 C08 Six (1993)  . . USA, California, Lassen (15) 1 DLS 24 C09 .  D. ponderosae . . 1 DLS 56 C10 .  D. jeffreyi P. jeffreyi USA, California, Sierra Nevada 10 C 843 C11 Harrington (1993)  . . .  DLS 554 C12 Six (1999)  . . .  DLS 776 C13 .  . . .  DLS 833 C14 .  . . .  DLS 681 C15 .  . . .  DLS 771 C16 .  . . USA, California, Lassen 10  DLS 173 C17 .  . . .  DLS 210 C18 .   107 Fungal species Beetle associate Host tree Collection site (Map no. a) No. isolates b Source c ID d Collector (Date sampled)  . . .  DLS 237 C19 .  . . USA, California, San Bernardino (16) 1 DLS 52 C20 Six (1993)  . . . 9 DLS 1560 C21 A. Hansen (2006)  . . .  DLS 1565 C22 .  . . .  UAMH (11156) C23 .  . . .  DLS 1588 C24 .  . . .  DLS 1595 C25 . Total number of isolates     166 62   G. aurea Dendroctonus sp. P. contorta Canada, BC, Invermere  1 CBS 438.69 UB Davidson ( 1963) Leptographium longiclavatum D. ponderosae  P. contorta Canada, BC, Kamloops 2 CB SLKW1436  LB Lee (2003)  D. jeffreyi  P. jeffreyi USA, California, Sierra Nevada   DLS 845 LC Six (1999) L. terebrantis D. ponderosae  P. contorta Canada, BC, Kamloops 3 CB 878AW1-2 TB1  Kim (2004)  . . .  CB LPKRLT-3 TB2  Kim (2003)  D. brevicomis  P. ponderosae USA, California, Sierra Nevada   C 418 TC  Harrington (2003) L. wingfieldii Tomicus pin iperda P. sylvestris France, Orl?ans  2 CBS 645.89 WF  Morelet (1984)  NA P. brutia  Greece, Thessaloniki   CBS 648.89 WG  Skarmoutsos (1987) a Generalized map location of collection sites corresponding to figure 3.2a b Number of isolates analyzed for the ecological assessment using single-locus sequencing; Samples isolated and identified in this study are bolded; Isolates from the same locality are originated from different sources (i.e. from beetles and/or galleries collected from different tree individuals) ; otherwise number of isolation sources are shown in parentheses  c Isolates selected for 15 single-gene phylogenies; Source of isolates: UAMH, University of Alberta Microfungus Collection and Herbarium, Canada; NOF, Culture Collection of Northern Forestry  Centre, Canada; ATCC, American Type Culture Collection, USA; CBS, Centraalbureau voor Schimmelcultures, Netherlands ; Isolates beginning with CB, DLS and C are from culture collections of C. Breuil, University of British Columbia, Canada;  D.L. Six, Univers ity of Montana, USA; and T.C. Harrington, Iowa State University, USA; respectively; Nine isolates chosen for the polymorphism discovery are shown in parentheses   d Letters indicate the location and colors indicate the host trees corresponding to figure 3.2a; Numbers indicate the number of isolates from each location   108 Table 3.2 Primer sequence and gene description for loci used in phylogenetic and population genetic analyses  Primer sequence 5'! 3' G. clavigera sequence length in bp* Forward Reverse Gene description (abbreviation) Total Exon  Intron  UTR  GenBank accession no. TCACGCCCACCGTTACCGACA  TGGAAATGGTCGGTGCCGAGGT  40S ribosomal protein S3 (40SRP)  742  585 0 157  HQ633911 - HQ633980  TCCAGACGAACCTGGTGCCGT  CAGGCGTCATCGAGCAAG CGA  alpha-tubulin 640  489  59 92  HQ633073 - HQ633142  ATGTGCAGGGTGGCGAGCGAA  GAATACCGCTCCGCTCGCACA  ATP-binding-cassette multidrug transporter (ABC) 549  549  0 0  HQ633143 - HQ633212  TGATTCGACTTTCCCCCT  CGTCGAACACAAACTCCT  GGAGTTTGTGTTCGACGAG  GAATGACAAGGCTATGAAGGGA  Anthranilate synthase (TRPG)  1,925 1,925 0 0  HQ633981 - HQ634049  TAAGGAAAGGGAGGGCGGT  TGGGTGCGTGATGAGCGA  ATTCCCCTCCCCTACTCC CTTCCATGTCCTCCTTCC  Metallo-peptidase (MPEP)  1,672 1,386  215 71  HQ634050 - HQ634118  GACATTGTAGAGGGCAGC  AGATG GGAGGTTGGAGAG  AGTAGAACACCGCCGACAG  CCGACCAAACACACCGCA  Cytochrome P450 (P450 I)  1,596 1,440  113  43   HQ633213 - HQ633282  TGCAGCAATGGGACCGGATGA  TCGTCACGTTCTCCCAGCGCT  P450 II  710 710 0 0  HQ633283 - HQ633352  CACACGGACCAACGACGA  CTCTCCTGCCCCTCTTCTC  Lipid acyl hydrolase (LAH) 1,123  1,123  0 0  HQ633353 - HQ633422  CTCTTCTTTGCCGGCCTTGCTGT  CGCAACGCAAACGCCAGAAGA  Fungal extracellular membrane protein  (CFEM I)  667 510 58 99  HQ633423 - HQ633492  GCGTCCATTGATCGGCGTGATGT  AACCGCCAACATGGCAACGG  CFEM II  491  427  64  0  HQ633493 - HQ633562  TGCTGTCGAGAACTGGAGGCGT  CGGCAGGACCTGGAACAGGAA  Lysophospholipase (LPL) 568 443  125 0  HQ633563 - HQ633632  CGGTCGCCCGCTCTACATTGA  CTCAGCCTCTAAGCCGTTGCCT  Phosphatidylinositol transferase  (PLT) 570 570 0 0  HQ633633 - HQ633702  TGCCGACAA GGTGGCCAAGTTC  GCGCAGCGCAACATTGACGACT  Peroxisomal-coenzyme A synthetase (PCAS) 685 117 24  544   HQ633703 - HQ633772  CACGACGACGAACTCCTCTCCCA  CAGGATGCCCTCGGCCTCTAAC  Anonymous I  455  296 3  156  HQ633773 - HQ633840  ACGCCGGCAAGACCTACACCA  TGCCAGACTGGTCCACATCTGCA  Anonymous II  805 240  61 504   HQ633841 - HQ633910  Concatenated dataset 13,198  10,810 722 1,666   *Base pair     109 Table 3.3 Fixed and shared polymorphisms between the two monophyletic clades in G. clavigera  Fixed polymorphisms  Shared polymorphisms    Exclusive polymorphisms to  Genetic differentiation a Locus Total Noncoding  Synonymous Replacement  Total SNPs  Indels (bp)  Gs Gc  Dxy (10 ? 3) SD of Dxy (10 ? 4) 40SRP 0 0 0 0  0 0 0  2 1  1.27 3.4 alpha-tubulin 1 0 1 0  0 0 0  0 0  1.56 4.1 ABC  2 0 0 2  0 0 0  1 0  3.69 9.4 TRPG 7 0 3 4  1 0 1 (1) *   6 1  4.79 9.5 MPEP 4 2 0 2  0 0 0  2 2  3.44 7.8 P450 I 6 3 2 1  0 0 0  3 3  4.48 9.3 P450 II 0 0 0 0  0 0 0  2 3  1.36 3.6 LAH 4 0 0 4  0 0 0  0 1  3.33 8.7 CFEM I  0 0 0 0  0 0 0  5 1  2.81 5.6 CFEM II  0 0 0 0  3 3 0  1 2  4.18 6.9 LPL 1 0 0 1  0 0 0  2 0  2.43 5.5 PLT 1 0 1 0  0 0 0  2 1  2.32 5.1 PCAS  2 2 0 0  0 0 0  1 2  3.10 7.7 Anonymous I 2 0 2 0  0 0 0  0 1  4.51 11.6 Anonymous II 3 3 0 0  1 1 0  2 2  7.00 14.2 Conca tenated dataset 33 10 9 14  5 4 1  29 20  3.71 5.7 *Coding regions  a Dxy, net nucleotide divergence for the  pairwise comparison of the two monophyletic clades in G. clavigera ;  SD, standard deviation    110 Table 3.4 Information on phylogenetic dataset sequenced from G. clavigera and its close relatives  Homoplasy level Locus Sample size Total number of characters Variable sites Parsimony informative characters (PI) Number of tree steps (TS) Number of MP trees P I/TS  CI  Nucleotide substitution model 40SRP  70 742  23 11 23 1 2.09 1.00 TrN alpha-tubulin 70 640  10 8 10 1 1.25 1.00 HKY  ABC 70 549  10 6 10 1 1.67 1.00 HKY  TRPG  69 1,925 37 33 45  3 1.36 0.89 TPM1uf+I  MPEP  69 1,672 30 28 31 2 1.11 0.96 HKY+I  P450 I  70 1,597 53 29 55 4  1.90 0.98 TrN P450 II  70 710 17 8 18 3 2.25 0.94  TRN LAH  70 1,123 25 15 26 2 1.73 1.00 TPM1uf+I  CFEM I  70 673 31 12 33 4  2.75 0.96 TIM1+I  CFEM II  70 491  27 12 29 5 2.42  0.93 TIM1+I  LPL  70 569 28 13 30 2 2.31 0.97 HKY+I  PLT  70 570 10 6 10 1 1.67 1.00 HKY  PCAS  70 706 27 9 27 1 3.00 1.00 HKY  Anonymous I  68 467  22 9 22 1 2.44  1.00 HKY  Anonymous II  70 805 46  27 47  2 1.74  0.98 GTR Concatenated dataset 70 13,239 402  226 503  100 2.23 0.81 GTR Information with outgroup taxon.      111     1 1 2        113 Figure 3.1 Single-locus phylogenies of 15 genes in G. clavigera and its four close relatives.   Bootstrap (BS>50) and posterior probabilities (PP>0.8) from MP and Bayesian analyses are shown along the branches. Thick branches indicate nodes with PP ! .95 and BS ! 70. The two bars indicate the G. clavigera monophyletic clades color-coded according to their beetle associates: MPB (gray) and JPB (green). Trees are rooted with G. aurea, except for TRPG and MPEP that miss the outgroup taxon and therefore are midpoint rooted. Refer to figure 3.2a and table 3.1 for color codes and lables.       114       115 Figure 3.2 Fungal collection sites and 15?gene phylogeny of Grosmannia clavigera complex  a) Map of western North America showing fungal collection sites where only one (MPB: gray) or two (MPB & JPB: green) G. clavigera beetle associates are present. Host tree species are color-coded and number of fungal isolates from each tree species is shown in parentheses. b) ML analysis of 15-gene combined dataset showing how the species recognized by PSR are related to each other and to other closely related species. Thick branches indicate nodes with ML and MP BS values of 100 and the Bayesian PP of 1.0. Gs and Gc monophyletic clades are labeled with bars color-coded according to beetle associates: MPB (gray) and JPB (green). Letters indicate the collection localities and colors indicate host tree species corresponding to the map and table 3.1. Dashed line indicates an adjustment of scale.     1 16                   117  Figure 3. 3 Recombination analysis   a) Split decomposition analysis of the 15-gene combined dataset. The scale represents phylogenetic distances between the haplotypes using a GTR+I+G substitution model with parameters estimated using JModelTest. Colored boxes represent the two G. clavigera monophyletic clades: Gc (green box) and Gs (gray box). The i nterconnected networks are suggestive of recombination and/or lineage sorting within both Gc and Gs clades. The center of network where it branches to the Gs and Gc groups was slightly netted (implying that the data support conflicting deep splits as expected from the among-gene incongruences, e.g. CFEM -II, as well as generally for deep branches) but it has been simplified and shortened to fit the graph. The labels refer to G. clavigera isolates listed in table 3.1). The I A values for Gs and Gc clades are shown and compared against histograms of I A values for 10,000 simulated recombined dataset.             118  Figure 3.4 Asexual and sexual stage in Gs  Reproductive structures in Gs. Light micrographs of asexual stage characterized with mononematous (a) and synnematous (b) conidiophores reproducing conidia (*). Light micrograph of sexual structure (c) characterized by a spherical ascocarp oozing ascospores (*).            119 Chapter 4 Comparative genomics of the pine pathogens and the beetle symbionts in the genus Grosmannia  4.1 Introduction  Over tens of millions of years conifer forests around the world have provided unique ecological niches for native bark beetles and their fungal symbionts. Interactions between conifer hosts, bark beetle vectors and their fungal associates have influenced the evolution of tree chemical defenses (e.g. terpenoids), beetles and fungal symbionts (Seybold et al. 2000; Farrell et al. 2001; Jordal 2013). Although beetle-tree-associated fungi have significant effects on forest ecosystems, knowledge has improved only recently about the specificity for host trees or beetle vectors in this group of fungi (Wingfield and Seifert 1993; Kurz et al. 2008). Currently, little is known about the genetic differences that are associated with fungal speciation and adaptation, and that may have been shaped by evolutionary processes. Fungal diversification and specialization for hosts may depend on genetic differences that include genomic rearrangements, gene losses/duplications, and coding and non-coding sequence variants that may be under selective pressure in particular genes (Aguileta et al. 2009; Stukenbrock et al. 2010; Manning et al. 2013). The extent of adaptive processes at the genome level can be quantified by identifying genomic differences within and between fungal lineages that have recently diverged and specialized onto different host trees (Stukenbrock et al. 2010).     120  In North America, tree-inhabiting beetles and their fungal symbionts are among the most diverse and damaging forest pests (Harrington 2005; Jordal and Cognato 2012). For example, in western Canada alone, the mountain pine beetle (MPB; Dendroctonus ponderosae) and its fungal associates have killed over 18 million hectares of Pinus contorta forests (http://www.for.gov.bc.ca/hfp/mountain_pine_ beetle/facts.htm; http://cfs.nrcan.gc.ca/pages/276), dramatically altering forest ecosystem dynamics and forest-dependent economic activities (Kurz et al. 2008). Further, the recent spread of the MPB-fungal complex into Alberta and Saskatchewan, and into P. banksiana raises the risk that the epidemic will spread eastward into and potentially across Canada?s boreal forests (Cullingham et al. 2011). Of the fungal associates, the ophiostomatoid (Sordariomycetes, Ascomycota) Grosmannia clavigera sensu lato is crucial to the epidemic as an obligate symbiont of MPB and a pathogen of P. contorta that can kill living trees through beetle mass colonization (Lee et al. 2006). This fungus forms a symbiotic relationship with MPB and its sister species the jeffrey pine beetle (JPB; D. jeffreyi). While the two beetles are morphologically and genetically similar, they are adapted to different host trees (Six and Paine 1997). JPB is highly specialized and colonizes only P. jeffreyi, while MPB primarily inhabits P. contorta but can also successfully colonizes more than 20 pine species, but not P. jeffreyi (Wood 1982).  Complexes consisting of beetles, trees, and fungi provide unique systems for understanding ecological divergence or speciation (Thompson 1994; DiGuistini et al. 2011; Massoumi Alamouti et al. 2011). Theoretical studies suggest that dispersal of the     121 plant pathogen between hosts, and aspects of the pathogen life cycle can promote ecological divergence; e.g. reproduction is frequently asexual, and sexual recombination is constrained because it occurs within a host?s tissues (Giraud et al. 2006; Giraud et al. 2008) . Concordant with this theoretical framework, protein-coding genealogies have identified two cryptic species within G. clavigera (Massoumi Alamouti et al. 2011). One species (Gs) is an exclusive associate of MPB and its primary host tree P. contorta, while the other (Gc) is found on localized populations of MPB and JPB where these beetles colonize the closely-related P. jeffreyi and P. ponderosa. Although the two Grosmannia lineages can occur in the same geographic region (e.g. California), no evidence of gene flow between Gs and Gc was detected based on sequence analysis of 15 nuclear coding loci, suggesting that host tree species and beetle population dynamics are important factors in the evolution and divergence of these fungi (Massoumi Alamouti et al. 2011).  Defining species boundaries in G. clavigera has been difficult, because this fungus belongs to a complex that consists of many related species that have little morphological and genetic variation (Tsuneda and Hiratsuka 1984;  Zambino and Harrington 1992; Six et al. 2003; Lim et al. 2004; Roe et al. 2010; Massoumi Alamouti et al. 2011; Roe et al. 2011) . Genetic variation has been characterized using allozymes in ten populations of Gc associated with JPB in California, and usin g multilocus sequencing, AFLP and microsatellite markers in the epidemic populations of Gs in British Columbia and Alberta (Canada), Idaho and Montana (USA) (Six and Paine 1999; Lee et al. 2007; Tsui et al. 2012) . Based on these surveys th e Gs epidemic populations     122 were divided into four distinct groups with some gene flow and admixture between the groups. Molecular evidence of random mating and linkage equilibrium suggests that both Gs and Gc have life cycles with sexual stages (Massoumi Alamoutiet al. 2011; Tsui et al. 2012). However, these fungi are predominatly found reproducing asexually inside the beetles? galleries under the bark. Sexual fruiting bodies have been found occasionally in nature for Gs and Gc from P. ponderosae  but not for Gc from P. jeffreyi, and fruiting bodies that demonstrate successful crosses have not been produced in the laboratory (Lee et al. 2007; Alamouti et al. 2011; Tsui et al. 2012). While our recent genealogical study identified cryptic species, this differentiation was not evident using other methods (e.g. multigene phylogenies or DNA finger printing). Fungal genome sequence and structure can be highly variable across populations and species (Galagan et al. 2005; Raffaele and Kamoun 2012), and single nucleotide polymorphisms (SNP) are now widely used for characterizing population genetics because they can also provide evolutionary information on ecological adaptation and speciation (Brumfield et al. 2003; Morin et al. 2004; Raffaele and Kamoun 2012).   Recently we reported the genome sequence of a Gs strain (slkw1407) isolated from P. contorta trees in the epidemic region (DiGuistini et al. 2011). The ~30 Mb genome assembly consisted of 18 supercontigs and 8,312 protein-coding gene models. We characterized some aspects of the functional genomics of the fungus, including its interaction with host-defense chemicals (Hesse-Orce et al. 2010; Wang et al. 2013). This work suggested that Grosmannia can tolerate, detoxify and utilize host defense chemicals. Given that host defense chemicals vary among pine species (Keeling and     123 Bohlmann 2006; Gerson et al. 2009; Boone et al. 2011; Hall, et al. 2013a;b), here we hypothesize that genes involved in host-pathogen interactions, secondary metabolite production and fungal interactions and differentiation, like cytochrome P450s, monooxygenases, membrane proteins such as ATP-binding cassette (ABC) and major facilitator superfamily (MFS) transporters, polyketide synthases genes (PKS), and vegetative incompatibility genes may have diverged to a greater extent in response to selection in different host environments.  In the current work, we use the reference Gs genome to enable comparative analysis of evolutionary divergence in distinct populations of Gc and Gs. We sequenced eleven strains, assembled their draft genome sequences, and reported a comprehensive assessment of intra and interspecies genomic variations relative to the Gs reference sequence. We applied genome-wide SNP phylogenies of twelve Grosmannia strains and gene genealogies of additional strains to test whether the genome dataset confirm our recent genealogical study that Gs and Gc are distinct lineages and whether it provides further evidence of ecological and/or geographic divergence in these fungi. Focusing on SNPs that are predicted to alter proteins encoded by genes noted above that are likely to be crucial for fungal colonization of host trees, we assess evidence for fungi diverging as they evolve to adapt to different pine species (P. contorta, P. jeffreyi, P. ponderosa ). We identify genes that show evidence of adaptive selection, and relate these variations to differences in fungal ecology and biology.      124 4.2 Material and methods  4.2.1 Fungal samples  We sequenced eight Gs genomes from two distinct populations of MPB-infested P. contorta trees: a) epidemic regions in Canada and the USA, and b) localized populations in small geographically isolated outbreaks in California. We also sequenced three genomes of the sibling species Gc. The sibling group included two strains from JPB-infested P. jeffreyi trees in California, as well as the Gc holotype described by Robinson-Jeffrey and Davidson (1968) from MPB-infested P. ponderosa trees in British Columbia (BC). We deposited these fungi at the University of Alberta Microfungus Collection and Herbarium (UAMH) along with additional strains used for SNP validations, phylogenies and physiological studies (Table 4.1, Appendix B.12).  4.2.2 Illumina paired-end library construction, sequencing and assembly   Fungal mycelia were grown on 2% malt extract (MEA; 33 g Oxoid malt extract agar, 10 g Technical agar No.3, and 1 L distilled water) overlaid with cellophane. DNA from the mycelia was extracted using the method of M?ller et al. (1992). DNA samples were processed at the Genome Science Center (GSC, Vancouver, BC, Canada) for paired-end sequencing following Illumina protocols (Illumina, Hayward, CA, USA). The library for each strain was amplified in a single flow-cell and sequenced to either 50 or 76     125 nucleotide base (nt) reads on the Illumina Genome Analyzer (GA) II or IIx following the manufacturer specifications.  Genomes were assembled from paired-end reads of 200 base DNA fragments generated by Illumina sequencing using the ABySS assembler v1.2.7 (Simpson et al. 2009). Reads that passed the chastity filter were assembled with a relative short kmer (25?31nt) for higher sensitivity. The resulting contigs were used as single end reads along with the original paired end data and reassembled with a higher kmer (35?61nt), which has a higher specificity (Appendix B.1). The assembly was cleaned and gaps closed using Anchor (www.bcgsc.ca/platform/bioinfo/software). Ambiguous base calls were resolved by mapping the reads back to the assembly using BWA v0.5.9 (Li and Durbin 2009) and calling consensus bases using SAMtools ?mpileup? v0.1.18 (Li et al. 2009). Assembled contigs and scaffolds larger than 200 nucleotides were ordered and oriented using MUMmer (Kurtz et al. 2004) based on the Grosmannia published genome (slkw1407; NCBI Genome PID: 39837; DiGuistini et al. 2009). Contigs that did not align with slkw1407 were ignored for our analysis. The assembly statistics after ordering and orientation are shown in Appendix B.2.  4.2.3  Gene predictions and ortholog determination   To detect Grosmannia orthologs, we first generated gene annotations for each draft genome using the homology-based gene predictor genBlastG (She et al. 2011). We     126 used protein sequences from the slkw1407 reference genome as the query (n=8,312) and the genome of another strain as the target database. Gene annotations and pairwise homology between the slkw1407?s gene models and those from each genome were assigned based on genBlastG hits with an E? value cutoff of ! 1e-10 and a query coverage of >50%. The genBlastG output can result in redundant gene predictions when the query gene belongs to a multigene family, paralogous genes, or tandem gene duplications. Given this, for downstream analyses we applied a filtering procedure so that each genomic region would contain only one gene prediction with the highest global sequence percent identity (PID) to the query. The filtering procedure was carried out as follows: (i) all gene predictions were sorted by PID, (ii) for each two overlapping gene model, if the overlapping region was > 5% of the length for either gene, then on ly the prediction with higher PID was kept, (iii) all gene models were required to have PID " 70% to the query, and (iv) to avoid assigning paralogs to query-target pairs, the best match had to have a PID 10% higher than the next best match. Non -overlapping gene models with high similarity to the same query were reported as putative paralogs and were removed from analysis. For genes with alternative splicing variants, the longest transcript was selected to represent the gene. After filtering incomplete genes and discarding genes with frame-shifts, which could have been caused by the draft quality of the genomes, only high quality 1:1 orthologous genes were retained for analysis. Gene models for all Grosmannia genomes are provided in annotation files.gff (https://tria website). Appendix B.3 summarizes genBlastG output used to find the pairwise homology between reference gene models and those of each draft genome.      127 4.2.4 Mapping and variant calling  We performed variant calling among the Grosmannia  genomes by mapping reads from each strain to the slkw1407 reference genome sequence. Before read mapping, we filtered raw reads to remove low-quality and duplicate sequences u sing PRINSEQ lite v0.17.1 (Schmieder and Edwards 2011) . We discarded reads that failed the Illumina chastity filter (Haridas et al. 2011), contained uncalled bases, and had an average Phred -scaled quality  of less than 10 in the last 20 base calls. For the retained reads, the initial (5?) five nucleotides showed GC -content bias (data not shown), and were trimmed, leaving 45 and 71 nt reads for mapping. We also filtered potential duplicate reads resulting from amplification of the identical DNA fragments during library preparation and sequencing. The numbers of reads used for SNV calling and processing steps are shown in Appendix B.4.  For each strain, filtered reads were mapped to the slkw1407 reference genome sequence using BWA v0.5.9 (Li and Durbin 2009). Initial mapping results were converted into the indexed and sorted Binary Alignment/Map (BAM) format usin g SAMtools v0.1.18 (Li et al. 2009) and Picard v1.54 (http://picard. sourceforge.net ). To enhance the quality of the alignments for more accurate variant detection, we used Genome Analysis Toolkit (GATK) (McKenna et al. 2010) to locally realign the BAM files in complex regions, e.g. containing insertions/deletions (indels). For each alignment, BWA assigned a mapping quality score (MAPQ). We used reads with MAPQ greater     128 than zero to estimate the coverage and average read depth of final BWA alignments, using BEDtools v2.13.4 (Quinlan and Hall 2010) . The individual BAM datasets are available at https://. Once reads from individual s trains were mapped to the slkw1407 genome, we used SAMtools ?mpileup? to assess variant sites, applying Base Alignment Quality computation and a ? C50 argument to minimize alignment artifacts and base-calling errors. Single nucleotide variants (SNVs) were i dentified using the Bayesian variant calling models implemented in ?bcftool? (Li et al. 2008). After consensus base calling, we filtered the initial variants for strand and distance biases ( p-value<0.0001) using SAMtools ?vcfutils.pl?. The final set of high -quality calls also required a candidate site to be biallelic and to meet the following criteria: minimum Phred-scaled base calling score of 20, MAPQs of at least 30, read depths of more than four and less than 500, and a minimum 10 nt distance from indels. Variant calls that failed to meet these criteria were likely to be false positives. Because SAMtools ?mpileup? assumes a dipl oid model and our samples represent haploid genomes, we also removed heterozygote calls.   4.2.5 Verification of variant calls  To estimate the robustness of SAMtools results, genomic variants were also assessed using the SNV calling algorithm implemented in GATK v1.40 (McKenna et al. 2010). This method also uses a Bayesian model to estimate the likelihood of a site harboring an alternative allele for each sample. GATK was run on the same BAM files as SAMtools, using default parameters. GATK raw -variant calls were filtered in the     129 same manner as the SAMtools calls (see above). To estimate SNV false positives in our dataset, we generated Illumina paired-end reads for the slkw1407 reference strain (Appendix B.4) and assessed variant calls for these reads mapped against their own published genome, and identified 1,796 high-quality SNVs. Because the alternate base was present in all eleven genomes and also in Illumina read alignments from the slkw1407, a large percentage (93.2%) of these changes likely represent errors in the reference genome assembly. We removed these ambiguous calls from the final SNV dataset. For sixteen additional Gc and Gs isolates (Table 4.1), we also used PCR and Sanger sequencing to validate SNPs in the nine candidate genes listed in Appendix B.5. For the eleven Grosmannia strains, we aligned homologous contigs of the candidate genes to those of slkw1407 genome and gene models using progressive Mauve 2.3.1 (Darling et al. 2010). Primers were designed based on the alignment using Geneious 5.1 (Biomatters Ltd, New Zealand). Amplicons were purified and sequenced at the Nucleic Acid Protein Service Unit at the University of British Columbia (Vancouver, Canada). All sequences are available at GenBank (accession nos.).  4.2.6 Functional annotations for SNVs  SNVs were annotated as coding (synonymous and non-synonymous), intronic, flanking and intergenic, with SNPeffect v.2.0.5 (Reumers et al. 2006), using the slkw1407 genome?s sequence and predicted gene models. We assigned flanking regions (i.e. UTRs and putative regulatory regions) of 1,000 nt upstream and     130 downstream of the initiation/termination codons of the annotated slkw1407 gene models, unless neighboring gene sequences were within this range; for such cases we truncated the assigned regions. We also characterized, as a set of variants, SNVs that result in the loss or gain of a stop codon, which likely affect the integrity of the protein products. Orthologous genes containing a premature stop codon were labeled as pseudogenes. We assessed the accuracy of stop-codon variant calls using expressed sequence tag libraries (EST) and RNA -seq data from slkw1407 (DiGuistini et al. 2009) as well as Illumina reads from more than one strain within each Gs and Gc group. Finally, we applied Gene ontology (GO) functional enrichment analysis (Molecular function or Biological process) on pseudogene candidates. The GO term associations were determined for each slkw1407 reference gene models using Blast2Go v2.5.0 with the default parameters (Conesa et al. 2005) . Blast2GO was also used for a GO functional enrichment analysis; for that we performed the Fisher's exact test with a FDR correction to obtain an adjusted p-value between the candidate genes and the whole genome annotation.  4.2.7  SNVs clustering and phylogenomics   We used the genome-wide SNV data to determine phylogenomic  relationships and the nucleotide divergence among Grosmannia genomes. To assess genome-wide-SNV clusters among a relatively small number of Grosmannia strains, we used the non-parametric AWclust (Gao and Starmer 2008) R package, because it requires no model     131 assumptions (e.g. Hardy?Weinberg equilibrium) and is based on hierarchical clustering of a distance matrix rather than on allele frequency variation. We compared the clustering results with inferences from maximum parsimony (MP) and Bayesian phylogenetic analyses, for which we concatenated the genomic SNV dataset into one continuous sequence for each strain (total character=103,430). MP trees were identified using PAUP* 4.0b10 (Swofford 2003) by heuristic searches with TBR branch-swapping and the MULPARS option, and 100 random sequence additions. Bayesian analyses used MrBayes 3.2 (Ronquist and Huelsenbeck 2003), under the best-fit substitution model selected by the Akaike information criterion (AIC) implemented in JModelTest 0.1.1 (Posada 2008). Each run consisted of four incrementally heated Markov chains, using default-heating values. The chains were initiated from a random tree and were run for 2 million generations with sampling every 1000 generations. To assess the confidence of phylogenomic analyses, MP bootstrap (BS) values were calculated with 1000 replicates and the heuristic option (Felsenstein 1985) using PAUP*, and Bayesian posterior probabilities (PP) were inferred with a 50% majority-rule consensus tree that was sampled after the likelihood scores had converged, using MrBayes. The stationarity of likelihood scores for sampled trees was assessed in Tracer v1.5 (Rambaut and Drummond 2009), and the convergence was assessed using cumulative posterior probability plots in AWTY (Nylander et al. 2008) to assess split frequency within and between MCMC runs. The roots of the resulting trees were inferred by midpoint rooting. Mean nucleotide divergence (Dxy) was calculated using the Maximum Composite Likelihood method implemented in Mega 5.0 (Tamura et al. 2011), and was averaged     132 across 1,000 bootstrap replicates. The SNP character matrix used in the cluster and phylogenetic analyses is deposited in TreeBASE (TB..: http://www.treebase.org).  4.2.8  Gene genealogies and concatenated data phylogeny   To assess the monophyly of phylogenetic clades (or to assess biogeographic traits) resolved using the genome-wide SNP dataset, we randomly selected nine gene loci (Appendix B.5) that showed putative fixed differences between distinct Gs and Gc populations and sequenced them in 16 additional strains (Table 4.1). For each of the nine gene datasets, we generated MP and statistical-parsimony genealogies using PAUP and TCS v. 1.13 (Clement et al. 2000). Gaps were treated as missing data and no weighting was introduced in the single-gene analyses. The nine gene loci were concatenated to conduct maximum likelihood (ML) analysis (with 1000 nonparametric replicates bootstrap) using RAxML-VI-HPC 7.0.4 (Stamatakis 2006), as well as the weighted parsimony, with the weighting inversely proportional to the number of parsimony informative characters at each locus. We also performed Bayesian analyses for each gene and for the combined dataset. For Bayesian and MP analyses and for assessing their confidence and best-fit model of sequence evolution, we used the same criteria as those applied to construct SNP phylogenies. Monophylies supported by both BS !70% and PP!95% were considered significant. The multigene dataset is deposited in TreeBase (TB.. http://www.treebase.org ).      133 4.2.9 Detecting signature of selection and rate of protein evolution   For sele ction analyses with Gs-Gc multiple alignments, we first searched for genes that were orthologous to the 8,312 gene models of the reference strain slkw1407. We found an average of 8,064 orthologs for the eleven assemblies, ranging from 7,876 to 8,222 in Gs and 7,973 to 8,198 in Gc (Appendix B.6). We retained orthologs to 7,340 slkw1407 genes that matched at least four of the eight Gs and/or at least two of the three Gc genomes (n=972) and removed 3,864 of these because they had either fewer than two coding differences (n=3,377) or zero divergence (D N +D S=0; n=487). The selection analyses included the remaining 3,476 orthologs, which contained 19,616 nucleotide differences in coding sequences with a median size of 1,749 aligned bases, after excluding the gaps (Appendix B.7). The average number of Gs? Gc genomes in the aligned datasets was n=8.6.   To detect signatures of positive selection in Grosmannia we applied different methods to Gs and Gc gene predictions that were orthologous to the slkw1407 gene models. Fi rst we compared polymorphisms within Gs (n=4 ? 8 strains) with fixed substitutions (i.e. divergence) between Gs and Gc sequences. We considered synonymous and nonsynonymous differences and used two Gc strains from P. jeffreyi and P. ponderosa as the outgroup taxa. We used Gs-Gc multiple alignments of all genes with at least two coding differences that were aligned by MAFFTv7.023 (Katoh et  al. 2002) for their entire coding regions, and applied the McDonald? Kreitman tests (MK;     134 McDonald and Kreitman 1991) implemented in the MK.pl script (Holloway et al. 2007). We assessed whether the ratio of nonsynonymous and synonymous was statistically independent of differences being polymorphic (PN:PS) or divergent (DN:DS), using Fisher?s exact test. For each gene, MK results for the direction and degree of departure from neutrality were summarized using the neutrality index (NI; Rand and Kann 1996), after adding one pseudocount to each mutation class to eliminate zero counts. We also reported an unbiased NI estimate for differences across all the genes (NITG; Stoletzki and Eyre-Walker 2011).   Next, we applied maximum likelihood methods implemented in the Codeml from PAMLV4.0 (Yang 2007). We estimated Gs?Gc pairwise distances at nonsynonymous (dN) and synonymous (dS) sites for each gene, by setting parameters as follow: seqtype = 1, CodonFreq = 2, Runmode = -2 and the transition-transversion ratio !  estimated from the data (Goldman and Yang 1994). For this test, we used pairwise alignments of single coding sequences from each species, generated for all Gs and Gc strains; the number of pairwise comparisons ranged from 8 to 27 per gene. Threshold dS values were determined by plotting dN as a function of dS, excluding outliers from the main distribution. To test for further evidence of positive selection, we applied the ?site-specific? models M1a/ M2a and M7/M8 (Nielsen and Yang 1998). Only gene alignments displaying more than three fixed (DN+DS ! 3) and/or synonymous (DS+PS ! 3) differences were considered for this additional test (Stoletzki and Eyre-Walker 2010). M1a assumes that codons contain only 0<dN/dS<1 or dN/dS=1. We compared this with the alternative model M2a, which allows dN/dS for a site to be less than, equal to or     135 greater than 1. If dN/dS is significantly greater than 1, then adaptive substitutions are assumed to have occurred to fix nonsynonymous differences between species. If dN/dS<1, adaptive evolution may still have occurred on some fraction of all differences, but cannot be inferred with certainty. We also compared the null model M7, which assumes a beta distribution of 0! dN/dS! 1 across sites with the alternative model M8, which allows for positive selection . The log-likelihoods for the null and alternative models were used to calculate a likelihood ratio test statistic, which was then compared against the !2 df = 2 distribution (Yang 2007). The positive selection hypothesis was accepted if both alternative models M2a and M8 provided a statistically significant better fit to the data. For all the analyses, we removed low frequency polymorphisms (singletons) to avoid biases caused by slightly deleterious mutations regarding the prevalence of adaptive divergence (Fay et al. 2001; Li et al. 2008).  4.2.10  Physiological assessments    We characterized the monoterpene utilization of (+)-limonene as a carbon source by Grosmannia  strains from three different pine trees P. contorta, P. jeffreyi and P. ponderosa. For this experiment, we selected five Gs and Gc strains from independent samples of each tree species (total n=15, Table 4.1). The three-day fungal cultures actively growing on MEA were transferred to glass plates containing yeast nitrogen base minimal medium (0.17% YNB, 1.5% granulated agar), where 200 ?l of (+)-limonene (Sigma, Oakville, ON) were added onto two (2 " 4 cm) strip filter papers that     136 were placed inside the lid of each glass plate. The plates were sealed with DuraSeal film (Laboratory Sealing Film; VWR, Mississauga, Ontario, Canada) and incubated at 22?C in a sealed g lass container. Limonene was re-supplied biweekly; after six weeks, the mycelial plugs treated with limonene were transferred to normal MEA plates to check whether the fungus was killed or survived the chemical treatment. The control was YNB minimal medium  without monoterpene.   4.3 Results  4.3.1 Genome assembly, orthologs determination and SNV variants  For the eleven Grosmannia strains, we obtained assemblies ranging from 27.7 to 32.4 Mb (Appendix B.2). We found no significant evidence of genome rearrangements for any of the sequenced strains compared to the slkw1407 reference genome (Appendices B.2 and B.13). The eleven strains shared more than 8000 genes with an average sequence identify of 98?0.4% between Gs and Gc genomes. On average only 3% of genes were missi ng or highly divergent (<70% sequence identity) relative to the reference gene models (Appendix B.6 and figure 4.1b).  Assessing coverage for variant calling , we noted that between 86 and 99% of the filtered reads mapped to the slkw1407 genome sequence , providing an average read depth between 22x to 58x per strain (Table 4.2). On average, 94.1% of the slkw1407     137 genome (i.e. ~27.4 Mbp) was covered by ! 5 uniquely mapped reads, with a range of 90.0% to 97.8% coverage across the eleven genomes. C overage for slkw1407 gene -prediction models on average showed slightly higher depth than for the whole genomes  (Table 4.2).   We compared the variants called by SAMtools and GATK, which showed a high percentage of overlapping SNVs (n=91,763) between the two methods, and  used the SAMtools results because it generated fewer unique calls ( 12.7% of total 105,104)  than GATK  (21.9% of total 117,449) . Of 198,362 putative variants, 105,104 SNVs and 9,907 indels passed quality control and filtering, yielding 115,011 high -confidence differences across the twelve Grosmannia genomes. After removing ambiguous calls that are likely to represent errors in the reference genome assembly, we obtain 103,430 SNV sites with a mean transition-to-transversion ratio of 3.4 (Appendix B.9). We estimated a false positive rate of 4.4 "10-6 or one per 24,590 nts and a false negative rate of 0.046% for the sequenced regions.   4.3.2  Functional classification of genomic variants   We classified nucleotide variants for their potential functional and/or adaptive significance by characterizing the level of intra- and interspecific differences in different genomic regions. From 103,430 SNVs across twelve Grosmannia genomes, we identified 36,017 variants within the slkw1407 gene models. Of these genic variants,     138 5,826 were intronic and 30,191 were in coding exons, 14,889 of which were synonymous and 15,302 nonsynonymous (Appendix B.9). Of the nongenic variants, 24,589 were located in our predicted ~6000kb gene -flanking regions and 42,880 were intergenic. Because gene models in slkw1407 can overlap (DiGuistini et al. 2009) , 56 of the genic SNVs were identified in more than one gene region  (e.g. a variant in a coding region as well as in an intronic region). Among the coding variants, 262 variants in 218 genes in the total Grosmannia genomes were predicted either to cause a premature stop codon (n=226) or to eliminate a stop codon (n=36). Of these 262 variants, 92 that had truncated proteins and 3 that had lost a stop codon occurred in only one genome, 155 were found in at least two genomes, and 12 were observed in all eleven genomes. The latter 12 variants may indicate that the slkw1407 genome sequence has an error or a low frequency allele in these positions. For this analysis we removed the 12 variants that occurred in all eleven genomes, as well as the 95 SNVs that were found in only one genome, which removed 85 genes. Of the remaining 133 genes, 85 were slkw1407 gene-models with known functions (n=86 for 71 genes with premature stop SNV; n=14 for 13 genes with stop-loss SNVs; and one gene showed both a stop -gain and a stop-loss SNV; Appendix B.10). Blast2Go enrichment analysis of genes with known functions identified enrichment of stop-codon variants for members with oxidoreductase activity (31%, p < 0.001) within both  biological process (BP) and molecular function (MF) classifications, followed by genes involved in transmembrane transporters (16.9%) and nucleotide binding activities (18%) in BP and MF, respectively (Appendix B.11). Some of the enriched oxidoreductases belonged to gene families with known roles in detoxification (Appendices B.10? 11). For example a flavoprotein monooxygenase     139 (CMQ-6740) in the slkw1407-gene cluster (figure 4.2a?b), which was reported to have a role in detoxification and/or utilization of host-tree defense chemicals (DiGuistini et al. 2011), showed a stop-codon in both Gc strains from P. jeffreyi.  For this gene, we confirmed the variant by slkw1407 EST and RNA-seq data (Appendix B.10), as well as by an independent PCR validation of additional strains (total Gc n=16, Gs=12 and two other species n=4), showing that this mutation is unique to the Gc strains from P. jeffreyi (figure 4.2c).  4.3.3  Divergence classification of genomic variants    Across the twelve Grosmannia  genomes, approximately 67% (n=70,018) of the total number of SNVs were parsimony informative in that multiple strains contained alternate nucleotide bases. The remaining SNVs (n=33,412) were unique differences (i.e. singletons) in that only one strain showed the alternate nucleotide base. To characterize intra and interspecific variants, we assigned the informative polymorphisms to three classes: fixed, exclusive and shared (Table 4.3). Most SNPs were either fixed (n=37,712), or were exclusive to the nine Gs (18,871) or the three Gc (n=9,685) strains; the rest (n=3,750) were present in both species. Within Gs, the eight re-sequenced genomes differed from the slkw1407 reference genome by an average of 12,859 SNPs and 3,315 short indels, corresponding to one SNP per 2,133 nucleotides in the ~27.4 Mbp covered regions (Tables 4.2, Appendix B.9). In contrast, the three Gc strains showed an average of 61,512 SNPs and 6,878 short indels, corresponding to one SNP     140 per 446 nucleotides. The mean single nucleotide divergence between the Gs and Gc genomes was estimated at 1.66 (? 0.006), which was respectively ~7 and 11 times higher than mean intraspecific divergence for Gc (0.24 ? 0.002) and Gs (0.15 ? 0.0006).   4.3.4 Clustering and phylogenomic analysis of SNVs      We assessed genetic distance and phylogenetic relationships among Grosmannia genomes by the AWclust non-parametric clustering (Gao and Starmer 2008) and phylogenetic analyses of SNV data. The AWClust resolved Grosmannia genomes into four clusters corresponding to Gs and Gc lineages that each formed additional sub-clusters according to the geographic regions and host tree associates of the fungal taxa (Appendix B.14). The maximum parsimony (MP) and Bayesian phylogenetic trees supported the results from cluster analysis and showed identical tree topologies that only differed in the placement of the slkw1407 reference strain either within the Gs isolates from Alberta or those from Rocky Mountains. Here we only showed the MP (figure 4.1a), which was best described by a single unrooted tree with consistency index of 0.79, and 0.97 when including only Gs (GsRef, GsB3, GsC1) and Gc (GcC2 and GcB1) from distinct populations. The MP tree provided high statistical support (BS = 100% and PP = 1.0) for the positioning of Gs and Gc strains into two divergent clades, and for additional subclades within each clade. As expected, slkw1407 grouped within the P. contorta ? infesting Gs strains, which formed a distinct clade from the Gc strains. In the Gc clade, the G. clavigera holotype that had been isolated from MPB-infested P.     141 ponderosae was in a different subclade than the two P. jeffreyi associates. Within the Gs, strains from MPB-epidemic regions in BC, Alberta and Rocky Mountains were significantly separated from the two strains from the localized California pop ulation. This pattern was also consistent with the SNP density for the latter two genomes, which showed almost twice as many differences as the reference strain and the epidemic strains (Appendix B.9).  4.3.5  Ecological and physiological assessments   To support SNP phylogenetic relationships among the twelve Grosmannia genomes and to assess the host and distribution ranges of distinct lineages, we sequenced nine gene loci (Appendix B.5) in 16 additional strains from localized populations of MPBs and JPBs in their  respective host trees P. contorta, P. ponderosa and P. jeffreyi (Table 4.1). Genealogies from each of these genes (Appendix B.15), as well as, the concatenated phylogeny (figure 4.3a) confirmed the genome-wide SNV results noted above. The single-locus genealogies supported the monophyly of the entire Gc and Gs clades in eight of the nine gene trees (BS !70% and PP !0.95, Appendix B.15). Within Gc, seven gene trees separated the taxa associated with JPBs (n=10) in California from the MPB associates infestin g P. ponderosa trees in BC, California and South Dakota (n=6). The Gc ? P. ponderosa subclade was statistically supported in the concatenated phylogeny (figure 4.3a), as well as in one single-gene tree (CMQ6965 ? ABC.C, Appendix B.15). The phylogeny from conca tenated loci was     142  also consistent with geographic isolation within Gs, with five strains from the localized population in California forming a monophyletic clade separated from the epidemic strains; but no single-gene tree or the concatenated dataset showed statistical support for the California localized group. The concatenated matrix of nine gene sequences resulted in 14,308 aligned nucleotide positions, 296 variable sites and 143 informative characters (Appendix B.5). The nine -gene species tree showed identical topology based on ML, MP and Bayesian analyses with minor differences in the placement of terminal taxa (ML, figure 4.3a).   Consistent with results from the genome-wide SNP analyses and nine-gene phylogenies, we showed that while P. ponderosa and P. jeffreyi associates are genetically very close they can be characterized with distinct pattern of (+) -limonene utilization. Consistent with P. jeffreyi producing a lower level of limonene than P. contorta and P. ponderosa, we found that no Gc isolates from P. jeffreyi grew on (+) -limonene minimum media, in contrast to all Gc isolates from P. ponderosa, as well as to all Gs and the closely related species from P. contorta, which did grow  (figure 4.3b) .   4.3.6  Signature of positive and purifying selections in Grosmannia  To test for positive adaptive selection in Gs? Gc orthologs, we compared the ratio of SNPs within Gc with the sequence divergence between Gs and Gc at nonsynonymous and synonymous positions (figure 4.4a ? c). Under a neutral model of molecular     143 evolution, the ratio of polymorphism to divergence should be comparable between these two site classes. We used the neutrality index (NI) to quantify deviations from the neutral expectation in 3,476 Grosmannia orthologs. Assuming that synonymous positions are selectively neutral, positive selection should result in an excess of amino acid divergence (?log10NI>0) while purifying selection will show excess of nonsynonymous polymorphisms (?log10NI<0), indicating weakly deleterious variations segregating within species. For Gs?Gc orthologs, we obtained a median ?log10NI value of less than zero (?0.05; Appendices B.7, B.16), which suggested that the majority of genes (n=1,755) in our dataset are subject to weak purifying selection. We also detected a statistically significant (p < 10-05) signal of purifying selection in the pooled analysis of all 3,476 genes (?log10NITG = ?0.11, pooled PN = 3,834, DN = 5,903, PS = 3,267, DS = 6,612). But only five genes showed significant evidence of purifying selection on a per-gene basis. While 1,215 genes showed ?log10NI > 0, indicating fewer amino acid polymorphisms within Gs relative to those between Gs and Gc, we only found 11 genes with statistically significant (p ! 0.05) excess of protein divergence between the two species. Six of the 11 genes were among the 42 Grosmannia orthologs showing an excess of nonsynonymous fixed differences between Gs and Gc (i.e. the 1.2% of the 3,476 Gs?Gc polymorphic genes having DN " 9; Appendix B.7 and figures 4.4a?c). Among genes exhibiting the strongest evidence for positive selection (i.e. ?log10NI > 0), we noted polyketide synthases (PKS; CMQ_4392, _5323, _5095 and _2677), a non-ribosomal peptide synthase (NRPS; CMQ_3566), ABC transporters (CMQ_6634,_6965, _6960), oxidoreductases (CMQ_1999, _ 5949) and an heterokaryon incompatibility gene (CMQ_742) (Table 4.4 and Appendix B.7). However,     144 no genes were significant for either positive or purifying selection after correction for multiple testing (Benjamini and Hochberg 1995) .  For detecting positive selection in our dataset, we also applied codon-based models and likelihood estimates of dN, dS and dN/dS ( ! ) ratios. Divergence estimates were made from Gs-Gc pairwise alignments of the 3,476 orthologs using a codon substitution model that takes into account possible biases such as codon preference and nucleotide composition (Yang and Nielsen 2000) . We estimated mean dS 0.0032?8.9"10-19, corresponding to an average of one mutation per 312 synonymous sites between Gs and Gc since the common ancestor. This number was lower than the genome-wide average (one mutation per 446 nts, relative to the slkw1407 reference strain), presumably due to selective constraints in the coding regions. The mean pairwise dN was lower than dS (0.0011?4.8"10-5), reflecting the expected stronger constraints on substitutions that changed amino acids. The overall mean for dN/dS in the 3,476 orthologous genes (i.e. excluding 289 genes with dS equal to zero) was 0.3 ?0.005. This value was similar to the ? log10NITG = ? 0.11 value obtained for the MK test, suggesting that a large majority of genes are conserved and evolve with dN/dS less than one (Appendices B.7, B.16).   The pairwise dN/dS ratio is a measure of the overall evolutionary constraint averaged across the sequences of the gene and may  be too conservative for detecting positively selected sites along a gene. Thus we applied ?site -specific? models to test for     145 further evidence of positive selection within a more divergent subset of the 3,476 orthologous genes, removing 2,263 genes that had fewer than three fixed (n=1,567) and/or synonymous (n=696) differences. For the remaining 1,213 genes, the site-based approach identified 82 genes statistically significant for the positively selected sites (!>1; p ! 0.05). For the majority of these significant genes (n=46), the MK test also estimated a summary statistic of positive selection ?log10NI > 0, indicating an excess of protein divergence by both methods (Appendices B.7?8). The genes exhibiting the strongest evidence for positively selected sites include polyketide synthases (PKS; CMQ_5095, _2687, _2677), an ABC transporter (CMQ_6965), CYP450s (CMQ_3491, _ 6107, _4067), oxidoreductases (CMQ_277, _5685), ankyrin-repeat containing proteins (CMQ_1651, _ 569), a heat repeat protein (CMQ_7934) and an autophagy protein (CMQ_7167) (Table 4.4). The summary statistics on selection from MK and from PAML generally agreed (Appendix B.8). However, the two methods both identified significant signal for positive selection in only one gene (PKS_5095, Table 4.4). Another two PKS genes (CMQ_4392 and _2677) showed a significant or marginally significant signal for positive selection with both methods before correction for multiple tests. After correction for multiple tests, the signal was no longer significant (PAML P= 0.07 and MK P=0.09). The number of 46 significant genes (4%) in our dataset is lower than the conventionally accepted significance level of 5% because majority of genes are conserved and evolve with !  less than one. Nonetheless, after correction for multiple testing, we identified at least seven genes that evolved with !  greater than one (P<0.0001). This indicated that even though the level of divergence between the Gs and Gc was low, there is statistically significant evidence for site-specific positive selection between Grosmannia      146 species. Results for all the genes are available in Appendices B.7?8; Table 4.4 shows only the genes with the strongest evidence of positive selection using both PAML ?site-model? and the MK test.   4.4 Discussion  In this study, to identify features common across distinct Grosmannia populations and species, we compare the genomes of twelve Grosmannia clavigera sensu lato strains, representing two known sibling species that have different ecological characteristics (Massoumi Alamouti et al. 2011). We first used genome assemblies to assess changes in genomic structure such as rearrangements and gene gains/losses, and then focused on variation at the gene and nucleotide levels. We identified a number of functional variants in genes potentially involved in secondary metabolism and chemical detoxification, reflecting fungal adaptation to the specific chemistries of host trees. The data and results generated are a resource for assessing and characterizing fungal populations in the present MPB epidemic as it continues to spread into new habitats, including the P. banksiana  boreal forest as well as in future MPB outbreaks. The approach described here can also be applied to other insect-vectored/tree-colonizing fungi.        147 4.4.1 Grosmannia draft genomes   Using Illumina sequencing, we assembled draft genomes from eleven Grosmannia strains that represent distinct populations of the two known cryptic species: Gs and Gc. We showed that the de novo assemblies in these fungi could be mapped over a large fraction of the Grosmannia reference genome, suggesting that the majority of the assembled contigs, and the genes they contain, lie in regions that are collinear within and between the cryptic species. The extensive similarities in gene content (large-scale synteny) and order (colinearity) (Hane et al. 2011) within a large fraction of aligned contigs suggested that the morphologically cryptic Grosmannia species have diverged recently (Appendix B.13). This is consistent with the previous gene genealogies of Grosmannia clavigera sense lato and few other close relatives, which suggested that these pine-infesting, beetle-associated taxa have yet to reach a reciprocal monophyly for all the loci (Massoumi Alamouti et al. 2011). While teleomorphs (i.e. sexual structures) have been found rarely in Gs and not yet in Gc from P. jeffreyi, evidence of repeat-induced point mutation (RIP) in the reference genome and recombination in the population, as well as the extensive synteny reported here suggest that both fungi have a sexual phase (DiGuistini et al. 2011; Massoumi Alamouti et al. 2011; Tsui et al. 2012).  Large scale structural changes can exceed nucleotide evolution in plant pathogens like Mycosphaerella and Fusarium spp., which retain lineage-specific chromosomal islands or even entire lineage-specific chromosomes (Cuomo et al. 2007; Stukenbrock     148 et al. 2010; Klosterman et al. 2011). In filamentous ascomycetes such structural changes may be attributed to relatively long divergence times or horizontal gene transfer (Hane et al. 2007; Desjardins et al. 2011; Hane et al. 2011; Klosterman et al. 2011). Here major structural changes that would uniquely distinguish the cryptic Grosmannia species were not evident in our draft assemblies. Instead, the distinct ecological differences and host preferences in these fungi appear to be driven mainly by local nucleotide changes.   4.4.2  Grosmannia genome - wide SNVs   Single nucleotide variations (SNVs) are the most abundant type of genetic variation reported for eukaryotic genomes (Brumfield et al. 2003; Morin et al. 2004). Detecting genome-wide nucleotide variants within and between species using high-throughput sequencing technologies depends on two factors: a) whether the non-reference allele are present in the strains sequenced, and b) the number of high-quality and accurately-mapped reads that overlap the variant sites. The greater than 100,000 novel SNVs that we identified occurred in similar densities in intergenic, regulatory and coding regions across the eleven strains, and provide the first comprehensive assessment of genome-wide intra- and interspecific nucleotide variants for this group of beetle-vectored fungal symbionts. These SNV calls likely somewhat underestimate the total nucleotide differences within and between Grosmannia genomes, given that at least 10% of the     149 reference genome had less than 5x read coverage ? a limitation expected for Illumina sequencing of repetitive and GC-rich genomic regions (Li et al. 2008; Wang et al. 2011).  The genome-wide frequencies of nucleotide variants within Grosmannia species were lower than in other filamentous ascomycetes, including the plant pathogens Magnaporthe oryzae, Mycosphaerella graminicola, Sclerotinia sclerotiorum and different Verticillium and Cochliobolus species, as well as human pathogens in the genera Coccidioides and Paracoccidioides, and the generalist saprophyte Neurospora crassa and species in the genus Aspergillus (Lambreghts et al. 2009; Ma et al. 2010; Neafsey et al. 2010; Amselem et al. 2011; Andersen et al. 2011; Desjardins et al. 2011; Klosterman et al. 2011; McCluskey et al. 2011; Stukenbrock et al. 2011; Xue et al. 2012; Condon et al. 2013). In these fungal species whole-genome intraspecific SNV densities range from one per 865 nucleotides in the corn pathogen C. heterostrophus to one per 132 bases in the human pathogen P. brasiliensis. These numbers are higher than the Grosmannia intraspecific variant frequency of one per 2,133 nucleotides and often higher than nucleotide divergence between Grosmannia sister species (i.e. one per 446 bases). Our intraspecific SNV frequencies were comparable to those of Fusarium graminearum, a global pathogen of cereal crops (Cuomo et al. 2007). This pathogen is a sordariomycete like the ophiostomatoid fungi; it differs from other filamentous ascomycetes, including G. clavigera, because it is homothallic (i.e. self-fertile) and rarely out-crosses (Cuomo et al. 2007; Tsui et al. 2013). F. graminearum?s inbreeding may be associated with lower nucleotide diversity, as is the case in other fungal and O?mycetes genomes (Tyler et al. 2006; Cuomo et al. 2007). The frequency     150 of genome-wide SNVs in the opportunistic human pathogen A. fumigatus  is similar to that for Grosmannia , and is surprisingly low compared to its close relatives (Rydholm et al. 2006; Rokas et al. 2007). A. fumigatus ? low nucleotide variance and its lack of population structure globally have been explained by the worldwide spread of this fungus having occurred too recently for mutations to have accumulated within and between populations (Rydholm et al. 2006).  Differences in genome-wide frequency of SNVs among filamentous fungi may be in part due to differences in their life histories and dispersal processes. Ascomycetes comparative genomics have largely focused on saprotrophs that have broad host ranges and on pathogens that have the ability to survive for extended periods as free-living saprophytes without a specific host. Such fungi tend to have more stable population sizes and higher genetic variation in natural populations (Thompson 1994; Barrett et al. 2008). In contrast, fungal symbionts like Grosmannia  have limited and specific ecological niches (beetle vectors and host trees) and are more likely to experience local population outbreaks, crashes and re-colonization than generalist and saprophytic fungi (Thompson 1994; Six and Paine 1999; Carroll et al. 2006; Smith et al. 2010; Roe et al. 2011; Tsui et al. 2012). After such crashes, long periods of low endemic population sizes are expected for both the beetle and its associated fungi. Such cycles promote loss of genetic variance within populations and generate between-population genetic differences, through genetic drift and adaptive selection. Consistent with the above, our results show that while Grosmannia fungi have lower overall genome-wide frequencies of nucleotide variants than other filamentous fungi, their     151 SNVs support distinguishing two cryptic species and also suggest phylogenetically and biogeographically structured lineages that may include at least one additional species.  4.4.3  Grosmannia SNV- phylogenomics   Using genome-wide SNVs, we generated a high-resolution phylogeny that separated the twelve Grosmannia strains into two divergent monophyletic clades, confirming our previous gene genealogy discrimination of the Gs and Gc sister cryptic species (Massoumi Alamoutiet al. 2011). If the two species share extensive polymorphism through introgression, or incomplete lineage sorting due to a recent split from a common ancestor, we would expect that inter- and intraspecific nucleotide differences would be correlated (Avise 2004; Kulathinal et al. 2009). Here, no such correlation was evident; interspecific nucleotide divergence was significantly (p<0.01) higher than the mean intraspecific variation within Gc and Gs, suggesting that gene flow between Grosmannia cryptic species was weak or absent. This was consistent with the low level of homoplasy in our SNV-phylogeny (CI=0.97), and with our previous gene genealogies using population-level samples (Massoumi Alamoutiet al. 2011). The statistical support for each Grosmannia SNV-phylogenetic group indicates that we can detect lineage-specific variants, and so may be able to identify functional variants that are likely important to Gs and Gc adaptation to distinct ecological niches or to divergence of other phylogenetic groups resolved here.       152 Our SNV -phylogeny divided the epidemic Gs strains into well-supported phylogenetic groups that were also identified previously using AFLP and microsatellite markers (Lee et al. 2007; Tsui et al. 2012) . In addition, within Gs, we found a more divergent subclade, separating the strains from localized populations in California from the epidemic BC subpopulations. The av erage pairwise nucleotide divergence between California and epidemic phylogenetic groups were more than twice as large than divergences within and among epidemic groups, likely due to California location being distant, in the southernmost part of the species? range, along the Great Basin Desert (Wood 1982). While genetic structures within localized Gs populations have not been documented before, they have been reported for the MPB populations using AFLP markers (Mock et al. 2007). MPB populations in California were more divergent compared to those from other epidemic and most of the localized populations, consistent with our results on the fungal associate. This consistency reflects the co-evolutionary association between the beetle and the fungus, as suggested for other similar insect-fungal associations (Marin et al. 2009). MPB divergence based on AFLP makers was not significantly higher than expected for the isolation by distance, and it was suggested to correlate with the phylogenetic pattern of P. contorta  trees experiencing a northward expansion into British Columbia and the Northwest Territories since the last glaciation period (Marshall et al. 2002; Mock et al. 2007). For the fungal associate, whether or not the Gs-California lineage warrants recognition as a species would require sampling additional isolates from the localized populations infesting P. contorta  trees in the southern and eastern portion of the species? range, preferably using SNV makers optimized for this application (Morin et al. 2004). Our previous     153  network analysis on a 15 -gene concatenated dataset of the population-level samples from California and epidemic regions had shown incongrurence among gene genealogies, inferring the evidence of either incomplete lineage sorting or recombination. Either of these processes could be occurring in Gs populations. They may well have resulted from a recent species divergence maintaining high population size during the ongoing epidemics, a typical scenario in incomplete lineage sorting. Recombination is also likely and indicative of the potential lack of species structures within Gs when phylogenomic analyses are applied to population-level samples.  Within the Gc clade, our whole-genome SNV -phylogeny indicates host-specific differentiation in Grosmannia by separating the JPB associate from the holotype isolated from MPB-infested P. ponderosa (PP) in BC (Robinson-Jeffrey and Davidson 1968). Consistent with these results, the protein-coding combined phylogeny of additional Gc strains suggested that one lineage (Gc-JP) is exclusively associated with the JPB infesting the host tree P. jeffreyi in California whereas the other (Gc-PP) was only found on MPBs infesting P. ponderosa trees. The Gc from P. ponderosa host species in different geographic areas (i.e. BC, South Dakota and California) was genetically closer than those collected from different host species (P. jeffreyi) in the same geographic region in California. While our data from P. ponderosa trees were limited, preventing us from assessing the extent of host-specificity across the MPB-localized USA populations, or the role of geographical isolat ion in speciation, overall, our results suggest that speciation process in these fungi can be attributed to the host-    154 tree species and the geographic isolation of the host species from the current epidemics.   The genome-wide SNV divergence between the Gc -J P and Gc -PP was only twice as large as the intraspecific differences, reflecting the recent divergence of these lineages. A recently diverged population may represent an early stage in speciation, which begins when populations become genetically separated through geographical isolation or through ecological selection, and when adaptation acts as barrier to gene flow, and leads to genetically cohesive populations that are called species because they are ?segments of separately evolving lineages? (de Queiroz 2007) . The genealogical nondiscordance criterion (Dettman et al. 2003)  and the phylogeny of nine informative (i.e. genes randomly selected because of their potential fixed differences between the P. ponderosa and P. jeffreyi associates) protein-coding loci suggest that Gc-JP and Gc -PP are independent evolutionary lineages. The SNV -phylogeny and gene genealogies were further supported by our current ecological data showing that each lineage was associated with distinct beetle and tree host species. Further characterization of lineage-specific SNVs at a population level would strengthen evide nce for the work reported here, which suggests that lineages within Gc likely warrant recognition as genealogical and ecological species.       155 4.4.4 Grosmannia genes involved in host adaptation and ecological divergence   A combination of life -history traits and sel ection imposed by host trees may have promoted speciation and ecological divergence in Grosmannia lineages, as shown for many plant pathogenic fungi (Giraud et al. 2006; Stukenbrock and McDonald  2008; Giraud et al. 2010). In pine trees, phenolics and terpenoids from oleoresin are key constitutive and inducible chemical defenses (Boone et al. 2011).  While monoterpenes (e.g. !-phellandrene and limonene) and heptane (a straight-chain alkane) are toxic to many pathogens and insects, beetle -fungal complexes have evolved efficient mechanisms to survive and become established in such environments (DiGuistini et al. 2011; Wang et al. 2013).  For Gs, functional genomics and transcriptomic data suggest that ABC transporters, genes associated with oxidative stress responses and fatty acid !-oxidation pathways, and gene clusters that co ntain cytochrome P450s, dehydrogenases, and monooxygenases are involved in overcoming tree defenses (Hesse-Orce et al. 2010; DiGuistini et al. 2011; Wang et al. 2013) . However, chemical defense systems d iffer quantitatively and qualitatively between species of pine and between different populations within a pine species (Keeling and Bohlmann 2006; Gerson et al. 2009; Boone et al. 2011) . For example, P. jeffreyi has lower level of limonene and higher level of heptane than P. contorta (Mirov and Hasbrouck 1976; Paine and Hanlon 1994; Smith 2000) . Limonene is one of the most toxic defen se chemicals for bark beetle -fungal complexes (Raffa 2001; Raffa et al. 2005) ; it influences MPB -attack density in epidemic regions and it is found at high concentrations in P. ponderosa populations that have been subject to beetle -fungal outbreaks (Sturgeon     156 1979; Clark et al. 2010) . Given this, host preferences among Grosmannia lineages may reflect different abilities to survive and adapt to host chemicals or to other biotic and abiotic stresses inside the host.   Changes in gene contents and in gene products are central mechanisms in fungal genome evolution. Genes lost or in the process of  being lost through pseudogenization have been shown in plant pathogens (Stukenbrock et al. 2010; Marcet -Houben et al. 2012; Raffaele and Kamoun 2012; de Wit et al. 2012)  and in the closely -related human-pathogenic yeasts Candida albicans  and C. dubliniensis  (Moran et al. 2011). Similarly, 1.3% of the protein -coding genes in the Gs and Gc genomes contain premature stop codons, indicating that the genes may hav e been pseudogenized.  Twenty -two of these genes have oxidoreductase activity, including those with known roles in stress response and detoxification like short -chain dehydrogenases, cytochrome P450s and monooxygenases. Among those, twenty appear to have be en lost in both Gc? PP and Gc? JP, or in only one of these lineages. For example, a flavoprotein monooxygenase identified in the Gs gene cluster potentially involved in terpenoid detoxification and/or utilization has been pseudogenized in all the Gc -JP strai ns tested (n=10). Sequencing of additional Gc -PP and Gs strains and two related species confirmed that the stop codon is unique to the P. jeffreyi associates. Consistent with our genomic and sequencing results, all Grosmannia fungi including three species  from P. contorta  (Gs, L. longiclavatum and L. terebran tis) and Gc strains from P. ponderosa were able to grow on limonene as a sole carbon source, but none of the Gc strains from P. jeffreyi grew or survived in this condition. Combined, these results su ggest that a number of     157 genes with potential roles in Gs host-adaptation are inactivated or are evolving to become pseudogenes in Gc lineages. For example, because P. jeffreyi  produces lower levels of monoterpenes (including limonene) than pine species in epidemic and localized populations, the Gc -JP lineage may no longer require certain genes. The Gc lineages likely have more pseudogenes than we report here, because we have characterized only those caused by stop -codons, and not those due to indels and/or frameshift mutations.  We assessed both purifying and positive selection in Grosmannia protein-coding genes. Under the assumption that synonymous changes are neutral, purifying selection is inferred when the ratio of nonsynonymous to synonymous substitutions is less than 1, while positive selection pressure is usually inferred when the ratio is greater than 1 (Wright and Andolfatto 2008) . Our genome -wide characterization of Gs ? Gc protein-coding evolution showed that most genes evolve under purifying selection (dN/dS=0.3 ? 0.005). Similar results have been reported for other filamentous ascomycetes, reflecting overall evolutionary constraints on protein-coding genes (Gu et al. 2005; Nielsen et al. 2005; Rokas 2009; Sharpton et al. 2009; Stukenbrock et al. 2010; 2011). In contrast, among all the variable Grosmannia  genes, only 46 showed significant evidence for positive selection (i.e. before correction for multiple testing, p<0.05), which is not surprising given the close similarity between Gs and Gc orthologs (dS=0.0032). We note, however, that current divergence -based selection methods have limited statistical power for closely related species (Li et al. 2008; Oleksyk et al. 2010) , and consequently we may have missed some genes with weaker signs of selection. For     158  instance, sequence diversity and divergence in our data suggested that 1,215 candidate genes were showing some signs of adaptive selection, but the test was only significant for eleven genes (p <0.05).   The most significant examples of evidence for positive selection were the four PKSs, one NRPS and three ABC transporters. The PKSs and NRPS families are key enzymes for producing secondary metabolites, which are involved in fungal host -colonization and pathogenicity (Kroken et al. 2003; Collemare et al. 2008; de Wit et al. 2012). Based on our ABC domain and phylogenetic analyses (data not shown), the three membrane transporters are classified in the ABC-C subfamily, and so have potential roles in either host-chemical defenses or secondary metabolite export (Kovalchuk a nd Driessen 2010). Other genes with putative functions in chemical detoxification or utilization included an oxidoreductase, an isoflavone reductase and three cytochrome P450s (DiGuistini et al. 2011; Lah et al. 2013) . We also found that some of these genes had putative role in nutrient uptake (a ferric reductase and a mono-carboxylate permease). Other genes were potentially involved in cell signaling (e.g. histidine kinase, phospholipase), fungal development and growth (e.g. membrane copper amine oxidase, cell morphogenesis, autophagy protein, heat-repeat protein, hit finger domain protein); and a few with putative roles in protein -protein interactions or self/nonself recognition (e.g. two ankyrin repeat proteins and a heterokaryon incompatibility protein) (Luhtala 2004; Fedorova et al. 2005; Bahn et al. 2006; Kohler et al. 2006; Liu and Gelli 2008; Soanes et al. 20 08; Pollack et al. 2009) . In summary, for Grosmannia lineages that are adapted to different pine trees, our results suggest that     159 many of the genes that are evolving under positive selection are involved in secondary metabolite synthesis and secretion, host-chemical detoxification and stress responses, nutrient uptake from the host plants, and hyphal growth and differentiation.  In conclusion, we have used the Grosmannia genomes to show relationships between ecology and biological functions that are maintained or that diverge during colonization of a range of pine host trees, which are themselves adapting to changing environmental conditions. Although the fungal population has expanded and contracted repeatedly over at least several hundred years, large-scale synteny, with conserved gene content and order, suggests that these closely related strains adapt to different pine hosts largely through local nucleotide changes. This initial genome-wide SNV dataset is a phylogenetic resource that can be extended into a more comprehensive characterization of Grosmannia ecology and population structure.             160 4.5 Tables and figures  Table 4.1 Fungal strains used in this study  Fungal species Beetle associate Host tree Collection site (Map no. a) Source b Code c Grosmannia sp. (Gs) Dendroctonus ponderosae  Pinus contorta ( Pc ) Canada, BC, Kamloops (1)  (UAMH 11150) *  GsB1     BC, Houston (2)  UAMH 11153 GsB2     . (UAMH 11348) GsB3    Pc x P. banksiana  Canada, Alberta, Fox Creek (3)  (UAMH 11353) GsA1     . (UAMH 11354) GsA2    Pc  Alberta, Cypress Hills (4)  (UAMH 11347) GsA3     USA, Montana, Hidden Valley (5)  (UAMH 11156) GsM1     USA, California, Sierra Nevada (6)  (UAMH 11349) GsC1     . UAMH 11350 GsC2     . CB 67F21  GsC3     . UAM H 11361  GsC4     . UAMH 11362  GsC5  Grosmannia clavigera (Gc) D. ponderosae  Pinus  ponderosae ( P p ) BC, Cache Creek (7)  (ATCC 18086) Gc B1    U SA, South Dakota, Black Hills (8)  UAMH 11369  Gc S1     . (UAMH 11370)  Gc S2     . (UAMH 11371)  Gc S3     . (CB 15B29C2 ) Gc S4     . CB 23B110C5  Gc S5     . CB 32B85C10  Gc S6     USA, California, Sierra Nevada (6) (UAMH 11372)  Gc C11     California, Lassen (9)  UAMH 11373  Gc C12   D. jeffreyi  Pinus  jeffreyi (Pj)  California, San Bernardino (10)  (UAMH 11351) GcC1     . (UAMH 11352) GcC2     . DLS 1560  GcC3     . DLS 1595  GcC4     California, Lassen (9)  UAMH 11377  GcC6      161 Fungal species  Beetle associate  Host tree  Collection site (Map no. a)  Source b  Code c    . (DLS 210) GcC7     California, Sierra Nevada (6) (C 843)  GcC5     . (UAMH 11375) GcC8     . DLS 681 GcC9    Pj x Pp California, Lassen (9) UAMH 11374 GcC10  Leptographium terebrantis D. ponderosae  Pc BC  UAMH9722    Pc x P. banksiana  Canada, Saskatchewan  AU 123-113  L. longiclavatum D. ponderosae  Pc BC  CB LKG+T2B    D. ponderosae  Pc BC  UAMH 4876  a General map location of collection sites corresponding to Appendix B.12  b *, The fungal strain used for the reference genome published by DiGuistini et al. 2011; Strains selected for Illumina sequencing are bolded and fungal strains used for the physiological assessment are shown in parentheses; Source of isolates: U AMH, University of Alberta Microfungus Collection and Herbarium, Canada; ATCC, American Type Culture Collection, USA; Isolates beginning with CB, DLS, AU, and C are from culture collections of C. Breuil, University of British Columbia, Canada; D.L. Six, University of Montana, USA; A. Uzunovic, FPInnovations, Canada; and T.C. Harrington, Iowa State University, USA, respectively;   c Letters indicate the location and numbers indicate the number of isolates from each location.            162 Table 4.2  Summary of the genomic and gene coverage data in the eleven sequenced genomes   IDs*  Covered genomic bases Genomic coverage Covered gene bases Gene coverage Unmapped reads  GsB1 (control) 28,791,583 (98.8%)  70 ! 15,507,591 (99. 9%)  47 ! 0.7%  GsB2 27,370,438 (94.0%)  27 ! 15,333,848 (98.8%)  30! 1.2%  GsB3 28,464,740 (97.7%)  48! 15,316,500 (98.7%)  48! 2.6%  GsA1 28,488,061 (97.8%)  58 ! 15,334,490 (98.8%)  61! 3.2%  GsA2 26,491,256 (91.0%)  35 ! 13,533,085 (87.2%)  78 ! 5.0%  GsA3 27,146,56 9 (93.2%)  24! 15,203,277 (98.0%)  29! 0.8%  GsM1 27,197,413 (93.4%)  22! 15,170,992 (97.8%)  33! 1.3%  GsC1 28,153,881 (96.6%)  50 ! 15,092,814 (97.3%)  57 ! 4.1%  GsC2 28,084,478 (96.4%)  49! 15,031,784 (96.9%)  63! 4.0%  GcB1.a 26,267,089 (90.1%)  14! 13,636,725 ( 87.9%)  20! 6.8%  GcB1.b 28,360,091 (97.4%)  55 ! 15,399,793 (99.3%)  50 ! 10.1 %  GcB1.ab 28,455,969 (97.7%)  79 ! 15,435,487 (99.5%)  77 ! 9.2%  GcC1 27,001,022 (92.7%)  25 ! 15,204,933 (98.0%)  29! 2.4%  GcC2.a 26,377,060 (90.5%)  36! 13,414,391 (86.5%)  54 ! 13.4 %  GcC2.b 27,130,509 (93.2%)  42! 14,137,799 (91.1%)  56 ! 12.1%  GcC2.ab 27,940,266 (95.9%)  72 ! 14,868,092 (95.8%)  87 ! 13.6 %  Average 27,425,586 (94.1%)  37 ! 14,701,029 (94.7%)  47 ! 5.1%  *IDs, "a" and "b" are results from two independent sequence lanes for the  same strain, "ab? results from two sequence runs combined for the same strain. The estimated coverage are based on filtered reads mapped to the slkw1407 reference genome sequence, which is ~ 29.1 Mbp after excluding gaps.           1 63  Table 4.3 Genome-wide characterization of fixed and shared polymorphisms between Gs and Gc lineages   Genomic Regions Fixed a Shared b Exclusive c to Gs (parsimony-informative) Exclusive to Gc (parsimony-informative) Total (parsimony-informative) Dxy (? SD) Total 37,712  3,750  35,765 (18,871)  26,203 (9,685)  103,430 (70,018)  1.66 (?0.006)  Intergenic  10,818  2,458  14,811  14,793  42,880  nc Flanking regions  11,148  559  8,308  4,574  24,589   Intronic  2,755  112  2,001  958  5,826   Synonymous 6,808  353  4,984  2,744 14,889   Nonsynonymous  6,116  264  5,601  3,059  15,040   stop gain ?  stop lost 73 ?  18  4 ?  1  80 ?  14  69 ?  3 226 ?  36   a Fixed polymorphisms are nucleotide sites, at which all Gs strains differ from all strains of Gc.   b Shared polymorphisms are sites for which multiple nucleotides are found in both Gs and Gc strains.   c Exclusive polymorphisms are those that are polymorphic in one species and invariant in the other.  nc, not calculated.            164  Table 4.4 The top 42 genes showing evidence of positive selection a  MK b  PAML c Genes Gene description PN DN PS DS p-value  M1a?M2a M7?M8 p-value dN/dS (%) CMQ_6634  ABC transporter 1 10 3  1 *   ns ns ns ns  CMQ_1999  Ferric reductase 0 5  2 0 *   . . . .  CMQ_5949  Putative oxidoreductase 0 5  3  1 *   . . . .  CMQ_742  Heterokaryon incompatibility  1 9 3  1 *   . . . .  CMQ_3665  Thermotolerance protein 1 4  4  0 *   . . . .  CMQ_3566  NRPS  0 21 4  13  *   . . . .  CMQ_5323  Polyketide synthase 1 17 4  5  *   . . . .  CMQ_5095 . 1 20 4 9 *  21 26 ** 5 5  (1.3)  CMQ_4392 . 0 29 6 8 * *  5 5 p=0.07 5.3  (na) CMQ_2677 . 3 8 2 5 p=0.09  26 26 ** 157  (0.18) CMQ_2687 . 3 7 2 2 ns  30 31 ** 577  (0.24)  CMQ_1651 Ankyrin repeat protein 1 27 0 9 .  27 23 ** 3 5  (4.70)  CMQ_569  . 1 49  3  31  .  12 13  **  6 (12.5)  CMQ_5699 Peroxin  1 5 1 2 .  35 37 ** 999 (0.08) CMQ_7934 Heat repeat protein 2 4 3 2 .  18 19 ** 306 (0.20) CMQ_8021  Heat shock transcription factor 1 3  4  0 .  8 8 *  158  (0.60) CMQ_6965  ABC transporter 1 10 3  3  .  8 8 * *  164  (0.60) CMQ_1224  Membrane copper amine oxidase 0 4  2 2 .  7 7 *  75  (0.70) CMQ_277 Isoflavone reductase 1 3 4 2 .  32 32 ** 58 (0.70) CMQ_5685  2-dehydropantoate 2-reductase 1 1 3  2 .  10 10 **  453  (0.06) CMQ_6107  Cytochrome p450 monooxygenase  0 5  1 2 .  10 16 **  67 (1.15)  CMQ_3491  . 1 4  1 2 .  7 7 *  61 (1.13)  CMQ_4067  . 0 5  1 4  .  7 7 *  230  (0.06) CMQ_1868  C2h2 finger domain containing protein 1 3  2 2 .  7 7 *  81 (0.48)  CMQ_938  GTP cyclohydrolase 1 2 2 3  .  7 7 *  80 (0.40)  CMQ_5377  Glycoside hydrolase 3  4  2 1 .  7 7 *  113  (0.60) CMQ_1846  C-terminal hydrolase 1 2 2 3  .  7 7 *  252  (0.18) CMQ_4864  Cell morphogenesis protein 1 7 3  4  .  7 7 *  73  (0.70) CMQ_7167  Autophagy protein 1 6 1 3  .  8 8 *  152  (0.60) CMQ_555  Death -box RNA helicase  2 4  2 1 .  8 8 *  147  (0.97) CMQ_3835  Dash complex subunit  0 5  1 2 .  7 7 *  74  (0.80) CMQ_6415  tRNA -guanine transglycosylase 1 2 2 2 .  12 12 **  935  (0.03)  CMQ_906  Monocarboxylate permease 1 5  1 2 .  7 6 *  72 (2.20)     165 MK b  PAML c Genes Gene description PN DN PS DS p-value  M1a?M2a M7?M8 p-value dN/dS (%) CMQ_4470 Sensor histidine kinase response 0 8 1 9 .  7 7 . 821 (0.01) CMQ_2502 Phospholipase  0 2 1 2 .  11 11 ** 884 (0.03) CMQ_3494 Ribosome biogenesis  0 4 1 1 .  9 9 . 393 (0.4) CMQ_3485 mtDNA inheritance protein 1 4 1 3 .  8 8 * 156 (0.33) CMQ_1204 Inositol polyphosphate phosphatase 2 2 3 2 .  8 8 . 148 (0.56) CMQ_3564 Dihydrodipicolinate synthetase 0 3 1 1 .  17 18 ** 889 (0.12) CMQ_4773 ATP-dependent DNA helicase 2 2 3 2 .  7 7 * 170 (0.20) CMQ_599 ATP-binding endoribonuclease 3 2 4 2 .  14 14 ** 103 (0.50) CMQ_3121 60S ribosomal protein 1 6 1 3 .  7 7 * 82 (0.80) a List of genes that showed the strongest evidence of positive selection using MK test and PAML ?site-model?; Genes that were found significant for positive selection by both methods and those that after correction for multiple testing remained significant are highlighted; Results are only shown for genes with a putative function; NS, not significant; NA, not applicable;  b PN, number of nonsynonymous polymorphisms within Gs; DN, number of nonsynonymous differences between Gs and Gc; PS, number of synonymous polymorphisms within Gs; DS, number of synonymous differences between Gs and Gc;  P-value, significant excess of nonsynonymous divergence in MK test using the Fisher exact test (P<0.05* or P<0.01**).   c M1a-M2a and M7-M8, twice the difference in the natural logs of the likelihoods of the two models being compared. This value is used in a likelihood ratio test along with 2 degrees of freedom; p-value, is an uncorrected value from !2 distribution to indicates the confidence with, which the null model can be rejected; dN/dS, nonsynonymous /synonymous substitution ratio (!  = dN/dS) under M8 model of variable !  ratios among sites, and the percent of codons placed in that class. Amino acid positions identified in the class of codons evolving under positive selection in M8 (posterior probability>0.90) are listed in Appendix B.8.          166         167  Figure 4.1 Grosmannia SNP-phylogenomics, gene content and amino acid similarity  a) Maximum parsimony analysis of 103,430 SNPs among 12 Grosmannia genomes. All branches have 100% bootstrap support and posterior probabilities of 1.0. The scale bar indicates the number of SNPs along each branch. A, B, C and M are the collection sites. GsB1 is the reference genome. Pc, Pj and Pp are the host tree species (Table 4.1). The grey box highlights Gs strains from epidemic regions. b) Genome -wide pairwise amino acid identity between 8,312 Grosmannia reference gene models and homologous proteins in the 11 other strains. Homologous proteins with high amino acid identity are likely orthologs.     168         169  Figure 4.2 Intra-/interspecific variants in the terpenoid-processing gene cluster   a) Alignment of 12 homologous supercontigs shows complete synteny and colinearity for all Grosmannia strains. GsB1 :  Grosmannia  reference genome. Locally collinear blocks (LCB, shown in the same color) have similar sizes among strains, and the 33 gene models that are potentially involved in limonene utilization or detoxification are in complete synteny. The apparent indel (purple LCB) is located in an intergenic region that may have been subjected to assembly error for a few Gs and Gc strains . To rule out the possibility of assembly error, a contig longer than the indel (puple LCB) is required, but these data were missing for all the strains showing the deletion. b) The 33 orthologous genes in the terpenoid-processing cluster showed relatively low numbers of polymorphisms (non-synonymous, PN ; synonymous: PS) and divergence (non-synonymous, D N ; synonymous, D S). Dots represent genes with less than two coding nucleotide differences. c) The flavoprotein monooxygenase (CMQ -6740) has a stop codon in its second exon that is unique to the Gc isolates from P. j efferyi.             170         171 Figure 4 .3 Grosmannia species phylogeny correlated with the host - tree species of different phylogenetic lineages and fungal lineage tolerance to a related host -defence chemical   a) Maximum likelihood (ML) analysis of a concatenated nine-gene dataset subdivides Gs and Gc strains into three well-supported clades according to host tree species. Thick branches indicate nodes with 100% support from ML, maximum parsimony and Bayesian analyses. The grey box highlights Gs strains from epidemic regions.  Arrows indicate total numbers of fixed differences between Gs-Gc, Gc?P. ponderosa and Gc?P. jeffreyi lineages. The tree is rooted with the outgroup taxa L. longiclavatum and L. terebrantis. Dashed line indicates an adjustment of scale. b) Physiological assessment comparing (+)-limonene utilization as a carbon source in Gs and Gc lineages and close relatives. The growth of the fungal lineages from P. contorta and P. ponderosa indicates their ability to tolerate and utilize this toxic chemical.              172   Figure 4.4 Comparison of divergence and polymorphism  a) Number of synonymous and non - synonymous Gs ? Gc fixed differences (divergences) in 7,340 orthologous gene models. b) Number of synonymous and non - synonymous SNPs within Gs strains in 7,340 orthologous gene models. Circular bands with alternating shades of grey represent 36 scaffolds in which the gene models are located. Grey triangles mark genes with t he largest numbers of non- synonymous divergences (e.g. PKS, ANK, ABC in a) and polymorphisms (e.g. ANK in b). c) Summary distributions of McDonald - Kreitman (MK) cell entries for the fixed differences and SNPs in 3,476 variable genes (Appendix B.7).      173  Chapter 5 Conclusions  This thesis highlights the importance of studying genomes in ophiostomatoid fungi, particularly for species of the genus Grosmannia , which include the most common associates of bark beetles and important tree pathogens like G. clavigera. Until now phylogenomic data in the form of sequences of many individual loci or fully sequenced genomes had not been used for establishing relationship among species, defining species boundaries, or assessing host adaptation and ecological divergence in the beetle- tree associated fungi. Here, we show that genome data are particularly relevant for assessing the systematics of this fungal group, which remains problematic at both interspecific and generic levels. Further, these large datasets can generate insights into the ecological and biological attributes of these organisms and evolutionary patterns that affect species diversity, and relationships of species with their vectors and host trees.  Sequenced genomes support comparing patterns of nucleotide substitutions for  orthologous regions within and between closely related species, and so can provide evidence for evolutionary forces and the processes that influence the evolution of fungal pathogens. When nearly complete genome sequences are available for many species, these comparisons are powerful. Until recently, genomic resources for the beetle - tree associated fungi were available only for G. clavigera. When we made the reference genome of this species available, one of the many research opportunities created was     174 to characterize evolutionary divergence between Grosmannia  strains. For this, we generated draft genomes for eleven additional G. clavigera strains from distinct populations including different beetle vectors and host trees. These genomes provided a unique opportunity for comparative analysis to establish species and population boundaries, and to reveal the potential role of host-tree adaptation and population dynamics in fungal evolution and divergence. Our analyses focus on generating a comprehensive assessment of genome-wide single nucleotide variations within and between the newly recognized species, and using this information to identify specific genes that are potentially involved in species divergence and host adaptation. Note that the work reported here, which demonstrates an approach for understanding the effects of genome divergence, should be extended to populations over wider geographic regions.   While sequence data for such work is now practical to generate, there is a need to adapt or develop analysis tools to such genomic data in order that they can be used efficiently and reliably to identify species diversity, population structures and processes affecting the evolution of the tree pathogens in their natural ecosystem. The work describe here is an important step toward this objective. We developed new genomic resources that provide a foundation for future development of phylogenomic and lineage-specific markers to assess species diversity and relationships within the ophiostomatoid fungi. As well, we identified functional variants that are potentially involved in host colonization and progressive divergence within and between the Grosmannia species. The resources that we created provide basic information for future     175  research on comparative and functional genomics of other species of this important group of fungi, and for related ophiostomatoid genera. A longer -term program that extends the demonstration reported here could be part of a multidisciplinary approach towards developing and implementing sustainable forestry management practices that reduce the intensity of beetle-fungal outbreaks.  5.1 Ophiostomatoids systematics limitation and status of the genus Grosmannia   Taxonomy has been a dynamic and progressive discipline for many fungal groups (Hawksworth et al. 2013) ; however, for beetle -tree associated fungi (ophiostomatoids), it has needed further development. In chapter 2, I used morphological characterization and a multigene-phylogenetic approach to evaluate the ophiostomatoid systematics by defining genera and species boundaries, and I revised the status of some Op hiostomatales genera, including Grosmannia. The results demonstrate the limitations of using classical morphological traits and molecular analyses based on single genes to address the taxonomy of this fungal group. The sexual form is less frequently  found for ophiostomatoids and thus many taxa are primarily classified based on the asexual or anamorph structures (Upadhyay 1981; Wingfield and Seifert 1993) . I show that no anamorph can unambiguously support the phylogenetic groups resolved by our multigene datasets. Molecular systematics based on multiple informative loci integrated with a more natural taxonomic scheme separate species of the genera  Grosmannia ,      176 Ambrosiella and Raffaelea into different clades corresponding to their distinct ecological niches and vector associates, i.e. bark- versus wood-boring beetles.   For this work I generated a multigene dataset for sixty-seven taxa that represent a diverse set of ophiostomatoid teleomorphs and anamorph genera. The results present the phylogenetic relationships of ophiostomatoids at a higher taxonomic level, including different genera (i.e. Ceratocystiopsis, Grosmannia and Ophiostoma) that are associated with the Dendroctonus ponderosae (MPB)-fungal outbreak in western North America (Lee et al. 2006a). The high resolution of our multigene phylogeny also confirms the monophyletic status of the genus Grosmannia, which was my primary question, since species of this genus were previously mixed with the ecological group of Ophiostomatales, which was collectively called ambrosia fungi. Finally, I show that ambrosia fungi, including the genera Ambrosiella and Raffaelea, are polyphyletic, forming at least six distinct phylogenetic clades, each of which I propose should be reassigned to a new genus. These results will influence future research on the Ophiostomatales? taxonomy, especially for the novel clades that need additional support with an expanded collection of fungal taxa and gene loci.  5.2  Defining species boundaries in Grosmannia clavigera  Cryptic species are increasingly reported for plant pathogens. The results in chapter 3 provide additional examples of cryptic species within G. clavigera, the well-known     177 fungal pathogen of P. contorta forests in western North America and the symbiont of two sister bark beetles: MPB and JPB. We found that genetic relatedness in Grosmannia lineages mainly corresponds with the host tree species, and that host specificity and/or preferences are more evident in the localized populations than in epidemic regions.   5.2.1 Polymorphism discovery and species recognition   Cataloging species diversity is a first step toward dev eloping an understanding of how various organisms interact with their environment, which is a key factor in establishing organisms' roles in the ecosystem. Despite previous systematics efforts, defining species in the G. clavigera complex was a challenge and required developing new molecular resources. I began this work by characterizing nucleotide variations in 67 gene loci and then sequencing 15 loci across 53 population samples. Using ?phylogenetic species recognition by genealogical concordance? and dif ferent population genetic analyses of this dataset, I identified two cryptic sibling species within the pathogen: Gc and Gs. The data suggested a potential history of recombination within both species, which in addition with our ecological data (i.e. occasional observance of sexual stage inside the older galleries of the beetle associates) allowed for names according to the teleomorphs genus. I retained the nomenclatural name G. clavigera  for the lineage (Gc) that is genetically and ecologically represented by the holotype (Robinson-Jeffrey and Davidson 1968) , while Grosmannia  species (Gs) is a     178  newly recognized species that remains to be described. This work also generated protein-coding sequences that were used for developing target-specific PCR -primers for the Grosmannia species and their close relatives (Khadempour et al. 2010).  5.2.2  Ecologically distinguishable lineags   Sequencing one of the informative loci across 166 isolates indicated that Gc is more specific to the closely related tree species P. jeffreyi and P. ponderosa in the USA, where these pines are infested by localized populations of respective beetle associates. In contrast, Gs is an exclusive associate of MPB and its primary host -tree P. contorta; however, in the current epidemic areas, it is also found in other pine species. This is an important finding because it suggests that both host tree species and beetle population dynamics are important factors in the evolution and divergence of these fungi. In regions with localized population, although the beetle-fungal population level is low and difficult to sample (Smith et al. 2010), we show that further investigation from different host trees (P. contorta, P. ponderosa and P. flexilis) in eastern and southern portion of the MPB range i.e., in areas that have not been reached by the current epidemics, is necessary for understanding the biological and ecological interaction between beetle-fungal complexes and the host trees. These interactions are not evident in the epidemic regions since increase in the beetle-fungal population levels has enabled these organisms to spread from their native range to other suitable habitat and to establish in new host trees (e.g. P. banksiana).     179 5.3 Population genomics in Grosmannia   While sequencing one genome per species allows insights into an organism's biology, inferring genetic variations that lead to adaptation in different environments, and to biological/physiological differences between populations and species requires more than one genome sequence for a species. The most comprehensive view on such variations is not gained by sequencing a few candidate genes, but by genome-wide studies of variants within and between closely related species, i.e. population genomics. This has become routine in non-model organisms with the advent of NGS systems like those from Illumina. In chapter 4, we use these technical advances to generate additional genomic resources from Grosmannia strains representing the two newly recognized species: Gs and Gc. Despite their close phylogenetic relationship these fungi inhabit distinct ecological niches, and I believe that adaptation to the specific chemistries of host tree is an important feature in their evolutionary divergence. My analysis of the new genomic resources indicates that this hypothesis likely holds true, or at least warrants further investigation for a set of genes potentially involved in host-colonization and/or pathogenicity.    5.3.1 Grosmannia draft genomes and genome-wide characterization of SNVs   Because Illumina systems generate deep sequencing data, the genomes for each of the eleven Grosmannia strains provided high coverage for much of the reference     180  genome, allowing for a comprehensive characterization of genome -wide sequence variation in these strains, which were isolated from a large geographic area and distinct ecological niches. In contrast to other eukaryotes, most filamentous fungi have compact genomes with relatively few repetitive sequences; this makes generating genome assemblies from short sequence reads affordable and technically less challenging than for higher eukaryotes (Nowrousian 2010). While finished assemblies tend to be essential for thoroughly characterizing lineage -specific differences, and for detecting recently evolved or rapidly evolving parts of genomes, draft genomes support identifying major structural changes and local change s like SNVs and short indels.   Large scale structural changes have been shown to exceed nucleotide evolution in plant pathogens like M ycosphaerella  and Fusarium spp., which have lineage-specific chromosomal islands or even entire lineage-specific chromosomes (Cuomo et al. 2007; Stukenbrock et al. 2010; Klosterman et al. 2011) . Such structural changes may be attributed to relatively long divergence times or horizontal gene transfer (Hane et al. 2011). In this work; however, draft assemblies showed no evidence of major structural changes in the genomes of the eleven Grosmannia  isolates, suggesting that these fungi have diverged relatively recently, consistent with the gene genealogies that we describe in the third chapter of this thesis. We found more than 100,000 SNVs, suggesting that distinct ecological differences and host preferences in these fungi may be driven mainly by local nucleotide changes, rather than by large-scale rearrangements.       181 5.3.2  Genome - wide SNV - phylogeny   The genome-wide phylogeny that I derived from the SNV dataset confirmed chapter three?s genealogical study of G. clavigera populations, which separate the twelve strains into divergent Gs and Gc monophyletic clades; and it further divides each clade into potentially distinct phylogenetic groups that had not been previously reported. Within  Gs, I suggested that strains from epidemic regions and localized populations might represent two evolutionary independent lineages that before applying the species criteria require further sampling in southern and eastern portions of the species? range and preferably using new molecular markers from our SNV dataset. Further, within Gc, we found additional evidence of host preference and/or specificity in the localized populations showing that JPB associates (JP) form a separate clade from the holotype isolated from MPB-infested P. ponderosa (PP) in BC (Robinson-Jeffrey and Davidson 1968). I examined the monophyly of Gc phylogenetic groups by genealogies of additional P. ponderosa associates collected from California (n=2), South Dakota (n=3) and the holotyp e as the only remainder from previous epidemics in BC. The ?genealogical nondiscordance criterion? (Dettman et al. 2003) suggested that Gc-JP and Gc-PP may warrant recognition as genealogical and ecological species. Further characterization of P. ponderosa associates at the population level using informative SNV markers should clarify the extent of host -specificity across the MPB -localized USA populations and/or the role of geographical isolation in their divergence.       182 Our network analysis in figure 3.3a  of chapter 3 may seem to disagree with the SNV subdivisions within Gs and Gc . While the network analysis allows visualizing conflicts among the 15 gene datasets, it cannot distinguish between the causes of conflicts, which may be recombination or incomplete lineage sorting (i.e. retention of ancestral polymorphisms through multiple speciation events). Further, none of the 15 gene trees were informative in resolving either Gs-California or Gc- P. ponderosa  groups, except for a peptidase locus that separated Gc strains from P. ponderosa , but with no statistical support. However, for recently diverged species the lack of diagnosable characters at some loci and a reticulate pattern of incomplete lineage sorting are not unexpected. Neutral coalescent theory indicates that the expected time to monophyly is often long even for a single nuclear gene, suggesting that many species exist, but have not yet achieve monophyly even for a few genes (Hudson and Coyne 2002; Knowles and Carstens 2007; Shaffer and Thomson 2007) . Nevertheless, potential signs of recombination, particularly within Gs, can provide important clues for avoiding over-estimating species diversity in these fungi. Additional analyses at the population level should be carried out using the informative SNP markers reported here.  5.3.3  Functional characterization of SNVs and their adaptive contribution   Nucleotide substitutions that generate adaptive variations can occur in both regulatory and coding sequences. In this thesis, while our analysis focuses on protein -coding genes, we also identified intra- and interspecific SNVs in potential intergenic and     183 regulatory regions. Both the coding and noncoding variants that I report should be good starting points for future studies on the functional genomics of the Grosmannia species and on the role of positive selection in the evolution of gene regulatory elements. For instance, SNVs characterized in regions upstream and downstream of Grosmannia gene models may cause adaptive variations, especially if they cause substitutions in regulatory sequences like promoters and terminators, and so they may activate or suppress gene expression. However, because the annotation of regulatory regions in Grosmannia genomes is currently putative, we only present the results for adaptive variants in coding regions, which provide a rich resource for future investigations of adaptive evolution and functional variation in the pathogens.  Most studies that seek to detect adaptive evolution in fungal pathogens have focused on specific likely candidate genes. Recently, however, genomic data has allowed ?reverse ecology? studies that search for genes under posit ive selection without a priori expectations (Li et al. 2008; Ellison et al. 2011; Gladieux et al. 2013) . Here, we use this approach to carry out a genome-wide assessment of polymorphism and divergence in 12 Gs ?Gc gene orthologs. Only 1.2% of 3,476 informative genes showed rapid amino acid evolution (i.e. non-synonymous fixed differences between Gs and Gc). Considering these genes, and other variable coding regions, we found evidence of positive selection in at least 46 genes ( p-value<0.05). Some of these genes were involved in secondary metabolite synthesis and secretion (PKS). Others had putative roles in exporting host-chemical defense chemicals or secondary metabolites (ABC?C     184 transporters), and in protein-protein interactions or self/nonself recognition (ankyrin repeat proteins and a heterokaryon incompatibility protein).   The between-species selection methods applied here (i.e. MK and dN/dS tests) showed limited statistical power for inferring positive selection between Grosmannia siblings. Generally, these methods can be used to identify non-recent selective forces, but they require a large number of nucleotide variants to exceed the background of mutational drift over long periods of species differentiation (Li et al. 2008; Oleksyk et al. 2010). Despite low divergence level between Gs and Gc lineages, we find statistically significant evidence of site-specific positive selection on a small set of genes, including those with a putative function in host-colonization and pathogenicity  (e.g. PKSs).   Characterizing the coding variants, we also identified a number of truncated or potential pseudogenes within Grosmannia  gene models. The majority of these genes (n=22) have potential oxidoreductase activity, including those with known roles in stress response and detoxification like cytochrome P450s, short-chain dehydrogenases and monooxygenases. These results are consistent with our physiological assessment, and suggest that the stop-codon variants may also reflect adaptation of Gs and Gc lineages to the specific chemistries of Pinus  contorta, ponderosa, and jeffreyi  trees. Both the potential Grosmannia  pseudogenes and the genes showing evidence of positive selection (e.g. PKS and ankyrin repeat proteins) are important candidates for being     185 functionally important in divergence and/or ecological adaptation of Grosmannia fungi. These warrant further validation and investigation.   5.4  Perspective on future work   This thesis provides a foundation for future work on the taxonomy and systematics of the Ophiostomatales . Different phylogenetic groups were redefined and newly suggested genera and novel species should be incorporated in future studies, to expand the collection of fungi isolated from different ecological and geographical sources, and to characterize their systematics using the new genome datasets. Genome-wide scans for positively selected genes that provide insight into adaptive evolution in Grosmannia lineages, the genetic basis of differences between and within species, and putative functions of some genes should be expanded. For example, the potential pseudogenes in P. jeffreyi associates or PKS genes with positive selection signatures might be important in host chemical detoxification and/or host colonization, and can be the focus of future comparative studies or experimental functional characterization. Adding other closely related species such as L. terebrantis and L. longiclavatum would increase the phylogenetic depth of the genome datasets, which can improve statistical power of the selection analyses and may permit new lineage- and clade-specific questions to be tested.       186 Very recently, in Alberta, the MPB-fungal complexes have succeeded in colonizing a new host-tree species, P. banksiana , adapting to the new chemical and physical environment present in this host. While P. banksi ana is more closely related to P. contorta than to P. ponderosa or P. jeffreyi , environmental conditions prevailing in the northern boreal forest could allow the symbiotic partners to continue to adapt and to spread eastward, with potentially important ecological and economic consequences. To infer recent selection pressures on the beetle-pathogen complexes as they exploit new hosts requires polymorphism data and population genomic-based analysis? the excess of non-synonymous to synonymous differences would not be great enough to be statistically significant using divergence-based selection analysis. The genomic datasets and informative markers for landscape population genomics provide a comprehensive resource for work on ongoing adaptation in this vector-pathogen complex, and the approach reported here could be extended to other beetle-tree associates.             187 References  Agapow P-M, Burt A. 2001. Indices of multilocus linkage disequilibrium. Mol. Ecol. Notes 1:101?102. Aguileta G, Lengelle J, Marthey S, et al. 2009. Finding candidate genes under positive selection in Non-model species: examples of genes involved in host specialization in pathogens. Mol. Ecol. 19:292?306. Alfaro RI, Humble LM, Gonzalez P, Villaverde R, Allegro G. 2007. The threat of the ambrosia beetle Megaplatypus mutatus (Chapuis) (=Platypus mutatus Chapuis) to world poplar resources. Forestry 80:471?479. Amselem J, Cuomo CA, van Kan JAL, et al. 2011. Genomic Analysis of the Necrotrophic Fungal Pathogens Sclerotinia sclerotiorum and Botrytis cinerea. PLoS Genet 7:e1002230. Andersen MR, Salazar MP, Schaap PJ, et al. 2011. Comparative genomics of citric-acid-producing Aspergillus niger ATCC 1015 versus enzyme-producing CBS 513.88. Genome Res. 21:885?897. Anderson PK, Cunningham AA, Patel NG, Morales FJ, Epstein PR, Daszak P. 2004. Emerging infectious diseases of plants: pathogen pollution, climate change and agrotechnology drivers. Trends Ecol. Evol. 19:535?544. Von Arx JA, Hennebert GL. 1965. Deux champignons ambrosia. Mycopathol. Mycol. Appl. 25:309?315. Avise JC, Ball RM. 1990. Principles of genealogical concordance in species concepts and biological taxonomy. In: Futuyma D, Antonovics J, editors. Oxford Surveys in Evolutionary Biology. Vol. 7. Oxford: Oxford University Press. p. 45?67. Avise JC. 2004. Molecular markers, natural history, and evolution. 2nd Revised edition. Sunderland, MA: Sinauer Associates Ayres MP, Lombardero MJ. 2000. Assessing the consequences of global change for forest disturbance from herbivores and pathogens. Sci. Total Environ. 262:263?286. Bahn Y-S, Kojima K, Cox GM, Heitman J. 2006. A Unique Fungal Two-Component System Regulates Stress Responses, Drug Sensitivity, Sexual Development, and Virulence of Cryptococcus neoformans. Mol. Biol. Cell 17:3122?3135.     188 Barrett LG, Thrall PH, Burdon JJ, Linde CC. 2008. Life history determines genetic structure and evolutionary potential of host-parasite interactions. Trends Ecol. Evol. 23:678?685. Batra L. 1963. Ecology of Ambrosia Fungi and Their Dissemination by Beetles. Trans. Kans. Acad. Sci. 66:213?236. Batra L. 1967. Ambrosia Fungi: A Taxonomic Revision, and Nutritional Studies of Some Species. Mycologia 59:976?1017. Batra LR. 1966. Ambrosia Fungi: Extent of Specificity to Ambrosia Beetles. Science 153:193?195. Batra LR. 1972. Ectosymbiosis between ambrosia fungi and beetles. i. Indian J. Mycol. Plant Pathol. 2:165?169. Baum D, Shaw KL. 1995. Experimental approaches to plant systematics - Monographs in Botany from the Missouri Botanical Garden 53. (Hoch PC, Stephenson AG, editors.). St, Louis: Missouri Botanical Garden De Beer ZW, Seifert KA, Wingfield MJ. 2013. A nomenclator for ophiostomatoid genera and species in the Ophiostomatales and Microascales. In: Seifert KA, de Beer ZW, Wingfield MJ, editors. The Ophiostomatoid Fungi: Expanding Frontiers. CBS Biodiversity Series. Utrecht, Netherlands: CBS-KNAW Fungal Biodiversity Centre. p. 245?322. De Beer ZW, Wingfield MJ. 2013. Emerging lineages in the Ophiostomatales. In: Seifert KA, De Beer ZW, Wingfield MJ, editors. Emerging lineages in the Ophiostomatales. 12. Utrecht, Netherlands: CBS-KNAW Fungal Biodiversity Centre. p. 21?46. Benade E, Wingfield MJ, Wyk PSV. 1995. Conidium Development in the Hyalorhinocladiella Anamorph of Ophiostoma ips. Mycologia 87:298. Benade E, Wingfield MJ, Wyk PSV. 1996. Conidium development in the Hyalorhinocladiella anamorphs of Ceratocystiopsis minuta-bicolor and Ophiostoma minus. Can. J. Bot. 74:891?897. Benjamini Y, Hochberg Y. 1995. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Stat. Soc. Ser. B Methodol. 57:289?300. Bentz BJ, R?gni?re J, Fettig CJ, Hansen EM, Hayes JL, Hicke JA, Kelsey RG, Negr?n JF, Seybold SJ. 2010. Climate Change and Bark Beetles of the Western United States and Canada: Direct and Indirect Effects. BioScience 60:602?613. Berbee ML, Taylor JW. 2001. Fungal Molecular Evolution: Gene Trees and Geologic Time. In: McLaughlin DJ, McLaughlin EG, Lemke PA, editors. The Mycota: a comprehensive     189 treatise on fungi as experimental systems for basic and applied research. Volume VII: Systematics and Evolution, Part B. Be rlin: Springer. p. 229?245. Bickford CP, Kolb TE, Geils BW. 2005. Host physiological condition regulates parasitic plant performance: Arceuthobium vaginatum subsp. cryptopodum on Pinus ponderosa. Oecologia 146:179 ?189. Bleiker K, Six D. 2009. Competition and Coexistence In a Multi-partner Mutualism: Interactions Between two Fungal Symbionts of the Mountain Pine Beetle In Beetle -attacked Trees. Microb. Ecol. 57:191 ?202. Bohlmann J. 2012. Pine terpenoid defences in the mountain pine beetle epidemic and in oth er conifer pest interactions: specialized enemies are eating holes into a diverse, dynamic and durable defence system. Tree Physiol. 32:943?945. Boone CK, Aukema BH, Bohlmann J, Carroll AL, Raffa KF. 2011. Efficacy of tree defense physiology varies with bark beetle population density: a basis for positive feedback in eruptive species. Can. J. For. Res. 41:1174 ?1188. Borden JH. 1984. Semiochemical -mediated aggregation and dispersal in the Coleoptera. In: Lewis T, editor. Insect Communication. London: Academic Press. p. 123?149. Brasier CM. 1987. The dynamics of fungal speciation. In: Rayner AD., Brasier CM, Moore D, editors. Evolutionary Biology of the Fungi. Cambridge, UK: Cambridge University Press. p. 231?260. Bruen TC, Philippe H, Bryant D. 2006. A Simple  and Robust Statistical Test for Detecting the Presence of Recombination. Genetics 172:2665?2681. Brumfield RT, Beerli P, Nickerson DA, Edwards SV. 2003. The utility of single nucleotide polymorphisms in inferences of population history. Trends Ecol. Evol.  18:249?256. Bryant D, Moulton V. 2004. Neighbor -net: an agglomerative method for the construction of phylogenetic networks. Mol. Biol. Evol. 21:255 ?265. Burnett J. 2003. Fungal Population and Species. Oxford, UK: Oxford University Press  Burt A, Carter DA, Koenig GL, White TJ, Taylor JW. 1996. Molecular markers reveal cryptic sex in the human pathogen Coccidioides immitis. Proc. Natl. Acad. Sci. U. S. A. 93:770?773. Butcher D. 1995. Muller?s ratchet, epistasis and mutation effects. Genetics 141:431?437. Carroll A, Aukema B, Raffa K, Linton DA, Smith G, Lindgren B. 2006. Mountain pine beetle outbreak development: The endemic--incipient epidemic transition. Victoria, BC: Canadian Forest Service      190 Cassar S, Blackwell M. 1996. Convergent Origins of Ambrosia Fungi. Mycologia 88 (4):596?601. Clark EL, Carroll AL, Huber DPW. 2010. Differences in the Constitutive Terpene Profile of Lodgepole Pine Across a Geographical Range in British Columbia, and Correlation with Historical Attack by Mountain Pine Beetle. Can. Entomol. 142:557?573. Clement M, Posada D, Crandall KA. 2000. TCS: a computer program to estimate gene genealogies. Mol. Ecol. 9:1657?1659. Coleman JJ, Rounsley SD, Rodriguez-Carres M, et al. 2009. The Genome of Nectria haematococca: Contribution of Supernumerary Chromosomes to Gene Expansion. PLoS Genet 5:e1000618. Collemare J, Billard A, B?hnert HU, Lebrun M-H. 2008. Biosynthesis of secondary metabolites in the rice blast fungus Magnaporthe grisea: the role of hybrid PKS-NRPS in pathogenicity. Mycol. Res. 112:207?215. Collins FS, Brooks LD, Chakravarti A. 1998. A DNA polymorphism discovery resource for research on human genetic variation. Genome Res. 8:1229?1231. Condon BJ, Leng Y, Wu D, et al. 2013. Comparative Genome Structure, Secondary Metabolite, and Effector Coding Capacity across Cochliobolus Pathogens. PLoS Genet 9:e1003233. Conesa A, G?tz S, Garc?a-G?mez JM, Terol J, Tal?n M, Robles M. 2005. Blast2GO: a universal tool for annotation, visualization and analysis in functional genomics research. Bioinformatics 21:3674?3676. Croll D, McDonald BA. 2012. The Accessory Genome as a Cradle for Adaptive Evolution in Pathogens. PLoS Pathog. 8. Cullingham CI, Cooke JEK, Dang S, Davis CS, Cooke BJ, Coltman DW. 2011. Mountain pine beetle host-range expansion threatens the boreal forest. Mol. Ecol. 20:2157?2171. Cuomo CA, G?ldener U, Xu J-R, et al. 2007. The Fusarium graminearum Genome Reveals a Link Between Localized Polymorphism and Pathogen Specialization. Science 317:1400?1402. Darling AE, Mau B, Perna NT. 2010. progressiveMauve: Multiple Genome Alignment with Gene Gain, Loss and Rearrangement. PLoS ONE 5:e11147. Dean RA, Talbot NJ, Ebbole DJ, et al. 2005. The genome sequence of the rice blast fungus Magnaporthe grisea. Nature 434:980?986.     191 Delorme L, Lieutier F. 1990. Monoterpene composition of the preformed and induced resins of Scots pine, and their effect on bark beetles and associated fungi. Eur. J. For. Pathol. 20:304?316. Desjardins CA, Champion MD, Holder JW, et al. 2011. Comparative Genomic Analysis of Human Fungal Pathogens Causing Paracoccidioidomycosis. PLoS Genet 7:e1002345. Dettman JR, Jacobson DJ, Taylor JW. 2003. A Multilocus Genealogical Approach to Phylogenetic Species Recognition in the Model Eukaryote Neurospora. Evolution 57:2703?2720. DiGuistini S, Liao NY, Platt D, et al. 2009. De novo genome sequence assembly of a filamentous fungus using Sanger, 454 and Illumina sequence data. Genome Biol. 10:R94. DiGuistini S, Ralph SG, Lim YW, Holt R, Jones S, Bohlmann J, Breuil C. 2007. Generation and annotation of lodgepole pine and oleoresin-induced expressed sequences from the blue-stain fungus Ophiostoma clavigerum, a Mountain Pine Beetle-associated pathogen. FEMS Microbiol. Lett. 267:151?158. DiGuistini S, Wang Y, Liao NY, et al. 2011. Genome and transcriptome analyses of the mountain pine beetle-fungal symbiont Grosmannia clavigera, a lodgepole pine pathogen. Proc. Natl. Acad. Sci. 108:2504?2509. Duong TA, Beer ZW de, Wingfield BD, Wingfield MJ. 2012. Phylogeny and taxonomy of species in the Grosmannia serpens complex. Mycologia 104:715?732. Eck SH, Benet-Pag?s A, Flisikowski K, Meitinger T, Fries R, Strom TM. 2009. Whole genome sequencing of a single Bos taurus animal for single nucleotide polymorphism discovery. Genome Biol. 10:R82. Ellison CE, Hall C, Kowbel D, Welch J, Brem RB, Glass NL, Taylor JW. 2011. Population genomics and local adaptation in wild isolates of a model microbial eukaryote. Proc. Natl. Acad. Sci. U. S. A. 108:2831?2836. Engel MS, Grimaldi DA. 2004. The first Mesozoic stephanid wasp (Hymenoptera: Stephanidae). J. Paleontol. 78:1192?1197. Farrell B, Sequeira A, O?Meara B, Normark B, Chung J, Jordal B. 2001. The evolution of agriculture in beetles (Curculionidae: Scolytinae and Platypodinae). Evolution 55:2011?2027.     192 Fay JC, Wyckoff GJ, Wu CI. 2001. Positive and negative selection on the human genome. Genetics 158:1227?1234. Fedorova ND, Badger JH, Robson GD, Wortman JR, Nierman WC. 2005. Comparative analysis of programmed cell death pathways in filamentous fungi. BMC Genomics 6:177. Fedorova ND, Khaldi N, Joardar VS, et al. 2008. Genomic islands in the pathogenic filamentous fungus Aspergillus fumigatus. PLoS Genet. 4:e1000046. Felsenstein J. 1985. Confidence limits on phylogenies: an approach using the bootstrap. Evolution 39:783?791. Forgetta V, Leveque G, Dias J, et al. 2013. Sequencing of the Dutch Elm Disease Fungus Genome Using the Roche/454 GS-FLX Titanium System in a Comparison of Multiple Genomics Core Facilities. J. Biomol. Tech. JBT 24:39?49. Forrest GI. 1980. Genotypic variation among native scots pine populations in scotland based on monoterpene analysis. Forestry 53:101 ?128. Fraedrich SW, Harrington TC, Rabaglia RJ, Ulyshen MD, Mayfield AE, Hanula JL, Eickwort JM, Miller DR. 2008. A Fungal Symbiont of the Redbay Ambrosia Beetle Causes a Lethal Wilt in Redbay and Other Lauraceae in the Southeastern United States. Plant Dis. 92:215?224. Francke-Grosmann H. 1952. Uber die Ambrosiazucht der Kiefernborkenkafer Myelophilus minor Htg. und Ips acuminatus Gyll. Medd. Statens Skogsforskning Inst. Stockh. 41:1?52. Francke-Grosmann H. 1967. Symbiosis. In: Henry SM, editor. Ectosymbiosis in Wood-Inhabiting Insects. Vol. 2. Academic, New York, London: Academic Press. p. 141?205. Funk A. 1970. Fungal symbionts of the ambrosia beetle Gnathotrichus sulcatus. Can. J. Bot. 48:1445?1448. Galagan JE, Henn MR, Ma L-J, Cuomo CA, Birren B. 2005. Genomics of the fungal kingdom: Insights into eukaryotic biology. Genome Res. 15:1620?1631. Gao X, Starmer JD. 2008. AWclust: point-and-click software for non-parametric population structure analysis. BMC Bioinformatics 9:77. Gayathri Samarasekera GDN, Bartell NV, Lindgren BS, Cooke JEK, Davis CS, James PMA, Coltman DW, Mock KE, Murray BW. 2012. Spatial genetic structure of the mountain pine beetle (Dendroctonus ponderosae) outbreak in western Canada: historical patterns and contemporary dispersal. Mol. Ecol. 21:2931?2948.     193 Gebhardt, Begerow, Oberwinkler. 2004. Identification of the ambrosia fungus of Xyleborus monographus and X. dryographus (Coleoptera: Curculionidae, Scolytinae). Mycol. Prog. 3:95?102. Gebhardt H, Oberwinkler F. 2005. Conidial development in selected ambrosial species of the genus Raffaelea. Antonie Van Leeuwenhoek 88:61?65. Gebhardt H, Weiss M, Oberwinkler F. 2005. Dryadomyces amasae: a nutritional fungus associated with ambrosia beetles of the genus Amasa (Coleoptera: Curculionidae, Scolytinae). Mycol. Res. 109:687?696. Geiser DM, Pitt JI, Taylor JW. 1998. Cryptic speciation and recombination in the aflatoxin-producing fungus Aspergillus flavus. Proc. Natl. Acad. Sci. U. S. A. 95:388?393. Gernandt DS, Hern?ndez-Le?n S, Salgado-Hern?ndez E, P?rez de La Rosa JA. 2009. Phylogenetic Relationships of Pinus Subsection Ponderosae Inferred from Rapidly Evolving cpDNA Regions. Syst. Bot. 34:481?491. Gerson EA, Kelsey RG, St Clair JB. 2009. Genetic variation of piperidine alkaloids in Pinus ponderosa: a common garden study. Ann. Bot. 103:447 ?457. Giraud T, Gladieux P, Gavrilets S. 2010. Linking the emergence of fungal plant diseases with ecological speciation. Trends Ecol. Evol. 25:387?395. Giraud T, Refr?gier G, Le Gac M, de Vienne DM, Hood ME. 2008. Speciation in fungi. Fungal Genet. Biol. 45:791?802. Giraud T, Villar?al LMMA, Austerlitz F, Le Gac M, Lavigne C. 2006. Importance of the life cycle in sympatric host race formation and speciation of pathogens. Phytopathology 96:280?287. Gladieux P, Devier B, Aguileta G, Cruaud C, Giraud T. 2013. Purifying selection after episodes of recurrent adaptive diversification in fungal pathogens. Infect. Genet. Evol. J. Mol. Epidemiol. Evol. Genet. Infect. Dis. 17:123?131. Goffeau A, Barrell BG, Bussey H, et al. 1996. Life with 6000 Genes. Science 274:546?567. Goldman N, Yang Z. 1994. A codon-based model of nucleotide substitution for protein-coding DNA sequences. Mol. Biol. Evol. 11:725?736. Goodwin SB, Ben M?Barek S, Dhillon B, et al. 2011. Finished Genome of the Fungal Wheat Pathogen Mycosphaerella graminicola Reveals Dispensome Structure, Chromosome Plasticity, and Stealth Pathogenesis. PLoS Genet 7:e1002070.     194 Gu Z, David  L, Petrov D, Jones T, Davis RW, Steinmetz LM. 2005. Elevated evolutionary rates in the laboratory strain of Saccharomyces cerevisiae. Proc. Natl. Acad. Sci. USA. 102:1092? 1097. Guerrero R. 1966. Una nueva especie de hongo inperfecto asociado con el coleoptero Platypus sulcatus Chapius. Revta Investnes Agropecu. Ser. 5 3:97? 103. Hane JK, Lowe RGT, Solomon PS, et al. 2007. Dothideomycete ? Plant Interactions Illuminated by Genome Sequencing and EST Analysis of the Wheat Pathogen Stagonospora nodorum. Plant Cell Online 19:3347? 3368. Hane JK, Rouxel T, Howlett BJ, Kema GH, Goodwin SB, Oliver RP. 2011. A novel mode of chromosomal evolution peculiar to filamentous Ascomycete fungi. Genome Biol. 12:R45. Haridas S, Breuill C, Bohlmann J, Hsiang T. 2011. A biologist?s  guide to de novo genome assembly using next-generation sequence data: A test with fungal genomes. J. Microbiol. Methods 86:368? 375. Haridas S, Wang Y, Lim L, et al. 2013. The genome and transcriptome of the pine saprophyte Ophiostoma piceae, and a comparison with the bark beetle-associated pine pathogen Grosmannia clavigera. BMC Genomics 14:373. Harrington T, McNew D, Steimel J, Hofstra D, Farrell R. 2001. Phylogeny and Taxonomy of the Ophiostoma piceae Complex and the Dutch Elm Disease Fungi. Mycologia 93:111? 136. Harrington T. 2005. Ecology and evolution of mycophagous bark beetles and their fungal partners. In: Vega F, Blackwell M, editors. Ecological and evolutionary advances in insect? fungal associations. New York, NY: Oxford University Press. p. 257 ? 291. Harrington TC, Fraedrich SW, Aghayeva D. 2008. Raffaelea lauricola, a new ambrosia beetle symbiont and pathogen on the Lauraceae. Mycotaxon 104:399? 404. Hartig T. 1844. Ambrosia des Bostrychibs dispar. Allegemeine Jagd-Ztg. NS:73 ? 74. Hartl DL, Clark AG. 2007. Principles of population genetics. Sunderland, Mass.: Sinauer Associates Hausner G, Reid J, Klassen G. 2000. On the phylogeny of members of Ceratocystis ss and Ophiostoma that possess different anamorphic states, with emphasis on the anamorph genus Leptographium, based on partial ribosomal DNA sequences. Can. J. Bot. 78:903?916.     195 Hausner G, Reid J, Klassen GR. 1993. On the phylogeny of Ophiostoma, Ceratocystis s.s., and Microascus, and relationships within Ophiostoma based on partial ribosomal DNA sequences. Can. J. Bot. 71:1249?1265. Hawksworth DL, Kirk PM, Sutton BC, Pegler DN. 1995. Ainsworth & Bisby?s dictionary of the fungi. Wallingford, Oxon, UK: CAB International Hawksworth DL, McNeill J, de Beer ZW, Wingfield MJ. 2013. Names of fungal species with the same epithet applied to different morphs: how to treat them. IMA Fungus 4:53?56. Hawksworth DL. 2001. The magnitude of fungal diversity: the 1.5 million species estimate revisited. Mycol. Res. 105:1422?1432. Hawksworth DL. 2011. A new dawn for the naming of fungi: impacts of decisions made in Melbourne in July 2011 on the future publication and regulation of fungal names. IMA Fungus 2:155?162. Hay CJ. 1956. Experimental crossing of mountain pine beetle with Black Hills beetle. Ann. Entomol. Soc. Am. 49:567?571. Heckman DS, Geiser DM, Eidell BR, Stauffer RL, Kardos NL, Hedges SB. 2001. Molecular Evidence for the Early Colonization of Land by Fungi and Plants. Science 293:1129?1133. Hesse-Orce U, DiGuistini S, Keeling CI, et al. 2010. Gene discovery for the bark beetle-vectored fungal tree pathogen Grosmannia clavigera. BMC Genomics 11:536. Hey J, Nielsen R. 2004. Multilocus methods for estimating population sizes, migration rates and divergence time, with applications to the divergence of Drosophila pseudoobscura and D. persimilis. Genetics 167:747?760. Higby PK, Stock MW. 1982. Genetic Relationships Between Two Sibling Species of Bark Beetle (Coleoptera: Scolytidae), Jeffrey Pine Beetle and Mountain Pine Beetle, in Northern California. Ann. Entomol. Soc. Am. 75:668?674. Hintz W, Pinchback M, de la Bastide P, Burgess S, Jacobi V, Hamelin R, Breuil C, Bernier L. 2011. Functional categorization of unique expressed sequence tags obtained from the yeast-like growth phase of the elm pathogen Ophiostoma novo-ulmi. BMC Genomics 12:431. Hofstetter RW, Mahfouz JB, Klepzig KD, Ayres MP. 2005. Effects of tree phytochemistry on the interactions among endophloedic fungi associated with the southern pine beetle. J. Chem. Ecol. 31:539?560.     196 Hogenhout SA, Van der Hoorn RAL, Terauchi R, Kamoun S. 2009. Emerging concepts in effector biology of plant-associated organisms. Mol. Plant-Microbe Interactions MPMI 22:115?122. Holloway AK, Lawniczak MKN, Mezey JG, Begun DJ, Jones CD. 2007. Adaptive Gene Expression Divergence Inferred from Population Genomics. PLoS Genet 3:e187. Hopkin AD. 1902. On the study of forest entomology in America. U.S. Department of Agriculture, Division of Entomology Hopkins A. 1909. Practical Information on the Scolytid Beetles on North American Forests : Part I - Barkbeetles of the Genus Dendroctonus. Washington D.C. Hsiau PTW, Harrington TC. 2003. Phylogenetics and adaptations of basidiomycetous fungi fed upon by bark beetles (Coleoptera: Scolytidae). Symbiosis 34:111?131. Huber DPW, Aukema BH, Hodgkinson RS, Lindgren BS. 2009. Successful colonization, reproduction, and new generation emergence in live interior hybrid spruce Picea engelmannii glauca by mountain pine beetle Dendroctonus ponderosae. Agric. For. Entomol. 11:83?89. Hudson RR, Coyne JA. 2002. Mathematical consequences of the genealogical species concept. Evol. Int. J. Org. Evol. 56:1557?1565. Hulcr J, Kolarik M, Kirkendall LR. 2007. A new record of fungus-beetle symbiosis in Scolytodes bark beetles (Scolytinae, Curculionidae, Coleoptera). Symbiosis 43:151?159. Hull CM, Raisner RM, Johnson AD. 2000. Evidence for mating of the ?asexual? yeast Candida albicans in a mammalian host. Science 289:307?310. Huson DH, Bryant D. 2006. Application of phylogenetic networks in evolutionary studies. Mol. Biol. Evol. 23:254?267. Huson DH, Lockhart PJ, Steel MA. 2005. Reconstruction of reticulate networks from gene trees. In: In Proceedings of the Ninth International Conference on Research in Computational Molecular Biology (RECOMB). Springer. p. 233?249. Hutchison LJ, Reid J. 1988. Taxonomy of some potential wood-staining fungi from New Zealand 1. Ophiostomataceae. New Zealand J. Bot. 26:63?81. Hynum BG, Berryman AA. 1980. Dendroctonus ponderosae (Coleoptera: Scolytidae) pre-aggregation landing and gallery initiation on lodgepole pine. Can. Entomol. 112:185?191.     197 Jacobs K, Kirisits T, Wingfield MJ. 2003. Taxonomic re -evaluation of three related species of Graphium, based on morphology, ecology and phylogeny. Mycologia 95:714? 727. Jacobs K, Wingfield MJ. 2001. Lep tographium species: tree pathogens, insect associates, and agents of blue-stain. St. Paul, MN: American Phytopathological Society Press James PMA, Coltman DW, Murray BW, Hamelin RC, Sperling FAH. 2011. Spatial Genetic Structure of a Symbiotic Beetle-Fungal System: Toward Multi-Taxa Integrated Landscape Genetics. PLoS ONE 6:e25359. Jones K, Blackwell M. 1998. Phylogenetic analysis of ambrosial species in the genus Raffaelea based on 18S rDNA sequences. Mycol. Res. 102 (6):661? 665. Jordal BH, Cognato AI. 2012. Molecular phylogeny of bark and ambrosia beetles reveals multiple origins of fungus farming during periods of global warming. BMC Evol. Biol. 12:133. Kasuga T, White TJ, Koenig G, et al. 2003. Phylogeography of the fungal pathogen Histoplasma capsulatum. Mol. Ecol. 12:3383? 3401. Kasuga T, White TJ, Taylor JW. 2002. Estimation of Nucleotide Substitution Rates in Eurotiomycete Fungi. Mol. Biol. Evol. 19:2318 ? 2324. Katoh K, Misawa K, Kuma K, Miyata T. 2002. MAFFT: a novel method for rapid multiple sequence alignment based on fast Fourier transform. Nucleic Acids Res. 30:3059? 3066. Katsir L, Chung HS, Koo AJK, Howe GA. 2008. Jasmonate signaling: a conserved mechanism of hormone sensing. Curr. Opin. Plant Biol. 11:428? 435. Keeling CI, Bohlmann J. 2006. Genes, enzymes and chemicals of terpenoid diversity in the constitutive and induced defence of conifers against insects and pathogens. New Phytol. 170:657? 675. Keeling CI, Yuen MM, Liao NY, et al. 2013. Draft genome of the mountain pine beetle, Dendroctonus ponderosae Hopkins, a major forest pest. Genome Biol. 14:R27.  Kelley S, Farrell B. 1998. Is specialization a dead end? The phylogeny of host use in Dendroctonus bark beetles (Scolytidae). Evolution 52:1731? 1743. Kelley ST, Farrell BD, Mitton JB. 2000. Effects o f specialization on genetic differentiation in sister species of bark beetles. Heredity 84:218? 227. Kemen E, Gardiner A, Schultz -Larsen T, et al. 2011. Gene Gain and Loss during Evolution of Obligate Parasitism in the White Rust Pathogen of Arabidopsis thaliana. PLoS Biol 9:e1001094.     198 Khadempour L, Massoumi Alamouti S, Hamelin R, Bohlmann J, Breuil C. 2010. Target-specific PCR primers can detect and differentiate ophiostomatoid fungi from microbial communities associated with the mountain pine beetle Dendroctonus ponderosae. Fungal Biol. 114:825?833. Khoshraftar S, Hung S, Khan S, Gong Y, Tyagi V, Parkinson J, Sain M, Moses AM, Christendat D. 2013. Sequencing and annotation of the Ophiostoma ulmi genome. BMC Genomics 14:162. Kim J-J, Lim YW, Wingfield MJ, Breuil C, Kim G-H. 2004. Leptographium bistatum sp. nov., a new species with a Sporothrix synanamorph from Pinus radiata in Korea. Mycol. Res. 108:699?706. Kirisits T. 2004. Fungal Associates of European Bark Beetles With Special Emphasis on the Ophiostomatoid Fungi. In: Lieutier F, Day KR, Battisti A, Gr?goire J-C, Evans HF, editors. Bark and Wood Boring Insects in Living Trees in Europe, a Synthesis. Springer Netherlands. p. 181?236. Kirk PM, Cannon PF, Minter DW, Stalpers JA. 2008. Ainsworth and Bisby?s Dictionary of the Fungi. Wallingford, UK: CABI Klepzig K, Wilkens R. 1997. Competitive Interactions among Symbiotic Fungi of the Southern Pine Beetle. Appl. Environ. Microbiol. 63:621?627. Klepzig KD, Six DL. 2004. Bark beetle fungal symbioses: Context dependency in complex associations. Symbiosis 37:189?206. Klosterman SJ, Subbarao KV, Kang S, et al. 2011. Comparative Genomics Yields Insights into Niche Adaptation of Plant Vascular Wilt Pathogens. PLoS Pathog 7:e1002137. Knowles L, Carstens B. 2007. Delimiting species without monophyletic gene trees. Syst. Biol. 56:887?895. Kohler GA, Brenot A, Haas-Stapleton E, Agabian N, Deva R, Nigam S. 2006. Phospholipase A2 and Phospholipase B Activities in Fungi. Biochim. Biophys. Acta 1761:1391?1399. Korbel JO, Urban AE, Affourtit JP, et al. 2007. Paired-end mapping reveals extensive structural variation in the human genome. Science 318:420?426. Kovalchuk A, Driessen A. 2010. Phylogenetic analysis of fungal ABC transporters. BMC Genomics 11:177.     199 Kroken S, Glass NL, Taylo r JW, Yoder OC, Turgeon BG. 2003. Phylogenomic analysis of type I polyketide synthase genes in pathogenic and saprobic ascomycetes. Proc. Natl. Acad. Sci. 100:15670? 15675. Krokene P, Solheim H. 1996. Fungal associates of five bark beetle species colonizing  Norway spruce. Can. J. For. Res. 26:2115? 2122. Krupkin AB, Liston A, Strauss SH. 1996. Phylogenetic Analysis of the Hard Pines (Pinus subgenus Pinus, Pinaceae) from Chloroplast DNA Restriction Site Analysis. Am. J. Bot. 83:489? 498. Kubono T, Ito S. 2002. Raffaelea quercivora sp. nov. associated with mass mortality of Japanese oak, and the ambrosia beetle (Platypus quercivorus). Mycoscience 43:0255?0260. Kuhnholz S. 2004. Chemical Ecology and Mechanisms of Reproductive Isolation in Ambrosia Beetles. Kulathi nal RJ, Stevison LS, Noor MAF. 2009. The Genomics of Speciation in Drosophila: Diversity, Divergence, and Introgression Estimated Using Low-Coverage Genome Sequencing. PLoS Genet 5:e1000550. Kurtz S, Phillippy A, Delcher AL, Smoot M, Shumway M, Antonescu C , Salzberg SL. 2004. Versatile and open software for comparing large genomes. Genome Biol. 5:R12? R12. Kurz WA, Dymond CC, Stinson G, Rampley GJ, Neilson ET, Carroll AL, Ebata T, Safranyik L. 2008. Mountain pine beetle and forest carbon feedback to climate change. Nature 452:987? 990. Lagerberg T, Lundberg G, Melin E. 1927. Biological and practical researches into blueing in pine and spruce. Sven. Skogsvardsforen 25:145? 272. Lah L, Haridas S, Bohlmann J, Breuil C. 2013. The cytochromes P450 of Grosmannia clavigera: Genome organization, phylogeny, and expression in response to pine host chemicals. Fungal Genet. Biol. 50:72? 81. Lambreghts R, Shi M, Belden WJ, et al. 2009. A high-density single nucleotide polymorphism map for Neurospora crassa. Genetics 181:767? 781. Langor DW, Spence JR, Pohl GR. 1990. Host effects on fertility and reproductive success of Dendroctonus ponderosae Hopkins (Coleoptera: Scolytidae). Evolution 44:609.     200 Langor DW. 1989. Host effects on the phenology, development, and mortality of field populations of the mountain pine beetle, Dendroctonus Ponderosae Hopkins (Coleoptera:Scolytidae). Can. Entomol. 121:149 ?157. Lanier GN, Wood, DL. 1968. Controlled mating, caryology, morphology and sex ratio in the Dendroctonus ponderosae complex. Ann. Entom ol. Soc. Am. 61:517?526. Latta RG, Linhart YB, Snyder MA, Lundquist L. 2003. Patterns of variation and correlation in the monoterpene composition of xylem oleoresin within populations of ponderosa pine. Biochem. Syst. Ecol. 31:451 ?465. Leach JG, Orr G, Chr istense C. 1934. The interrelationships of bark beetles and blue-staining fungi in felled Norway pine timber. J. Agric. Res. 49:315 ?341. Lee S, Hamelin RC, Six DL, Breuil C. 2007. Genetic diversity and the presence of two distinct groups in Ophiostoma clav igerum associated with Dendroctonus ponderosae in British Columbia and the northern Rocky Mountains. Phytopathology 97:1177?1185. Lee S, Kim J, Breuil C. 2005. Leptographium longiclavatum sp. nov., a new species associated with the mountain pine beetle, Dendroctonus ponderosae. Mycol. Res. 109:1162?1170. Lee S, Kim J, Fung S, Breuil C. 2003. A PCR -RFLP marker distinguishing Ophiostoma clavigerum from morphologically similar Leptographium species associated with bark beetles. Can. J. Bot. 81:1104 ?1112. Lee S, Kim J -J, Breuil C. 2006a. Diversity of fungi associated with mountain pine beetle, Dendroctonus ponderosae, and infested lodgepole pines in British Columbia. Fungal Divers. 22:91?105. Lee S, Kim J -J, Breuil C. 2006b. Pathogenicity of Leptographium longic lavatum associated with Dendroctonus ponderosae to Pinus contorta. Can. J. For. Res. 36:2864 ?2872. Li H, Durbin R. 2009. Fast and accurate short read alignment with Burrows ?Wheeler transform. Bioinformatics 25:1754?1760. Li H, Handsaker B, Wysoker A, et al . 2009. The Sequence Alignment/Map format and SAMtools. Bioinformatics 25:2078 ?2079. Li H, Ruan J, Durbin R. 2008. Mapping short DNA sequencing reads and calling variants using mapping quality scores. Genome Res. 18:1851?1858. Li R, Fan W, Tian G, et al. 2010. The sequence and de novo assembly of the giant panda genome. Nature 463:311?317.     201 Li YF, Costello JC, Holloway AK, Hahn MW. 2008. ?Reverse ecology? and the power of population genomics. Evol. Int. J. Org. Evol. 62:2984?2994. Librado P, Rozas J. 2009. DnaSP v5: a software for comprehensive analysis of DNA polymorphism data. Bioinformatics 25:1451 ?1452. Lieutier F, Yart A, Salle A. 2009. Stimulation of tree defenses by Ophiostomatoid fungi can explain attack success of bark beetles on conifers. Ann. For. Sci. 66:801?801. Lim Y, Alamouti S, Kim J, Lee S, Breuil C. 2004. Multigene phylogenies of Ophiostoma clavigerum and closely related species from bark beetle-attacked Pinus in North America. FEMS Microbiol. Lett. 237:89?96. Liu M, Gelli A. 2008. Elongation Factor 3, EF3, Associates with the Calcium Channel Cch1 and Targets Cch1 to the Plasma Membrane in Cryptococcus neoformans. Eukaryot. Cell 7:1118?1126. Logan JA, Powell JA. 2001. Ghost forests, global warming, and the mountain pine beetle (Coleoptera: Scolytidae). Am. Entomol. 47:160?173. Luhtala N. 2004. Bro1 coordinates deubiquitination in the multivesicular body pathway by recruiting Doa4 to endosomes. J. Cell Biol. 166:717?729. Lynch M, B?rger R, Butcher D, Gabriel W. 1993. The Mutational Meltdown in Asexual Populations. J. Hered. 84:339?344. Ma L-J, van der Does H Charlotte, Borkovich KA, et al. 2010. Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature 464:367?373. Ma L-J, van der Does H. Charlotte, Borkovich KA, et al. 2010. Comparative genomics reveals mobile pathogenicity chromosomes in Fusarium. Nature 464:367?373. Maddison WP, Knowles LL. 2006. Inferring Phylogeny Despite Incomplete Lineage Sorting. Syst. Biol. 55:21?30. Magee BB, Magee PT. 2000. Induction of mating in Candida albicans by construction of MTLa and MTLalpha strains. Science 289:310?313. Malloch D, Blackwell M. 1993. Dispersal biology of the Ophiostomatoid Fungi. In: Wingfield MJ, Seifert KA, Webber JF, editors. Ceratocystis and ophiostoma: taxonomy, ecology, and pathogenicity. St. Paul, Minnesota: American Phytopathological Society Press. p. 195?206. Manning VA, Pandelova I, Dhillon B, et al. 2013. Comparative Genomics of a Plant-Pathogenic Fungus, Pyrenophora tritici-repentis, Reveals Transduplication and the Impact of Repeat     202 Elements on Pathogenicity and Population Divergence. G3 GenesGenomesGenetics 3:41?63. Marcet-Houben M, Ballester A-R, Fuente B de la, Harries E, Marcos JF, Gonz?lez-Candelas L, Gabald?n T. 2012. Genome sequence of the necrotrophic fungus Penicillium digitatum, the main postharvest pathogen of citrus. BMC Genomics 13:646. Mardis ER. 2008. Next-generation DNA sequencing methods. Annu. Rev. Genomics Hum. Genet. 9:387?402. Marin M, Preisig O, Wingfield BD, Kirisits T, Wingfield MJ. 2009. Single sequence repeat markers reflect diversity and geographic barriers in Eurasian populations of the conifer pathogen Ceratocystis polonica. For. Pathol. 39:249?265. Marshall DH, Newton C, Ritland K. 2002. Chloroplast phylogeography and evolution of highly polymorphic microsatellites in lodgepole pine ( Pinus contorta). TAG Theor. Appl. Genet. Theor. Angew. Genet. 104:367?378. Marvaldi AE, Sequeira AS, O?Brien CW, Farrell BD. 2002. Molecular and morphological phylogenetics of weevils (coleoptera, curculionoidea): do niche shifts accompany diversification? Syst. Biol. 51:761?785. Massoumi Alamouti S, Kim J-J, Humble LM, Uzunovic A, Breuil C. 2007. Ophiostomatoid fungi associated with the northern spruce engraver, Ips perturbatus, in western Canada. Antonie Van Leeuwenhoek 91:19?34. Massoumi Alamouti S, Tsui CKM, Breuil C. 2009. Multigene phylogeny of filamentous ambrosia fungi associated with ambrosia and bark beetles. Mycol. Res. 113:822?835. Massoumi Alamouti S, Wang V, DiGuistini S, Six DL, Bohlmann J, Hamelin RC, Feau N, Breuil C. 2011. Gene genealogies reveal cryptic species and host preferences for the pine fungal pathogen Grosmannia clavigera. Mol. Ecol. 20:2581?2602. Matute DR, McEwen JG, Puccia R, et al. 2006. Cryptic Speciation and Recombination in the Fungus Paracoccidioides brasiliensis as Revealed by Gene Genealogies. Mol Biol Evol 23:65?73. Mayden RL. 1997. The denouement in the saga of the species problem. In: Claridge MF, Dawah HA, Wilson MR, editors. Species: The Units of Biodiversity. London: Chapman and Hall. p. 381?424.     203 McCluskey K, Wiest AE, Grigoriev IV, Lipzen A, Martin J, Schackwitz W, Baker SE. 2011. Rediscovery by Whole Genome Sequencing: Classical Mutations and Genome Polymorphisms in Neurospora crassa. G3 Genes Genomes Genet. 1:303?316. McDonald JH, Kreitman M. 1991. Adaptive protein evolution at the Adh locus in Drosophila. Nature 351:652?654. McKenna A, Hanna M, Banks E, et al. 2010. The Genome Analysis Toolkit: A MapReduce framework for analyzing next-generation DNA sequencing data. Genome Res. 20:1297?1303. Metzker ML. 2010. Sequencing technologies - the next generation. Nat. Rev. Genet. 11:31?46. Miller JR, Koren S, Sutton G. 2010. Assembly algorithms for next-generation sequencing data. Genomics 95:315?327. Minter DW, Kirk PM, Sutton BC. 1982. Holoblastic phialides. Trans. Br. Mycol. Soc. 79:75?93. Minter DW, Kirk PM, Sutton BC. 1983. Thallic phialides. Trans. Br. Mycol. Soc. 80:39?66. Mirov NT, Hasbrouck J. 1976. The Story of Pines. First Edition. Bloomington, IN: Indiana University Press Mirov NT. 1948. The terpenes (in relation to the biology of genus Pinus). Annu. Rev. Biochem. 17:521?540. Mock KE, Bentz BJ, O?neill EM, Chong JP, Orwin J, Pfrender ME. 2007. Landscape-scale genetic variation in a forest outbreak species, the mountain pine beetle (Dendroctonus ponderosae). Mol. Ecol. 16:553?568. M?ller EM, Bahnweg G, Sandermann H, Geiger HH. 1992. A simple and efficient protocol for isolation of high molecular weight DNA from filamentous fungi, fruit bodies, and infected plant tissues. Nucleic Acids Res. 20:6115?6116. Moran GP, Coleman DC, Sullivan DJ. 2011. Comparative genomics and the evolution of pathogenicity in human pathogenic fungi. Eukaryot. Cell 10:34?42. Morin PA, Luikart G, Wayne RK, the SNP workshop group. 2004. SNPs in ecology, evolution and conservation. Trends Ecol. Evol. 19:208?216. M?nch E. 1907. Die Blaufaule des Nadelholzes. Naturw Ztschr F Forst U Landw 5:531?573. Neafsey DE, Barker BM, Sharpton TJ, et al. 2010. Population genomic sequencing of Coccidioides fungi reveals recent hybridization and transposon control. Genome Res. 20:938?946.     204 Nielsen R, Bustamante C, Clark AG, et al. 2005. A scan for positively selected genes in the genomes of humans and chimpanzees. PLoS Biol. 3:e170. Nielsen R, Yang Z. 1998. Likelihood models for detecting positively selected amino acid sites and applications to the HIV-1 envelope gene. Genetics 148:929?936. Nowrousian M, Stajich JE, Chu M, et al. 2010. De novo Assembly of a 40 Mb Eukaryotic Genome from Short Sequence Reads: Sordaria macrospora, a Model Organism for Fungal Morphogenesis. PLoS Genet 6:e1000891. Nowrousian M. 2010. Next-Generation Sequencing Techniques for Eukaryotic Microorganisms: Sequencing-Based Solutions to Biological Problems. Eukaryot. Cell 9:1300?1310. Nylander JAA, Wilgenbusch JC, Warren DL, Swofford DL. 2008. AWTY (are we there yet?): a system for graphical exploration of MCMC convergence in Bayesian phylogenetics. Bioinformatics 24:581?583. O?Donnell K. 1992. Ribosomal DNA internal transcribed spacers are highly divergent in the phytopathogenic ascomycete Fusarium sambucinum (Gibberella pulicaris). Curr. Genet. 22:213?220. Okada G, Seifert K, Takematsu A, Yamaoka Y. 1998. A molecular phylogenetic reappraisal of the Graphium complex based on 18S rDNA sequences. Can J Bot. Oleksyk TK, Smith MW, O?Brien SJ. 2010. Genome-wide scans for footprints of natural selection. Philos. Trans. R. Soc. B Biol. Sci. 365:185 ?205. Paine TD, Hanlon CC. 1994. Influence of oleoresin constituents from Pinus ponderosa and Pinus jeffreyi on growth of mycangial fungi from Dendroctonus ponderosae and Dendroctonus jeffreyi. J. Chem. Ecol. 20:2551?2563. Paine TD, Raffa KF, Harrington TC. 1997. Interactions Among Scolytid Bark Beetles, Their Associated Fungi, and Live Host Conifers. Annu. Rev. Entomol. 42:179?206. Paulin-Mahady AE, Harrington TC, McNew D. 2002. Phylogenetic and taxonomic evaluation of Chalara, Chalaropsis, and Thielaviopsis anamorphs associated with Ceratocystis. Mycologia 94:62?72. Pawlowska TE, Taylor JW. 2004. Organization of genetic variation in individuals of arbuscular mycorrhizal fungi. Nature 427:733?737. Pollack JK, Harris SD, Marten MR. 2009. Autophagy in filamentous fungi. Fungal Genet. Biol. 46:1?8.     205 Posada D, Crandall K. 1998. MODELTEST: testing the model of DNA substitution. Bioinformatics 14:817?818. Posada D. 2008. jModelTest: phylogenetic model averaging. Mol. Biol. Evol. 25:1253 ?1256. Pringle A, Baker DM, Platt JL, Wares JP, Latg? J-P, Taylor JW. 2005. cryptic speciation in the cosmopolitan and human pathogenic fungus Aspergillus fumigatus. Evolution 59:1886?1899. De Queiroz K. 1998. The general lineage concept of species, species criteria, and the process of speciation. In: Howard DJ, Berlocher SH, editors. Endless forms: species and speciation. New York: Oxford University Press. p. 57?75. De Queiroz K. 1999. The general lineage concept of species and the defining properties of the species category. De Queiroz K. 2007. Species Concepts and Species Delimitation. Syst. Biol. 56:879?886. Quinlan AR, Hall IM. 2010. BEDTools: a flexible suite of utilities for comparing genomic features. Bioinforma. Oxf. Engl. 26:841?842. Raffa, Aukema B, Erbilgin N, Klepzig KD, Wallin KF. 2005. Recent advances in phytochemistry. Toronto, Canada: Elsevier Raffa K. 1988. The mountain pine beetle in western North America. In: Berryman A, editor. Dynamics of forest insect populations patterns, causes, implications. New York, NY: Plenum Press. p. 506?550. Raffa K. 2001. Mixed Messages Across Multiple Trophic Levels: The Ecology of Bark Beetle Chemical Communication Systems. Chemoecology 11:49?65. Raffa KF, Berryman AA. 1982. Gustatory cues in the orientation of Dendroctonus ponderosae (Coleoptera: Scolytidae) to host trees. Can. Entomol. 114:97?104. Raffa KF, Berryman AA. 1983. The role of host plant resistance in the colonization behavior and ecology of Bark Beetles (Coleoptera: Scolytidae). Ecol. Monogr. 53:27?49. Raffaele S, Farrer RA, Cano LM, et al. 2010. Genome Evolution Following Host Jumps in the Irish Potato Famine Pathogen Lineage. Science 330:1540?1543. Raffaele S, Kamoun S. 2012. Genome evolution in filamentous plant pathogens: why bigger can be better. Nat. Rev. Microbiol. 10:417?430. Rand DM, Kann LM. 1996. Excess amino acid polymorphism in mitochondrial DNA: contrasts among genes from Drosophila, mice, and humans. Mol. Biol. Evol. 13:735?748.     206 Reid RW. 1963. Biology of the Mountain Pine Beetle, Dendroctonus monticolae Hopkins, in the East Kootenay Region of British Columbia.: III. Interaction between the Beetle and its Host, with Emphasis on Brood Mortality and Survival. Can. Entomol. 95:225 ?238. Reinhardt JA, Baltrus DA, Nishimura MT, Jeck WR, Jones CD, Dangl JL. 2009. De novo assembly using low-coverage short read sequence data from the rice pathogen Pseudomonas syringae pv. oryzae. Genome Res. 19:294 ?305. Renwick JAA, Pitman GB. 1979. An attractant isolated from female Jeffrey pine beetles, Dendroctonus jeffreyi. Environ. Entomol. 8:40 ?41. Reumers J, Maurer -Stroh S, Schymkowitz J, Rousseau F. 2006. SNPeffect v2.0: a new step i n investigating the molecular phenotypic effects of human non-synonymous SNPs. Bioinformatics 22:2183 ?2185. Rice AV, Thormann Markus N., Langor DW. 2007. Mountain pine beetle associated blue -stain fungi cause lesions on jack pine, lodgepole pine, and lodgepole ! jack pine hybrids in Alberta. Can. J. Bot. 85:307 ?315. Rice AV, Thormann M.N., Langor DW. 2007. Virulence of, and interactions among, mountain pine beetle associated blue-stain fungi on two pine species and their hybrids in Alberta. Can. J. Bot. 85: 316?323. Richardson DM. 2000. Ecology and biogeography of Pinus. Cambridge, England: Cambridge University Press  Robinson RC. 1962. Blue stain fungi in lodgepole pine (Pinus contorta Doug. var. latifolia Engelm.) infested by the mountain pine beetle (Dendro ctonus monticolae Hopk.). Robinson -Jeffrey RC, Davidson RW. 1968. Three new Europhium species with Verticicladiella imperfect states on blue-stained pine. Can. J. Bot. 46:1523 ?1527. Roe AD, Rice AV, Bromilow SE, Cooke JEK, Sperling FAH. 2010. Multilocus sp ecies identification and fungal DNA barcoding: insights from blue stain fungal symbionts of the mountain pine beetle. Mol. Ecol. Resour. 10:946 ?959. Roe AD, Rice AV, Coltman DW, Cooke JEK, Sperling FAH. 2011. Comparative phylogeography, genetic differentiation and contrasting reproductive modes in three fungal symbionts of a multipartite bark beetle symbiosis. Mol. Ecol. 20:584 ?600. Roeper RA. 1995. Patterns of mycetophagy in Michigan ambrosia beetles. Mich. Acad. 26:153 ?161.     207 Rokas A, Payne G, Fedorova ND, et al. 2007. What can comparative genomics tell us about species concepts in the genus Aspergillus? Stud. Mycol. 59:11?17. Rokas A. 2009. The effect of domestication on the fungal proteome. Trends Genet. 25:60?63. Rollins F, Jones K, Krokene P, Solheim H. 2001. Phylogeny of Asexual Fungi Associated with Bark and Ambrosia Beetles. Mycologia 93 (5):991?996. Ronquist F, Huelsenbeck JP. 2003. MrBayes 3: Bayesian phylogenetic inference under mixed models. Bioinforma. Oxf. Engl. 19:1572?1574. Rumbold CT. 1941. A blue stain fungus, Ceratostomella montium n. sp., and some yeasts associated with two species of Dendroctonus. J. Agric. Res. 62:589?60l. Rydholm C, Szakacs G, Lutzoni F. 2006. Low Genetic Variation and No Detectable Population Structure in Aspergillus fumigatus Compared to Closely Related Neosartorya Species. Eukaryot. Cell 5:650?657. Safranyik L, Carroll A l., R?gni?re J, et al. 2010. Potential for range expansion of mountain pine beetle into the boreal forest of North America. Can. Entomol. 142:415?442. Safranyik L, Carroll A. 2006. The biology and epidemiology of the mountain pine beetle in lodgepole pine forests. In: Safranyik L, Wilson B, editors. The mountain pine beetle: a synthesis of biology, management, and impacts on lodgepole pine. Victoria, BC: Canadian Forest Service. p. 3?66. Safranyik L, Linton DA. 1983. Brood production by three spp. of Dendroctonus (Coleoptera: Scolytidae) in bolts from host and nonhost trees. Entomol. Soc. Br. Columbia 80:10?13. Safranyik L, Wilson WR. 2006. The mountain pine beetle: a synthesis of biology, management, and impacts on lodgepole pine. Victoria, BC: Pacific Forestry Centre Savluchinske Feio S, Gigante B, Roseiro JC, Marcelo-Curto MJ. 1999. Antimicrobial activity of diterpene resin acid derivatives. J. Microbiol. Methods 35:201?206. Schirawski J, Mannhaupt G, M?nch K, et al. 2010. Pathogenicity Determinants in Smut Fungi Revealed by Genome Comparison. Science 330:1546?1548. Schmieder R, Edwards R. 2011. Quality control and preprocessing of metagenomic datasets. Bioinforma. Oxf. Engl. 27:863?864. Scott DB, Du Toit JW. 1970. Three new Raffaelea species. Trans. Br. Mycol. Soc. 55:181?186. Seifert KA, Okada G. 1993. Graphium anamorphs of Ophiostoma species and similar anamorphs of other ascomycetes. In: Wingfield MJ, Seifert Keith A., Webber JF, editors.     208 Ceratocystis and Ophiostoma: Taxonomy, Ecology and Pathology. St. Paul, MN: American Phytopathological Society Press. p. 27?41. Seifert KA, Samuels GJ. 2000. How should we look at anamorphs. Stud. Mycol. 45:5?18. Selosse M-A, Le Tacon F. 1998. The land flora: a phototroph-fungus partnership? Trends Ecol. Evol. 13:15?20. Seybold SJ, Bohlmann J, Raffa KF. 2000. Biosynthesis of coniferophagous bark beetle pheromones and conifer isoprenoids: evolutionary perspective and synthesis. Can. Entomol. 132:697?753. Shaffer HB, Thomson RC. 2007. Delimiting Species in Recent Radiations. Syst. Biol. 56:896?906. Sharpton TJ, Stajich JE, Rounsley SD, et al. 2009. Comparative genomic analyses of the human fungal pathogens Coccidioides and their relatives. Genome Res. 19:1722?1731. She R, Chu JS-C, Uyar B, Wang J, Wang K, Chen N. 2011. genBlastG: using BLAST searches to build homologous gene models. Bioinformatics 27:2141?2143. Simpson JT, Wong K, Jackman SD, Schein JE, Jones SJM, Birol I. 2009. ABySS: a parallel assembler for short read sequence data. Genome Res. 19:1117?1123. Six D, Harrington TC, Steimel J, McNew D, Paine TD. 2003. Genetic relationships among Leptographium terebrantis and the mycangial fungi of three western Dendroctonus bark beetles. Mycologia 95:781?792. Six DL, Paine TD, Hare JD. 1999. Allozyme diversity and gene flow in the bark beetle, Dendroctonus jeffieyi (Coleoptera : Scolytidae). Can. J. For. Res. 29:315?323. Six DL, Paine TD. 1997. Ophiostoma clavigerum Is the Mycangial Fungus of the Jeffrey Pine Beetle, Dendroctonus jeffreyi. Mycologia 89:858?866. Six DL, Paine TD. 1998. Effects of mycangial fungi and host tree species on progeny survival and emergence of Dendroctonus ponderosae (Coleoptera: Scolytidae). Environ. Entomol. 27:1393?1401. Six DL, Paine TD. 1999. Allozyme diversity and gene flow in Ophiostoma clavigerum (Ophiostomatales: Ophiostomataceae), the mycangial fungus of the Jeffrey pine beetle, Dendroctonus jeffreyi. Can. J. For. Res. 29:324. Six DL, Wingfield MJ. 2011. The role of phytopathogenicity in bark beetle-fungus symbioses: a challenge to the classic paradigm. Annu. Rev. Entomol. 56:255?272.     209 Six DL. 2003. Bark Beetle-Fungus Symbioses. In: Bourtzis K, Miller TA, editors. Insect Symbioses. CRC Press. p. 97?114. Slatkin M, Maddison WP. 1990. Detecting isolation by distance using phylogenies of genes. Genetics 126:249?260. Smith GD, Carroll AL, Lindgren BS. 2010. Facilitation in bark beetles: endemic mountain pine beetle gets a helping hand. Agric. For. Entomol. 13:37?43. Smith JM, Smith NH, O?Rourke M, Spratt BG. 1993. How clonal are bacteria? Proc. Natl. Acad. Sci. U. S. A. 90:4384 ?4388. Smith RH. 2000. X ylem monoterpenes of pines: distribution, variation, genetics, function. Berkeley, Calif: USDA  Soanes DM, Alam I, Cornell M, et al. 2008. Comparative Genome Analysis of Filamentous Fungi Reveals Gene Family Expansions Associated with Fungal Pathogenesis. PLoS ONE 3:e2300. Solheim H, Krokene P. 1998. Growth and virulence of mountain pine beetle associated blue-stain fungi, Ophiostoma clavigerum and Ophiostoma montium. Can. J. Bot. 76:561?566. Spanu PD, Abbott JC, Amselem J, et al. 2010. Genome Expansion and Gene Loss in Powdery Mildew Fungi Reveal Tradeoffs in Extreme Parasitism. Science 330:1543?1546. Spatafora J, Blackwell M. 1994. The polyphyletic origins of ophiostomatoid fungi. Mycol. Res. 98 (1):1?9. Spooner B, Roberts P. 2010. Fungi (Collins New Naturalist Library, Book 96). HarperCollins UK  Stajich JE, Berbee ML, Blackwell M, Hibbett DS, James TY, Spatafora JW, Taylor JW. 2009. Primer -- The Fungi. Curr. Biol. CB 19:R840?R845. Stamatakis A. 2006. RAxML-VI-HPC: maximum likelihood-based phylogenetic analyses with thousands of taxa and mixed models. Bioinformatics 22:2688 ?2690. Stock MW, Amman GD, Higby PK. 1984. Genetic Variation Among Mountain Pine Beetle (Dendroctonus ponderosae) (Coleoptera: Scolytidae) Populations from Seven Western States. Ann. Entomol. Soc. Am. 77:760?764. Stock MW, Amman GD. 1980. Genetic differentiation among mountain pine beetle populations from lodgepole pine and ponderosa pine in northeast Utah. Ann. Entomol. Soc. Am. 73:472?478. Stock MW, Guenter JD. 1979. Isozyme Variation Among Mountain Pine Beetle (Dendroctonus ponderosae) Populations in the Pacific Northwest. Environ. Entomol. 8:889?893.     210 Stoletzki N, Eyre-Walker A. 2010. The positive correlation between dN/dS and dS in mammals is due to runs of adjacent substitutions. Mol. Biol. Evol. [Internet]. Available from: http://mbe.oxfordjournals.org/content/early/2010/11/29/molbev.msq320 Stoletzki N, Eyre-Walker A. 2011. Estimation of the Neutrality Index. Mol. Biol. Evol. 28:63?70. Stukenbrock Eva H., Bataillon T, Dutheil JY, Hansen TT, Li R, Zala M, McDonald BA, Wang J, Schierup MH. 2011. The making of a new pathogen: Insights from comparative population genomics of the domesticated wheat pathogen Mycosphaerella graminicola and its wild sister species. Genome Res. Stukenbrock E. H., Bataillon T, Dutheil JY, Hansen TT, Li R, Zala M, McDonald BA, Wang J, Schierup MH. 2011. The making of a new pathogen: Insights from comparative population genomics of the domesticated wheat pathogen Mycosphaerella graminicola and its wild sister species. Genome Res. 21:2157?2166. Stukenbrock EH, Bataillon T. 2012. A Population Genomics Perspective on the Emergence and Adaptation of New Plant Pathogens in Agro-Ecosystems.Heitman J, editor. PLoS Pathog. 8:e1002893. Stukenbrock EH, Christiansen FB, Hansen TT, Dutheil JY, Schierup MH. 2012. Fusion of two divergent fungal individuals led to the recent emergence of a unique widespread pathogen species. Proc. Natl. Acad. Sci. U. S. A. 109:10954?10959. Stukenbrock EH, J?rgensen FG, Zala M, Hansen TT, McDonald BA, Schierup MH. 2010. Whole-Genome and Chromosome Evolution Associated with Host Adaptation and Speciation of the Wheat Pathogen Mycosphaerella graminicola. PLoS Genet. 6. Stukenbrock EH, McDonald BA. 2008. The Origins of Plant Pathogens in Agro-Ecosystems. Annu. Rev. Phytopathol. 46:75?100. Stukenbrock EH. 2013. Evolution, selection and isolation: a genomic view of speciation in fungal plant pathogens. New Phytol. 199:895?907. Sturgeon KB, Mitton JB. 1982. Evolution of bark beetle communities. In: Mitton JB, Sturgeon K. B., editors. Bark Beetles in North American Conifers. Austin, TX: Texas University Press. p. 350?384. Sturgeon KB, Mitton JB. 1986a. Allozyme and Morphological Differentiation of Mountain Pine Beetles Dendroctonus ponderosae Hopkins (Coleoptera: scolytidae) Associated with Host Tree. Evolution 40:290?302.     211 Sturgeon KB, Mitton JB. 1986b. Allozyme and Morphological Differentiation of Mountain Pine Beetles Dendroctonus ponderosae Hopkins (Coleoptera: scolytidae) Associated with Host Tree. Evolution 40:290?302. Sturgeon KB. 1979. Monoterpene Variation in Ponderosa Pine Xylem Resin Related to Western Pine Beetle Predation. Evolution 33:803. Swofford D, Olsen G, Waddell P, Hillis D. 1996. Phylogenetic inference. In: Hillis D, Moritz C, Mable B, editors. Molecular Systematics. Sunderland, MA: Sinauer Associates. p. 407?514. Swofford D. 2003. PAUP*: phylogenetic analysis using parsimony, version 4.0b10. Tamura K, Dudley J, Nei M, Kumar S. 2007. MEGA4: Molecular Evolutionary Genetics Analysis (MEGA) Software Version 4.0. Mol. Biol. Evol. 24:1596 ?1599. Tamura K, Peterson D, Peterson N, Stecher G, Nei M, Kumar S. 2011. MEGA5: molecular evolutionary genetics analysis using maximum likelihood, evolutionary distance, and maximum parsimony methods. Mol. Biol. Evol. 28:2731?2739. Taylor J, Jacobson D, Fisher M. 1999. The evolution of asexual fungi: Reproduction, Speciation and Classification. Annu. Rev. Phytopathol. 37:197?246. Taylor JW, Jacobson DJ, Kroken S, Kasuga T, Geiser DM, Hibbett DS, Fisher MC. 2000. Phylogenetic species recognition and species concepts in fungi. Fungal Genet. Biol. 31:21?32. Taylor TN, Hass H, Remy W. 1992. Devonian Fungi: Interactions with the Green Alga Palaeonitella. Mycologia 84:901. Taylor TN, Osborn JM. 1996. The importance of fungi in shaping the paleoecosystem. Rev. Palaeobot. Palynol. 90:249?262. Templeton AR. 1983. Phylogenetic Inference From Restriction Endonuclease Cleavage Site Maps with Particular Reference to the Evolution of Humans and the Apes. Evolution 37:221. Thompson JN. 1994. The coevolutionary process. Chicago, IL: University of Chicago Press Trapnell C, Salzberg SL. 2009. How to map billions of short reads onto genomes. Nat. Biotechnol. 27:455?457. Tsui CK-M, DiGuistini S, Wang Y, Feau N, Dhillon B, Bohlmann J, Hamelin RC. 2013. Unequal Recombination and Evolution of the Mating-Type (MAT) Loci in the Pathogenic Fungus Grosmannia clavigera and Relatives. G3 Genes Genomes Genet. 3:465?480.     212 Tsui CKM, Feau N, Ritland CE, Massoumi Alamouti S, DiGuistini S, Khadempour L, Bohlmann J, Breuil C, Hamelin RC. 2009. Characterization of microsatellite loci in the fungus, Grosmannia clavigera, a pine pathogen associated with the mountain pine beetle. Mol. Ecol. Resour. 9:1500?1503. Tsui CKM, Roe AD, El-Kassaby YA, Rice AV, Alamouti SM, Sperling F a. H, Cooke JEK, Bohlmann J, Hamelin RC. 2012. Population structure and migration pattern of a conifer pathogen, Grosmannia clavigera, as influenced by its symbiont, the mountain pine beetle. Mol. Ecol. 21:71?86. Tsuneda A, Currah R. 2006. Toward a deeper understanding of the nature of pleomorphism in conidial fungi. Reports Tottori Mycol. Inst. Jpn. 44:1?52. Tsuneda A, Hiratsuka Y. 1984. Sympodial and annellidic conidiation in Ceratocystis clavigera. Can. J. Bot. 62:2618?2624. Tyler BM, Tripathy S, Zhang X, et al. 2006. Phytophthora Genome Sequences Uncover Evolutionary Origins and Mechanisms of Pathogenesis. Science 313:1261?1266. Upadhyay H, Kendrick W. 1974. A New Graphium-like Genus (Conidial State of Ceratocystis). Mycologia 66:181?13. Upadhyay HP. 1981. Monograph of Ceratocystis and Ceratocystiopsis. Athens: University of Georgia Press Upadhyay HP. 1993. Classification of the ophiostomatoid fungi. In: Wingfield MJ, Seifert KA, Webber .JF., editors. Ceratocystis and ophiostoma: taxonomy, ecology, and pathogenicity. St. Paul, Mirmesota.: American Phytopathological Society Press. p. 7?13. Verrall AF. 1943. Fungi associated with certain ambrosia beetles. Agric. Res. 66:135?144. Vilgalys R, Hester M. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J. Bacteriol. 172:4238?4246. Wang J, Wang W, Li R, et al. 2008. The diploid genome sequence of an Asian individual. Nature 456:60?65. Wang W, Wei Z, Lam T-W, Wang J. 2011. Next generation sequencing has lower sequence coverage and poorer SNP-detection capability in the regulatory regions. Sci. Reports [Internet] 1. Available from: http://www.nature.com/srep/2011/110805/srep00055/full/srep00055.html     213 Wang Y, Lim L, DiGuistini S, Robertson G, Bohlmann J, Breuil C. 2013. A specialized ABC efflux transporter GcABC-G1 confers monoterpene resistance to Grosmannia clavigera, a bark beetle-associated fungal pathogen of pine trees. New Phytol. 197:886?898. Wheeler DA, Srinivasan M, Egholm M, et al. 2008. The complete genome of an individual by massively parallel DNA sequencing. Nature 452:872?876. Whitney H. 1971. Association of Dendroctonus ponderosae (Coleoptera: Scolytidae) with blue stain fungi and yeasts during brood development in lodgepole pine. Can. Entomol. 103:1495?1503. Whitney HS, Farris SH. 1970. Maxillary Mycangium in the Mountain Pine Beetle. Science 167:54?55. William J, Ballard O. 1996. Combining data in phylogenetic analysis. Trends Ecol. Evol. 11:334. Wingfield MJ, De Beer ZW, Slippers B, Wingfield BD, Groenewald JZ, Lombard L, Crous PW. 2012. One fungus, one name promotes progressive plant pathology. Mol. Plant Pathol. 13:604?613. Wingfield MJ, Seifert KA, Webber .J.F. eds. 1993. Ceratocystis and ophiostoma: taxonomy, ecology, and pathogenicity. St. Paul, Mirmesota: American Phytopathological Society Press Wingfield MJ, Seifert KA. 1993. Ceratocystis and Ophiostoma: Taxonomy, Ecology, and Pathogenicity. (Webber JF, editor.). Amer Phytopathological Society De Wit PJGM, van der Burgt A, ?kmen B, et al. 2012. The Genomes of the Fungal Plant Pathogens Cladosporium fulvum and Dothistroma septosporum Reveal Adaptation to Different Hosts and Lifestyles But Also Signatures of Common Ancestry. PLoS Genet 8:e1003088. Wood S. 1963. A revision of the bark beetle genus. Dendroctonus Erichson (Coleoptera: Scolytidae). Gt. Basin Nat 23:1?117. Wood S. 1982. The Bark and ambrosia beetles of North and Central America (Coleoptera, Scolytidae): a taxonomic monograph. Great Basin naturalist memoirs, no. 6 Wright SI, Andolfatto P. 2008. The Impact of Natural Selection on the Genome: Emerging Patterns in Drosophila and Arabidopsis. Annu. Rev. Ecol. Evol. Syst. 39:193?213. Xue M, Yang J, Li Z, et al. 2012. Comparative Analysis of the Genomes of Two Field Isolates of the Rice Blast Fungus Magnaporthe oryzae. PLoS Genet 8:e1002869.     214 Yamaoka Y, Hiratsuka Y, Maruyama PJ. 1995. The ability of Ophiostoma clavigerum to kill mature lodgepolepine trees. Eur. J. For. Pathol. 25:401?404. Yamaoka Y, Swanson RH, Hiratsuka Y. 1990. Inoculation of lodgepole pine with four blue-stain fungi associated with mountain pine beetle, monitored by a heat pulse velocity (HPV) instrument. Can. J. For. Res. 20:31?36. Yang Z, Nielsen R. 2000. Estimating Synonymous and Nonsynonymous Substitution Rates Under Realistic Evolutionary Models. Mol. Biol. Evol. 17:32?43. Yang Z, Swanson WJ. 2002. Codon-Substitution Models to Detect Adaptive Evolution that Account for Heterogeneous Selective Pressures Among Site Classes. Mol. Biol. Evol. 19:49?57. Yang Z. 2007. PAML 4: Phylogenetic Analysis by Maximum Likelihood. Mol. Biol. Evol. 24:1586?1591. Zambino PJ, Harrington TC. 1992. Correspondence of Isozyme Characterization with Morphology in the Asexual Genus Leptographium and Taxonomic Implications. Mycologia 84:12. Zipfel RD, de Beer ZW, Jacobs K, Wingfield BD, Wingfield MJ. 2006. Multi-gene phylogenies define Ceratocystiopsis and Grosmannia distinct from Ophiostoma. Stud. Mycol. 55:75?97. Z??iga G, Cisneros R, Hayes JL, Macias-samano J. 2002. Karyology, Geographic Distribution, and Origin of the Genus Dendroctonus Erichson (Coleoptera: Scolytidae). Ann. Entomol. Soc. Am. 95:267?275.            215 Appendices  Appendix A: Supplementary information supporting chapter 3      216 A.1 List of isolates, sampling locations, collection resources and deposited culture collections  Fungal pecies Isolate Other culture collectios Host tree/substrate Beetle associate Location Collection site Grosmannia sp. (Gs clade)  UAMH 4585   Pinus contorta  Dendroctonus ponderosae  Ca nada, British Columbia  Riske Creek    NOF 1280  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Terry Fox    CB H55 UAMH 11153  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H18   Pinus contorta  Dendroctonus po nderosae Canada, British Columbia  Houston    CB H19  UAMH 11348  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H21   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H22   Pinus contorta  Dendroctonus  ponderosae Canada, British Columbia  Houston    CB H41   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H42   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H43   Pinus contorta  Dendroctonus ponder osae Canada, British Columbia  Houston    CB H48   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB H50   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Houston    CB SLA11  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Tweedsmuir Park   CB W14  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Williams lake   CB W6 -1   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Williams lake   CB SLKW1407 UAMH 11150  Pinus con torta  Dendroctonus ponderosae  Canada, British Columbia  Kamloops   CB 200-1-14 UAMH 11151  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Kamloops   CB KDW4  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Kelowna   CB DPKGT1B UAMH 11152  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Kelowna   CB M6  Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Manning Park   CB M3   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Manning Park   CB M11   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Manning Park   CB M44   Pinus contorta  Dendroctonus ponderosae  Canada, British Columbia  Manning Park   CB M46   Pinus contorta  Dendroctonus ponderosae  Canada, British Colum bia Manning Park   UAMH 4818   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Westcastle   NOF 842  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Carbondale    NOF 2893  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Blairmore   CB B5 UAMH 11154  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B6  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B10 UAMH 11155  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B14   Pinus contor ta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B20  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB BW26   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B1   Pinus contorta  Dendroctonus ponderosae  Canada, Albe rta Banff   CB B16   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B17   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B19   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B21   Pinus contorta  Dend roctonus ponderosae  Canada, Alberta  Banff   CB BW22   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB BW27   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB BW28   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB B101   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Banff   CB CHMC3 UAMH 11347  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Cypress Hills    CB CHDSC7 UAMH 11355  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Cypress  Hills    CB CHEBC10  UAMH 11356  Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Cypress Hills    CB CHIHC2   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  Cypress Hills    CB CHMPB8   Pinus contorta  Dendroctonus ponderosae  Canada, Alberta  C ypress Hills      217  Fungal pecies Isolate Other culture collectios Host tree/substrate Beetle associate Location Collection site   CB HV14  Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV30 UAMH 11357  Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV4   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV6   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV8   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV9  UAMH 11358  Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV15  UAM H 11156  Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV20   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV21   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB HV24   Pinus contorta  Dendroctonus ponderosae  USA, Montana  Hidden Valley    CB D1128 UAMH 11359  Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB D1151 UAMH 11360  Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB D1129  Pinus contorta  Dend roctonus ponderosae  USA, Idaho  Hell roaring    CB D1131  Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB D1135   Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB D1137   Pinus contorta  Dendroctonus ponderosae  USA, Ida ho Hell roaring    CB D1146  Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB D1153   Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB HR7 -20(2)  Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    CB HR 8-5   Pinus contorta  Dendroctonus ponderosae  USA, Idaho  Hell roaring    DLS 1061  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    DLS 1037  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 12G13 UAMH 11349  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 23G23  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 55B11 UAMH 11350  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 68B21 UA MH 11361  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 710G16 UAMH 11362  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 11S26  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 11L24  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 23L11  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 23S11  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 24S12  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 24G22  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 24G24  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 55L22   Pinus cont orta Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 55S11   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 55G23   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 56G14   Pinus contorta  Dendr octonus ponderosae  USA, California  Sierra Nevada    CB 67F21   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 68S13  Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 68G12  Pinus contorta  Dendroctonus po nderosae  USA, California  Sierra Nevada    CB 79B34   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 79G31   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB 79L21   Pinus contorta  Dendroctonus ponderosae  U SA, California Sierra Nevada    CB 710G23   Pinus contorta  Dendroctonus ponderosae  USA, California  Sierra Nevada    CB Pa?9 UAMH 11363  Pinus albicaulis  Dendroctonus ponderosae  Canada, British Columbia Nelson    CB Pa?6 UAMH 11364  Pinus albicaulis  Dendrocton us ponderosae  Canada, British Columbia Nelson    CB Pa? 1  Pinus albicaulis  Dendroctonus ponderosae  Canada, British Columbia Nelson    CB Pa? 5   Pinus albicaulis  Dendroctonus ponderosae  Canada, British Columbia Nelson    CB Pa-10  Pinus albicaulis  Dendrocton us ponderosae  Canada, British Columbia Nelson    CB GCA02 UAMH 11365  Pinus strobiformi  Dendroctonus ponderosae  USA, Arizona  Pinale?o Mountains    CB GCA04 UAMH 11366  Pinus strobiformi  Dendroctonus ponderosae  USA, Arizona  Pinale?o Mountains      218 Fungal pecies Isolate Other culture collectios Host tree/substrate Beetle associate Location Collection site   CB GCA05   Pinus strobiformi Dendroctonus ponderosae USA, Arizona  Pinale?o Mountains    CB GCA07   Pinus strobiformi Dendroctonus ponderosae USA, Arizona  Pinale?o Mountains    CB GCA08   Pinus strobiformi Dendroctonus ponderosae USA, Arizona  Pinale?o Mountains    CB GCA13   Pinus strobiformi Dendroctonus ponderosae USA, Arizona  Pinale?o Mountains    CB GCA14   Pinus strobiformi Dendroctonus ponderosae USA, Arizona  Pinale?o Mountains    CB PY2?3b UAMH 11367  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kam loops   CB PY8?8 UAMH 11368  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB PY1 ? 4   Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB PY1(G1 -5)A   Pinus ponderosae Dendroctonus ponderosae Cana da, British Columbia  Kamloops    CB PY2(G1 -4)A  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB PY4 ? 4  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB PY4(G5 -8)D  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB PY8 ? 7  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kamloops    CB KDPT5   Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kelowna    CB KDLPT3   Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kelowna    CB KGBPT2   Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kelowna    DPLKGAPT6   Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kelowna    CB KGW5  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Kelowna  Grosmannia clavigera (Gc clade) ATCC18086  Pinus ponderosae Dendroctonus ponderosae Canada, British Columbia  Cache Creek    CB 15B29C1 UAMH 11369  Pinus ponderosae Dendroctonus ponderosae USA, So uth Dakota Black Hills   CB 15B29C2   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 16B17C3   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 16B24C4   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota Black Hills   CB 23B110C5   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 34B94C6 UAMH 11370  Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 24B166C7   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 24B166C8 UAMH 11371  Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 32B85C9   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 32B85C10   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 34B94C11   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 34B94C12   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 11B8C13   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 11B23C14   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   CB 21B69C15   Pinus ponderosae Dendroctonus ponderosae USA, South Dakota  Black Hills   DLS 15 UAMH 11372  Pinus ponderosae Dendroctonus ponderosae USA, California  Sierra Nevada   DLS 24 UAMH 11373  Pinus ponderosae Dendroctonus ponderosae USA, California  Lassen   DLS 56 UAMH 11374  Pinus ponderosae x P. Jeffreyi Dendroctonus ponderosae USA, California  Lassen   C843  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 554  CMW 15398 / UAMH 11375  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 776  CMW 15785  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 833  UAMH 11376  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 681  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 690  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 771  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 417   Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 651   Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 792  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Sierra Nevada   DLS 108  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Lassen   DLS 120  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Lassen   DLS 122  CMW 15394  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Lassen   DLS 126  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Lassen   DLS 140  Pinus jeffreyi Dendroctonus jeffreyi USA, California  Lassen     219 Fungal pecies Isolate  Other culture collectios  Host tree/substrate  Beetle associate  Location Collection site   DLS 173  CMW 15395 /  UAMH 11377 Pinus jeffreyi  Dendroctonus jeffreyi  USA, California Lassen   DLS 190  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California Lassen   DLS 210   Pinus jeffreyi  Dendroctonus jeffreyi  USA, California Lassen   DLS 235  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California Lassen   DLS 237  UAMH 11378 Pinus  jeffreyi  Dendroctonus jeffreyi  USA, California Lassen   DLS 52  CMW 15783 Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1560   Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1565   Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1575  UAMH 11351 Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1588  UAMH 11352 Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1595   Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1561  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1571  Pinus jeffreyi  Dendroctonus jeffrey i USA, California San Bernardino Mountains   DLS 1581  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains   DLS 1591  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California San Bernardino Mountains Grosmannia aurea CBS 438.69   P inus contorta Dendroctonus sp. Canada, British Columbia Invermere Leptographium longiclavatum CB SLKW1436   Pinus contorta  Dendroctonus ponderosae Canada, British Columbia Kamloops   C 845  Pinus jeffreyi  Dendroctonus jeffreyi  USA, California Sierra Nevada Leptographium terebrantis CB 878 AW 1 -2  Pinus contorta  Dendroctonus ponderosae Canada, British Columbia Kamloops   CB LPKRLT -3  Pinus contorta  Dendroctonus ponderosae Canada, British Columbia Kamloops   C 418  Pinus ponderosae  D. brevicomis  USA, California Sierra Nevada Leptographium wingfieldii CBS 645.89   P. sylvestris  Tomicus piniperda  France,  Orl?ans   CBS 648.89    P. brutia  NA Greece,  Thessaloniki G. clavigera  host tree   Total number    Pinus contorta    92    Pinus albicaulis    5    Pin us strobiformi   7    Pinus ponderosae    32    Dendroctonus ponderosae   136    Dendroctonus jeffreyi/Pinus jeffreyi    30    Total number of isolates   166    Isolates selected for the phylogenetic analysis are bolded.    2 2 0  A.2 Primer sequences and gene de scriptions for 67 G. clavigera loci screened for polymorphisms  Locus no. Gene no. Gene description (abbreviation)  Primers sequence  Target scaffold (Length of target scaffold)  Target gene (Length of target gene)  Length of sequence fragment  ATGTGCAGGGTGGCGAGCGAA 5 49  1  1 ABC transp orter 1 (ABC)  GAATACCGCTCCGCTCGCACA GCLVSC_144 (2,496,812)  GLEAN_6117 (6,801)    GTTGACGTGCTTGATGAGG  857  2  2   ABC transporter  2  AGACGGACGAATAGGGAG  GCLVSC_82 (1,001,357)  GLEAN_1593 (2,205)    AGGA AGAGCGAAGAGGCA  810  3  AGGATGGGCTGGTACGGA  GCLVSC_113 (3,089,927)  GLEAN_3921 (3,354)    AATCCCGATGCTTGCCCT  804  4  CTGCCTCTTCGCTCTTCCT  GCLVSC_113 (3,089,927)  GLEAN_3921 (3,354)    AAACTCCGCCTTCAGCCA  830  5  3 ABC transporter 3  ATCACCGCTTCTCCACCA  GCLVSC_113 (3,089,927)  GLEAN_3921 (3,354)    CTCTGTTTGCATCACTTCTC  865  6  ATCATGGCTCTACCTCCT  GCLVSC_144 (2,496,812)  GLEAN_1678 (4,563)    GTGTACCCCAAAGCCACC  770  7 CAAAATCGCCACAGTCCTC  GCLVSC_144 (2,496,812)  GLEAN_1678 (4,563)    TCTCGTTGTCTATGTCG CT  1,297  8  4  Lipases 1  AAAACTCCTGCCCAAACC  GCLVSC_144 (2,496,812)  GLEAN_1678 (4,563)    CGATGTTTTGGCCGTTGT  872  9 5  Lipases 2  CTCTGCGTGTCCTGGTTT  GCLVSC_113 (3,089,927)  GLEAN_4044 (2,286)    ATGCGCCTCCACTTGTCC  720  10  ACTTCTACACCGACTTCCTCCT  GCLVSC_173 (2,341,8 15)  GLEAN_5805 (2,748)    CACACGGACCAACGACGA 1,125  11  6 Lipid acyl hydrolase (LAH)  CTCTCCTGCCCCTCTTCTC GCLVSC_173 (2,341,815)  GLEAN_5805 (2,748)    GGAGACAGCGGGTATAGAG  857  12  7 Multidrug facilitator superfamily transporter 1  CAAGCAGATATGGAAACGGA  GCLVSC_89 (1,124,797)  GLEAN_4905 (516)    TCCCAAACAAAC AGGCCA  769  13  ATCCACACACAGCATCAG  GCLVSC_132 (1,063,191)  GLEAN_851 (1,695)    GTGGTTCTCAGTCTTCTCGT  641  14  8  Multidrug facilitator superfamily transporter 2  TGATGCTGTGTGTGGATGG  GCLVSC_132 (1,063,191)  GLEAN_851 (1 ,695)    GACATTGTAGAGGGCAGC 8 47  15  AGATGGGAGGTTGGAGAG GCLVSC_113 (3,089,927)  GLEAN_8113 (1,641)    AGTAGAACACCGCCGACAG 792  16  9 Cytochrome P450 1 (P450 I)  CCGACCAAACACACCGCA GCLVSC_113 (3,089,927)  GLEAN_8113 (1,641)    TGCAGCAATGGGACCGGATGA 710  17  10  Cytochrome P450 2 (P450 II)  TCGTCACGTTCTCCCAGCGCT GCLVSC_173 (2,341,815)  GLEAN_1091 (1,578)    TGAACGAGACGCTGCGTCTGATG  560  18  11 Cytochrome P450 3  TGCAGCTTCAGCGTCAAC AGGGT  GCLVSC_167 (1,996,734)  GLEAN_4745 (2,010)    CTTTGTCATCCACCCGCT  575  19  CATCCATCTCCTCGGCCT  GCLVSC_132 (1,063,191)  GLEAN_5485 (1,569)    GATCGCAGACATTGGCCT  818  20  12  Cytochrome P450 4  CCTCCTCCACACCTTCAC  GCLVSC_132 (1,063,191)  GLEAN_5485 (1,569)    TAAGGAAAGGGAGGGCGGT 913  21  13  Metallo-peptidase (MPEP)  TGGGTGCGTGATGAGCGA GCLVSC_113 (3,089,927)  GLEAN_8241 (1,485)        221 Locus no. Gene no. Gene description (abbreviation) Primers sequence Target scaffold (Length of target scaffold) Target gene (Length of target gene) Length of sequence fragment  ATTCCCCTCCCCTACTCC 780 22   CTTCCATGTCCTCCTTCC GCLVSC_113 (3,089,927) GLEAN_8241 (1,485)   TGATTCGACTTTCCCCCT 886 23 CGTCGAACACAAACTCCT GCLVSC_144 (2,496,812) GLEAN_1886 (2,325)   GGAGTTTGTGTTCGACGAG 975 24 14 Anthranilate synthase (TRPG) GAATGACAAGGCTATGAAGGGA GCLVSC_144 (2,496,812) GLEAN_1886 (2,325)   GCTGTCATGTCTGGCGTGTCCA 595 25 15 beta-tubulin TCTTCGCCATACGCCTCCTCG GCLVSC_141 (93,906) GLEAN_4834 (1,341)   TCCAGACGAACCTGGTGCCGT 641 26 16 alpha-tubulin CAGGCGTCATCGAGCAAGCGA GCLVSC_108 (1,834,804) GLEAN_6984 (1,353)   TCACGCCCACCGTTACCGACA 752 27 17 40S ribosomal protein S3 (40SRP) TGGAAATGGTCGGTGCCGAGGT GCLVSC_140 (4,437,291) GLEAN_2722 (795)   CGGTCGCCCGCTCTACATTGA 570 28 18 Phosphatidylinositol transferase  (PLT) CTCAGCCTCTAAGCCGTTGCCT GCLVSC_168 (797,636) GLEAN_6502 (1,017)   TGGCGGCAAGATCACGGAGGA 600 29 19 Acetyltransferase  CCATTCGCTCGTGCTCTCGCTT GCLVSC_140 (4,437,291) GLEAN_2685 (1,404)   GAAGGGTGGTGCCAAGAAGGTGA 549 30 20 Glyceraldehyde 3-phosphate dehydrogenase  TCTTCCGCCCCTCCCTCGTAGT GCLVSC_108 (1,834,804) GLEAN_7049 (954)   CGTGACGAAGAGCGGCATCGA 489 31 21 Alcohol dehydrogenase TGCAGCACGAGCTTGCCGAACT GCLVSC_140 (4,437,291) GLEAN_2576 (1,170)   TGCTGTCGAGAACTGGAGGCGT 568 32 22 Lysophospholipase  CGGCAGGACCTGGAACAGGAA GCLVSC_156 (1,463,716) GLEAN_6592 (726)   TCGAGGAGAACGCCGAGCTGA 666 33 23 putative septin TCGAGCGAACGCTGGTACCACT GCLVSC_160 (595,886) GLEAN_3045 (1,137)   TCACTGCCACCACCGTCTTTCCT 648 34 24 Thiol-specific antioxidant  GCACGAAGAAAGCACGCAGGA GCLVSC_132 (1,063,191) GLEAN_5524 (642)   CATGCGTGTTCGGCCTTCTGCT 726 35 25 Superoxide dismutase  TGCCGACCCAACTGGCTTAGCCT GCLVSC_97 (2,272,097) GLEAN_5060 (687)   AGATCGAGTCCACGTCACCGCA 758 36 26 Aspartic proteinase GGTTGCGCCATTGAAGACGACA GCLVSC_160 (595,886) GLEAN_3069 (1,263)   CGTCGTTCCTGCCGCTGATTGA 515 37 27 Polyketide synthase 1  ACCGTCCTCCACTCCGATCAGCT GCLVSC_144 (2,496,812) GLEAN_1845 (6,522)   TGGAAACCGAGGACAGGCGAACT 564 38 28 Polyketide synthase 2 ACGCCGGTCTCACGAGAAATCCA GCLVSC_160 (595,886) GLEAN_3011 (6,633)   TGCCGACAAGGTGGCCAAGTTC 687 39 29 Peroxisomal-coenzyme A synthetase (PCAS) GCGCAGCGCAACATTGACGACT GCLVSC_167 (1,996,734) GLEAN_0350 (1575)   CTCTTCTTTGCCGGCCTTGCTGT 667 40 30 Common fungal extracellular membrane protein 1  (CFEM I) CGCAACGCAAACGCCAGAAGA GCLVSC_173 (2,341,815) GLEAN_5868 (582)   GCGTCCATTGATCGGCGTGATGT 491 41 31 CFEM 2  (CFEM II) AACCGCCAACATGGCAACGG GCLVSC_156 (1,463,716) GLEAN_3507 (564)   TGTTCCGTGCCTTCGATGCCCT 824 42 32 CFEM 3 AGGTATCGCTGACCACATCCCGA GCLVSC_168 (797,636) GLEAN_1411 (921)   CTCTCTTTCGTCCTGGCTCTCGG 260 43 33 CFEM 4 CAAGACTCCGAGTCCGCCATTG GCLVSC_161 (1,111,432) GLEAN_4535 (366)   TGCTTCTGCTGCTCTGCGTGGT 1,229 44 34 CFEM 5 TGGACCGACGTAGTTGTGGCCGT GCLVSC_132 (1,063,191) GLEAN_827 (1,551)       222 Locus no.  Gene no. Gene description (abbreviation)  Primers sequence  Target scaffold (Length of target scaffold) Target gene (Length of target gene) Length of sequence fragment  ATCGCACCCTTGACTACCGCCA 608 45 35 CFEM 6 GGTGACAGAAAAGAGGACGCGCT GCLVSC_142 (425,798) GLEAN_5021 (627)   TCACGCCATACCAGATCC 833 46 AAACTCCCTCCTTCATCCA GCLVSC_179 (1,213,309) GLEAN_4283 (3,915)   AGGAGTACGAGGCCACAG 607 47 36 Anonymous 1 GCCACAGAACAGATAACGA GCLVSC_179 (1,213,309) GLEAN_4283 (3,915)   CCATCCGACAACAACACC 871 48 37 Anonymous 2 TCCTCCACGTCATCCTCCA GCLVSC_140 (4,437,291) GLEAN_2807 (603)   CTCTATTCCCCTTCAACTCTCTC 825 49 38 Anonymous 3 CTTCTCCCACCCCTTGAC GCLVSC_144 (2,496,812) GLEAN_1731 (765)   TTTCGTGTTGATCGCCTG 594 50 39 Anonymous 4 CCACCACACAATTGCTCC GCLVSC_161 (1,111,432) GLEAN_3159 (717)   CAAAGCAACCCGCAAAACAC 878 51 40 Anonymous 5 ACGTAGCGACCGACAACC GCLVSC_108 (1,834,804) GLEAN_6881 (1,062)   TCACATACCGGAGCCACCA 820 52 41 Anonymous 6 AAAACAAAGAACGGCACCCA GCLVSC_144 (2,496,812) GLEAN_1625 (765)   GAGGGAAGGAAAAGGGGA 629 53 42 Anonymous 7 CTGGCTAAATCTCTCTTCGT GCLVSC_140 (4,437,291) GLEAN_2380 (342)   CATCCTCACCCTTCAACC 810 54 ACTTAAACCACGCCAGAC GCLVSC_140 (4,437,291) GLEAN_7708 (546)   CTTCGCCCAACGGTACCA 912 55 43 Anonymous 8 CAGCCAAAAAAGGGGGCCA GCLVSC_140 (4,437,291) GLEAN_2314 (1,362)   ATCTTCCTCTCTTTCCACAC 740 56 44 Anonymous 9 CACTTTGTTTCCGCTCCT GCLVSC_173 (2,341,815) GLEAN_1181 (618)   TATTGTTGAGATTGGCCCTG 830 57 TTGGATATCTTGTGCTGCTG GCLVSC_113 (3,089,927) GLEAN_4069 (5,592)   CCATCGCTGCCAAATGTC 810 58 CTCCTGGCACGCTATCTC GCLVSC_113 (3,089,927) GLEAN_4069 (5,592)   AGGTGTCGTTGTTGATGTGG 730 59 45 Anonymous 10 GTTTGTGCTGGTGTCGGT GCLVSC_113 (3,089,927) GLEAN_4069 (5,592)   ACCGCATTTTCCTCATCAC 900 60 CCCTCTCACAATCTCCCA GCLVSC_113 (3,089,927) GLEAN_7855 (6,297)   CTTCTTCGCGTTCTTGCT 790 61 GGCATTTCTGGCTTGGGT GCLVSC_113 (3,089,927) GLEAN_7855 (6,297)   AGAGGGTGAAGTGAAAAGG 907 62 TGAAGAAGGGTGACGAGG GCLVSC_113 (3,089,927) GLEAN_7855 (6,297)   TATCCCAGTTCGCCAGCA 783 63 46 Anonymous 11 AAGAACAACCCGAAGGAG GCLVSC_113 (3,089,927) GLEAN_7855 (6,297)   CACGACGACGAACTCCTCTCCCA  458 64 4 7  Anonymous 12 (Anonymous I)  CAGGATGCCCTCGGCCTCTAAC  GCLVSC_167 (1,996,734) GLEAN_4597 (381)   TGCCAGACTGGTCCACATCTGCA  805 65 4 8  Anonymous 13 (Anonymous II)  ACGCCGGCAAGACCTACACCA  GCLVSC_140 (4,437,291) GLEAN_7505 (432)   TGGTCCACAGTACCATCCCGTCA 661 66 49 Anonymous 14 TGCATGTGCCCACGTCCACGA GCLVSC_113 (3,089,927) GLEAN_3737 (603)   ACCGCGAAGATGCCAGGCAAACA 631 67 50 Anonymous 15 GAGAACGCAACAGAACGCAGCCA GCLVSC_173 (2,341,815) GLEAN_5862 (870)       223 Genes and primer sequences selected for further phylogenetic and population genetic analyses are bolded  TreeBASE URL for concatenated alignment of the 67 loci (49,853 base pairs): http://purl.org/phylo/treebase/phylows/ study/TB2:S11355      224 A.3 Polymorphism summaries and diversity indices within the two monophyletic clades in G. clavigera  Number of single nucleotide polymorphisms Diversity indices a Gene predictions Species clade Total Singleton Noncoding Synonymous  Replacement  Indels (length in base pairs) Number of haploty pes Gene diversity (H) Diversity ! (10-3) Diversity " (10-3) Gs   2 1 0 1 1 0 3 0.27 0.37 0.63 40SRP  Gc  1 1 0 1 0 0 2 0.09 0.12 0.37 Gs   0 0 0 0 0 0 1 0.00 0.00 0.00 alpha tubulin Gc  0 0 0 0 0 0 1 0.00 0.00 0.00 Gs   1 1*  0 0 1 0 2 0.05 0.09 0.43 ABC  Gc  0 0 0 0 0 0 1 0.00 0.00 0.00 Gs   6 1*  0 1 5 0 4 0.65 1.30 0.73 TRPG  Gc  1 1 0 1 0 0 2 0.09 0.05 0.14 Gs   2 1 0 1 1 0 3 0.14 0.09 0.28 MPEP  Gc  2 1*  0 1 1 0 3 0.51 0.33 0.33 Gs   3 1*  0 1 2 0 4 0.38 0.26 0.44 P450 I  Gc  3 2*  0 0 3 0 4 0.39 0.27 0.33 Gs   2 1 0 2 0 0 3 0.23 0.39 0.66 P450 II  Gc  3 1 0 2 1 0 4 0.50 1.01 1.16 Gs   0 0 0 0 0 0 1 0.00 0.00 0.00 LAH Gc  1 0 0 0 1 0 2 0.24 0.22 0.24 Gs   5 1 2 3 0 0 4 0.58 2.53 1.77 CFEM I Gc  1 0 0 0 1 0 2 0.42 0.62 0.41 Gs   4 0 2 1 1 0 4 0.73 3.78 1.92 CFEM II Gc  5 0 1 2 2 0 4 0.73 3.60 2.81 Gs   2 1 0 1 1 0 3 0.50 0.91 0.83 LPL  Gc  0 0 0 0 0 0 1 0.00 0.00 0.00 Gs   2 0 0 2 0 0 3 0.43 0.80 0.83 PLT  Gc  1 1 0 1 0 0 2 0.09 0.16 0.48 Gs   1 1 1 0 0 0 2 0.05 0.07 0.34 PCAS  Gc  2 2 2 0 0 0 3 0.17 0.27 0.80 Gs   0 0 0 0 0 0 1 0.00 0.00 0.00 Anonymous I  Gc  1 1 1 0 0 0 2 0.09 0.20 0.60 Gs   3 1 2 1 0 2 (3) 3 0.39 0.96 0.88 Anonymous II  Gc  3 0 1 2 0 0 4 0.65 0.99 1.02 Total Within G. clavigera 62 86 18 18 33 35 2 (6) 58 0.99 2.00 1.37 Total within Gs  40 33 10 7 14 12 2 (6) 36 0.99 0.68 0.59 Total within Gc  22 24 10 5 10 9 0 22 1.00 0.41 0.50 * Coding regions   a H, haplotype/gene diversity which corresponds to the probability that two randomly chosen individuals are different at a chosen locus; !, nucleotide diversity based on average number of nucleotide differences per site between two sequences (Nei 1987); ", nucleotide diversity based on the number of segregating sites (Watterson 1975)        225 A.4 Gs morphology compared with those of the G. clavigera holotype  Fungal isolates Grosmannia sp. (Gs clade) G. clavigera  (Gc clade)a Host tree Pinus contorta  Pinus ponderosa  Insect associate Dendroctonus ponderosae  Dendroctonus ponderosae  Anamorph (asexual reproduction) Mononematous Leptographium and synnematous conidiophores Mononematous Leptographium and synnematous conidiophores Conidiophore length including conidiogenous apparatus (?m) 80?1150 100?1040 Conidium shape Clavate and cylindrical to oblong Clavate and cylindrical to obclavate Conidium size (clavate ?m) / (others ?m) 16.5?62 ! 4?6 / 14.8 ! 2.3 35?68 ! 4.2?5.6 / 12.6 !  3.8  Teleomorph (sexual reproduction) Grosmannia Grosmannia Ascocarp shape (color) spherical cleistothecia b  (black) spherical cleistothecia (black) Ascocarp size 245?570 250?640 Ascospore shape reniform (cucullate in side view) reniform (cucullate in side view) Ascospore size 3.5?5.0 ! 2.5?4.0 3.5?5.6 ! 2.8?4.2 a Observations by Robinson and Davidson (1968) and Six and Paine (1997)  Lee et al. (2003) have observed ascocarps with short necks ranging from 20?65 ?m             2 26  A.5  Haplotype network   Traditional phylogenetic methods that produce bifurcating gene trees cannot accurately portray genealogical relationships at the intraspecific level, due to different population processes such as recombination and/or pre sence of ancestral alleles. Therefore, for each of the 15 gene datasets, we generated parsimony networks of G. clavigera haplotypes using TCS 1.21 (Clement et al. 2000). This method estimates the unrooted tree and provides a 95% plausible set for parsimoni ous relationships between all haplotypes. Network approaches can account for homoplasy by introducing loops where recombination or frequent mutation could have occurred in the history of the genealogy.           2 2 7        228 Appendix B: Supplementary information support ing chapter 4  B.1  Main features of primary genome sequence data  StrainID  Sequencing Platform Read Length  Low Kmer  High Kmer  GsB2 Illumina Genome Analyzer II 50 25 35 GsB3 Illumina Genome Analyzer IIx 75 31 61 GsA1 Illumina Genome Analyzer IIx 75 31 61 GsA2 Illumina Genome Analyzer IIx 75 31 61 GsA3 Illumina Genome Analyzer II 50 25 35 GsM1 Illumina Genome Analyzer II 50 25 35 GsC1 Illumina Genome Analyzer IIx 75 31 61 GsC2 Illumina Genome Analyzer IIx 75 31 61 GcB1.a Illumina Genome Analyzer II 50 25 35 GcB1.b Illumina Genome Analyzer IIx 75 31 61 GcC1 Illumina Genome Analyzer II 50 25 35 GcC2 Illumina Genome Analyzer IIx 75 31 61      229  B.2  Assembly statistics of Grosmannia genomes   de novo contigs   after ordering and orientating de novo contigs against the refernece sequence b   Aligned contigs  c Strain a Assembly Size  Number of Scaffolds N50  N90  Gaps  Percent Gaps   Assembly Size  Number of Scaffolds N50  N90  Gaps  Percent Gaps    bp length % GsB2  27,738,425  1,348 95,703  22,052  40,805  0.15   28,453,2 82 91  2,219,886  787,389  2,460,073  8.6   27,931,310  101 GsB3  30,914,973  2,924  163,110  10,928  15,386  0.05   29,213,916  140 1,984,429  594,737  940,778  3.2  30,794,128  100 GsA1  29,882,447  2,597  159,048  10,755  12,448 0.04  29,066,433  144 2,242,497  588,345  1,068, 256  3.7  29,957,163  100 GsA2  29,997,851  2,327 111,753  8,908  20,875  0.07  28,802,836  143 2,220,345  594,681  919,606  3.2  29,952,967  100 GsA3  27,996,749  1,376  92,960  19,371  47,237 0.17  28,153,243  85  2,260,542  655,482  2,617,547  9.3   28,041,899  100 GsM1  28,888,169  1,414 102,139  21,466  49,831  0.17  27,600,675  96  2,166,132  756,581  2,504,286  9.1   28,985,091  100 GsC1  31,128,695  2,942  177,573  9,769  31,673  0.10  28,880,088 146  1,981,110  594,384  1,338,407 4.6   30,521,551  98  GsC2  32,243,025  2,397  221,770 13,462  20,957  0.06   28,560,494  146  1,834,826  588,324  1,217,577  4.3  31,761,788  99  GcB1.a  28,694,637  4,000 32,813 6,628  28,776  0.10  28,615,521  127 2,180,445  581,727  2,158,396  7.5   27,719,482  97  GcB1.b  30,843,415  1,409  133,367  17,320 459  0.00  29,424,681  123 1,983 ,195  591,692  1,320,203 4.5   29,143,054  94  GcC1  28,399,467  1,353  98,789  20,245  42,002 0.15   27,978,235  70 2,216,111  744,378 2,583,449  9.2   27,851,123  98  GcC2  32,445,864  2,787 155,238  6,569  20,119  0.06    28,845,018  136  1,993,212  591,042  1,263,041  4.4   30,323,732 93  a IDs, "a" and "b" are results from two independent sequence lanes for the same strain.   b The reference genome published by DiGuistini et al. 2011   c The total length and percentage of de-novo contigs aligned to the reference sequence over at least 100 bp with 95% identity. For strains with multiple contigs mapped to the same location in the reference genome, the total length and percentage of aligned contigs are higher than original de novo assembly values, representing a measure of redundancy and divergence from the reference genome     230 B.3  Summary of genBlastG output used for gene annotations of each Grosmannia draft genome, with pairwise homology (PID) with reference gene models    (Data are available upon request)     231 B.4  Total number of sequence  reads and filtering steps used for SNV a calling  Total number of Paired -end and single-end reads after removing Fungal strain IDs b Raw paired-end reads  (C ) Low?quality reads Duplicate reads   (%)  Total bp Estimated coverage UAMH 11150  GsB1 (control) 41,327,658  (50) 31,374,695 19,483,575  ?  7,523,362  (26) 2,092,073,040 70! UAMH 11153  GsB2 18,812,843  (50) 11,248,893 7,089,534  ?  2,317,892   (27) 742,363,200 25! UAMH 11348  GsB3 25,340,644  (76) 14,919,663 8,110,381  ?  3,200,348   (35) 1,378,898,810 47! UAMH 11353  GsA1 35,991,059  (76) 18,087,186 10,091,552  ?  3,643,716  (34) 1,691,704,220 58! UAMH 11354  GsA2 37,723,693  (76) 13,259,161 5,973,601  ?  2,517,982   (45) 1,027,028,064 35! UAMH 11347  GsA3 16,682,452  (50) 10,109,303 6,492,213  ?  2,053,087  (26) 676,688,085 23! UAMH 11156  GsM1 14,617,945  (50) 9,092,931 5,931,623  ?  1,668,522   (26) 608,929,560 21! UAMH 11349  GsC1  33,385,453  (76) 18,585,344 8,446,641  ?  3,848,853   (44) 1,472,691,585 51! UAMH 11350  GsC2  33,834,154  (76) 19,419,436 8,090,204  ?  3,814,108   (49) 1,419,610,636 49! ATCC 18086  GcB1.a 19,701,172  (50) 5,851,822 4,307,999  ?  833,338    (19) 425,220,120 15! ? GcB1.b 32,988,736  (76) 19,737,773 10,016,836  ?  4,232,013  (38) 1,722,863,635 59! ? GcB1.ab 52,689,908 25,589,595 14,324,835  ?  5,065,351  (34) 2,148,083,755 74! UAMH 11351  GcC1  17,815,964  (50) 10,334,042 6,606,133  ?  2,075,845   (26) 687,964,995 24! UAMH 11352  GcC2.a  33,250,787  (76) 14,324,332 6,807,253  ?  2,752,816   (43) 1,162,079,862 40! ? GcC 2.b 34,219,918  (76) 17,494,252 7,647,398  ?  3,452,387   (46) 1,331,049,993 46! ? GcC2.ab  67470705 (76) 31,818,584 14,454,651  ?  6,205,203  (48) 2,326,984,246 80! a We use the term SNV instead of the common term SNP (Single Nucleotide Polymorphism) bec ause SNPs are normally defined relative to a population and imply a minimum minor allele frequency whereas we are interested in finding all sequence variants that do not match the reference genome sequence, regardless of their frequency in the population.  b IDs, "a" and "b" are results from two independent sequence lanes for the same strain, "ab? results from two sequence runs combined for the same strain. The estimated coverage are based on filtered reads mapped to the slkw1407 reference genome sequence, which is ~ 29.1 Mbp after excluding gaps.   c nucleotide bases      232 B.5  Primer sequences used in the SNV validation and in the phylogenetic and population genetic analyses   Primer sequence 5' ?? > 3'  sequence length in bp*  Phylogenetic information  a Gene Forward  Reverse Total Exon  Intron GenBank accession no. Gene description (abbreviation)  Total  characters Variable sites Parsimony informative  MP trees CGAACGCCTCTGGCTCTCCATTG GCCTGCCGGAAATGTGACGTTG 1,258  3 2 xxxxxx CMQ_5562  ATCCGTGGTCCAGCAGTC AGCACGACCTTCTCCAGC 1,722 1 0 xxxxxx ABC transporter (subfamily C) 2,843  32 18 1 AGACCTCTCATGGCCTACTGCG AACATGCGGCAGCAGTCCCAAC 1,064  4  4  xxxxxx CMQ_6993.6634  ACACGGGCGAGCTTCTCAACAG TTCCACCACCTCCCAGAGTCCCA 1,182 3 2 xxxxxx ??  2,140  52  23 1 CGCGAGCTTTGGTGTTTCTGCC ACCAGCAGGCTGACTAGCGACA 1,074  1 0 xxxxxx CMQ_6965  TTCATTGGAGAGCGAGGCGCTG TGCGCAGCCAGTCGACCAATAC 943  2 1 xxxxxx ??  1,826 36 18 1 CMQ_861  GGCCAGCAACGGCATCTTTGAC TGCCACAGACCAAAGCCTGGAC 563  2 1 xxxxxx ABC transporter (subfamily G) 494  19 11 3 CMQ_4184  AAGCACGCCCGCTATGCACTC TGCTGACAGTTGTTGGCTGCCG 1,087 1 0 xxxxxx ??  1,055  18 8 1 TAAGGAAAGGGAGGGCGGT GGAGTAGGGGAGGGGAAT 913 1 1 xxxxxx CMQ_3826  TCGCTCATCACGCACCCA CTTCCATGTCCTCCTTCC 780 1 0 xxxxxx Metallo-peptidase 1,585  18 5  1 TGTGTCCAAGGCATTCCCCGAC CACAAACAGCGGCGTCGAGTTG 1,414  1 0 xxxxxx CMQ_5095  AAGGTGCTCCTGATGATGCGGC AATGGCAGCCGATGTGGCAGAG 1,008 2 1 xxxxxx Polyketide synthase  1,786 44  27 1  ACAGTCCGGAGAGCGTGACCATC AGGTTGGCGAACAAGTCCTGGG 1,213 1 0 xxxxxx CMQ_5323  AGTCCTTCTCTCGCGGCATCTCC AACCAGCATGTTCCGCACCTCG 1,331 2 1 xxxxxx Polyketide synthase  1,539  47  25  1 CMQ_6740  TTTGGCATCTCCAAGCCCTGCG GCGTCATCCAGACGGTCATCAGC 1,191 2 1 xxxxxx Flavoprotein monooxygenase  1,040  30 8 1 Nine -gene combined dataset          To be submitted   14,308  296 143  40  a Phylogenetic information for each and combined nine gene aligned dataset sequenced from additional Gs and Gc strains including the outgroup taxa       233 B.6 Grosmannia gene-model summaries  Fungal IDs Homologous genes 100% 99% 98?96% 95?90% 89-70% PID<70% Genes with no hit duplicates/paralogs GsB2 8,129 5,715 1,486 537 238 153 103 80 49 GsB3 8,222 5,855 1,521 495 219 132 63 27 46 GsA1 8,155 5,808 1,440 534 240 133 78 79 46 GsA2 8,107 5,345 1,428 738 382 214 129 76 42 GsA3 7,988 5,522 1,485 583 242 156 139 185 61 GsM1 7,876 5,399 1,493 587 248 149 191 245 58 GsC1 8,018 5,069 1,974 585 239 151 154 140 91 GsC2 7,956 4,885 2,073 592 257 149 146 210 50 GcB1 8,198 3,108 3,747 851 297 193 89 27 41 GcC1 7,973 2,907 3,560 966 326 210 193 150 64 GcC2 8,079 2,959 3,624 931 359 202 140 97 70 Average 8,064           130 120        234 B.7 McDonald-Kreitman (MK) test results and the mean Gs-Gc pairwise rate of protein-coding divergence (dN/dS)   (Data are available upon request)   B.8 PAML ?site-model? test of positive selection for 1,213 Gs-Gc orthologous genes   (Data are available upon request)                2 35  B.9 Variants identified using the published Grosmannia genome as the reference sequence  Single nucleotide variations (SNV) Fungal IDs All Ts/Tv  Intergenic  Intronic  Flanking  Synonymous coding Nonsynonymous coding  Nonsynoymous    stop gain ?  stop lost Indels GsB1 (control) 1,796   297  202  526  223  540  03 ?  05  GsB2  10,242  2.7  3,866  609 2,458  1,576  1,706  23 ?  09 3,117  GsB3  12,196  3.1  5,573  630  2 ,521  1,607  1,837   23 ?  10  3,240  GsA1  10,405  3.1  4,405  567  2,333  1,479  1,602  16 ?  09 3,170  GsA2  10,332  3.0  4,482  589 2,333  1,338  1,568 15 ?  10 3,154  GsA3  9,941  2.8  3,919  616 2,384  1,442  1,565 14 ?  09 3,095  GsM1  10,139  2.8  3,804  677  2,474  1,566 1,596 17  ?  10 3,082  GsC1  19,098 3.4  8,867  980 4,100  2,453  2,668  22 ?  13  3,781  GsC2  20,521  3.4  9,362  1,110 4,503  2,651  2,858  35 ?  09 3,882  Average 12,859  3.0  5,535  722  2,888  1,764  1,925    3,315  GcB1.a  58,663  3.9  19,918 3,921  15,819 9,754  9,150 103 ?  33  6,751  G cB1.b 59,926  3.8  21,169  3,918  15,806 9,763  9,170  103 ?  33  6,853  GcB1.ab  60,435  3.9  21,617  3,925  15,832  9,782  9,179  103 ?  33  6,862  GcC1  60,808 3.8  20,968  3,989  16,318  9,895 9,522  125 ?  29  6,814  GcC2.a  63,204  3.8  23,883  3,915  16,230  9,694  9,375  117 ?  29  6,934  GcC2.b  63,219  3.7  23,886  3,913  16,230  9,700  9,382  118 ?  29  6,951 GcC2.a b 63,294  3.7  23,945  3,916  16,234  9,705  9,386  118 ?  29  6,957  Average 61,512  3.8  22,177  3,943  16,128  9,794  9,362    6,878  Total  103,430   42,880 5,826 24,589 14,889 15,040 226 ? 36 9,906 The reference genome published by DiGuistini et al. 2011      236  B.10 Gene models containing intra- and interspecific stop-codon variants    Scaffold   Stop codon    RNA evidence a Gene Gene description   ID from to   gain lost position   RNAseq and/or EST annotation SNP class CMQ_7030  salicylate hydroxylase  SC_105  16,604  20,349   X  - 19,630     exclusive to Gs  CMQ_6551  siderophore biosynthesis protein  SC_108  1,570,163  1,573,981   X  - 1,573,138   Yes  confirmed fixed between Gs and Gc CMQ_6582  autophagy protein  SC_108  1,362,950  1,364,821   X  - 1,363,994   Yes  confirmed fixed between Gs and Gc CMQ_6740 flavoprotein monooxygenase   SC_108 1,030,358 1,032,964  X  - 1,031,164  Yes confirmed exclusive to Gc from JP CMQ_6988  tol-like protein  SC_108  1,250,926  1,254,937   X  - 1,254,412   No nc exclusive to Gs  CMQ_3349  arginyl-tRNA synthetase   SC_113  2,283,134  2,284,072   - X  2,284,072   Yes  confirmed fixed between Gs and Gc CMQ_3907  cytochrome p450 monooxygenase   SC_113  2,304,173  2,308,001   X  - 2,037,15 0  Yes  confirmed fixed between Gs and Gc CMQ_3907  cytochrome p450 monooxygenase   SC_113  2,304,173  2,308,001   X  - 2,037,706   Yes  confirmed exclusive to Gc from JP CMQ_3795  siderochrome-iron transporter  SC_113  1,184,097  1,186,007  X  - 1,184,320   Yes  confirmed fixed between Gs and Gc CMQ_3862  tyrosinase  SC_113  1,090,713  1,093,110   X  - 1,091,358   Yes  confirmed fixed between Gs and Gc CMQ_3890  pre-mRNA splicing factor   SC_113  388,304  389,549   - X  388,307   Yes  confirmed fixed between Gs and Gc CMQ_3894  major facilitator superfamily transporter  SC_113  721,228 722,896  X  - 722,513   Yes  confirmed exclusive to Gs  CMQ_3894  major facilitator superfamily transporter  SC_113  721,228 722,896  X  - 722,575   Yes  confirmed fixed between Gs and Gc CMQ_4004  esterase/lipase   SC_113  1,182,730  1,183,390   X  - 1,182,746   Yes  confirmed fixed between Gs and Gc CMQ_4014  stomatin family protein  SC_113  946,286  949,981   X  - 948,073   Yes  confirmed fixed between Gs and Gc CMQ_4014  stomatin family protein  SC_113  946,286  949,981   X  - 948,145   Yes  confirmed fixed between Gs and Gc CMQ_4014  stomatin family protein  SC_113  946,286  949,981   X  - 948,163   Yes  confirmed fixed between Gs and Gc CMQ_4097  extracellular scp domain containing protein  SC_113  851,732  852,407   X  - 852,126   Yes  confirmed fixed between Gs and Gc CMQ_4102  duf1680 domain containing protein  SC_113  1,009,005  1,011,413   X  - 1,010,651   Yes  confirmed exclusive to Gc from JP CMQ_4148  conidiophore development protein hyma  SC_113  1,266,837  1,270,034   X  - 1,266,939   Yes  confirmed fixed between Gs and Gc CMQ_1788  amino acid permease  SC_124  96,510  98,566   X  - 96,879  Yes  nc exclusive to Gs  CMQ_8025  class 5 chitinase 1   SC_125  37,549  43,430   X  - 39,034   Yes  nc exclusive to Gc from JP CMQ_8025  class 5 ch itinase 1  SC_125  37,549  43,430   X  - 39,035   Yes  nc exclusive to Gc from JP CMQ_8025  class 5 chitinase 1   SC_125  37,549  43,430   X  - 39,920   Yes  nc exclusive to Gc from JP CMQ_8025  class 5 chitinase 1   SC_125  37,549  43,430   X  - 41,011   Yes  nc exclusive to Gc from JP CMQ_8027  FAD -binding domain containing protein  SC_125  7,180 8,998  X  - 8,569   Yes  confirmed exclusive to Gc from JP CMQ_5120  ankyrin repeat-containing protein  SC_132  856,946  865,104   X  - 864,916   Yes  confirmed fixed between Gs and Gc CMQ_5120  ankyrin repeat-containing protein  SC_132  856,946  865,104   X  - 865,045   Yes  confirmed fixed between Gs and Gc CMQ_5339  sodium/phosphate symporter   SC_132  246,066  250,738   X  - 249,907   Yes  confirmed fixed between Gs and Gc CMQ_5339  sodium/phosp hate symporter  SC_133  246,066  250,738   X  - 249,904   Yes  confirmed fixed between Gs and Gc CMQ_5356  steroid monooxygenase  SC_132  332,989  334,789   X  - 334,270   Yes  confirmed exclusive to Gc from JP CMQ_5374  dynamin family protein  SC_132  485,507  487,91 6  X  - 487,815   Yes  confirmed fixed between Gs and Gc CMQ_1184  intracellular serine protease  SC_140  4,027,986  4,031,141   X  - 4,028,662   Yes  confirmed fixed between Gs and Gc CMQ_1274  ankyrin repeat protein  SC_140  3,535,897  3,538,054   - X  3,535,900   Yes  confirmed fixed between Gs and Gc CMQ_1290  nacht and tpr domain containing protein  SC_140  261,976 266,833   X  - 262,855   Yes  nc exclusive to Gc CMQ_1290  nacht and tpr domain containing protein  SC_140  261,976 266,833   X  - 263,124   Yes  nc exclusive to Gc CMQ_1340  cript family protein  SC_140  2,471,746  2,472,341   X  - 2,472,335   Yes  nc exclusive to Gs  CMQ_1422  cellulase family protein  SC_140  1,492,574  1,493,183   X  - 1,492,647   Yes  confirmed fixed between Gs and Gc CMQ_1422  cellulase family protein  SC_140  1,492,574  1,493,183   X  - 1,492,889   Yes  confirmed fixed between Gs and Gc     237   Scaffold    Stop codon     RNA evidence a Gene  Gene description    ID  from  to   gain lost position   R N Aseq and/or EST  annotation SNP class  CMQ_1455  isp4 protein   SC_140  3,859,995 3,862,532  X  - 3,860,274   Yes  nc exclusive to Gc from JP  CMQ_1644  4 -coumarate-CoA ligase  SC_140  2,219,024  2,220,788  X  - 2,219,571  Yes  nc exclusive to Gc from JP  CMQ_7268  nb-arc and tpr domain containing protein  SC_173  757,407  760,211  X  - 256,915  Yes  nc fixed between Gs and Gc CMQ_369  major facilitator superfamily transporter   SC_140  632,447  636,531  X  - 633,792  Yes  confirmed fixed between Gs and Gc CMQ_407  cytochrome p450 monooxygenase   SC_140  317,966 319,318  X  - 319,213  Yes  confirmed fixed between Gs and Gc CMQ_547  fungal specific transcription factor  SC_140  4,238,452  4,240,119   X  - 4,239,599   Yes  confirmed exclusive to Gc from JP  CMQ_547  fungal specific transcription factor  SC_140  4,238,452  4,240,119   X  - 4,239,602   Yes  confirmed exclusive to Gc from JP  CMQ_761  periodic tryptophan protein 2  SC_140  3,047,416  3,050,257  - X  3,050,255  Yes  nc fixed between Gs and Gc CMQ_781  acetate kinase  SC_140  44,718  47,502   X  - 47,091   Yes  nc exclusive to Gc CMQ_984  duf1479 domain containing protein   SC_140  565,877 566,990  X  - 566,939  Yes  confirmed fixed between Gs and Gc CMQ_836, CMQ_1107  vacuolar ATPase, hypothet ical protein  SC_140 SC_140  3494217, 3494740  3495330, 3494989   X  - 3,494,963   Yes  confirmed fixed between Gs and Gc CMQ_5397  zinc finger, mynd-type domain containing protein  SC_142  89,680 96,004   X  - 95,970  No nc exclusive to Gc from JP  CMQ_1833  polyketide synthase  SC_180  46,088  51,288  X  - 1,290,261  No nc exclusive to Gs  CMQ_2716  dead deah box DNA helicase  SC_144  360,004  368,123  X  - 360,280  Yes  confirmed fixed between Gs and Gc CMQ_2761  glycosyl transferase  SC_144  271,187 274,967   X  - 271,221  Yes  confirmed fixed between Gs and Gc CMQ_2880  major facilitator superfamily transporter toxin efflux pump  SC_144  763,037 767,468   X  - 763,265  Yes  confirmed fixed between Gs and Gc CMQ_2915  FAD -binding domain containing protein  SC_144  1,301,553 1,302,916  X  - 1,301,799  Yes  confirmed exclusive to Gs  CMQ_3057  metallo-beta-lactamase superfamily protein  SC_144  1,296,201 1,298,402   X  - 1,297,842   Yes  confirmed exclusive to Gs  CMQ_5826  kinase  SC_146  235,947  236,928  X  - 236,799  No nc exclusive to Gc from JP  CMQ_8029  integral membrane protein  SC_150  4,571  5,455   X  - 5,257  Yes  nc exclusive to Gc from JP  CMQ_2296  succinate dehydrogenase cytochrome b560 subunit SC_156  524,590  527,376  X  - 525,351  Yes  nc exclusive to Gs  CMQ_2316  bacilysin biosynthesis oxidoreductase  SC_156  1,202,328 1,203,381  - X  1,202,329  Yes  confirmed fixed between Gs and Gc CMQ_2320  ubiquitin thiolesterase  SC_156  1,373,084  1,376,324   X  - 1,374,505   Yes  confirmed fixed between Gs and Gc CMQ_2390  alpha beta hydrolase fold protein  SC_156  865,934  868,702  X  - 866,838  Yes  confirmed exclusive to Gs  CMQ_8064  major facilitator superfamily transporter   SC_161  426,923  428,356   - X  428,354   Yes  confirmed fixed between Gs and Gc CMQ_8079  hlh transcription factor  SC_1 61 650,035 654,037   X  - 650,646   Yes  confirmed shared CMQ_8216  ankyrin unc44   SC_161  662,004  663,579  - X  663,579  Yes  nc shared CMQ_8227  FAD dependent oxidoreductase superfamily   SC_161  332,380 336,708  X  - 335,539  Yes  confirmed fixed between Gs and Gc CMQ_8251  short chain dehydrogenase reductase  SC_161  157,508 158,403   X  - 158,250  No nc fixed between Gs and Gc CMQ_5958  glycerol-3-phosphate acyltransferase  SC_167  344,180  349,668   X  - 346,655   Yes  confirmed exclusive to Gs  CMQ_6209  kinesin light chain  SC_167  432,239  434,567   - X  434,565   Yes  confirmed exclusive to Gs  CMQ_6468  salicylate hydroxylase  SC_167  212,906 214,272   X  - 213,363  Yes  confirmed exclusive to Gs  CMQ_1913  major facilitator superfamily transporter multidrug resistance  SC_168  293,517 295,417   - X  293,520  Yes  confirmed fixed between Gs and Gc CMQ_1913  major facilitator superfamily transporter multidrug resistance  SC_168  293,517 295,417   X  - 295,077  Yes  confirmed fixed between Gs and Gc CMQ_7076  2og-Fe oxygenase fami ly protein  SC_173  179,330 180,443   X  - 180,436   Yes  confirmed fixed between Gs and Gc CMQ_7358  beta-glucosidase  SC_173  1,697,690 1,706,226  X  - 1,703,615  Yes  confirmed fixed between Gs and Gc CMQ_7517  short chain dehydrogenase reductase  SC_173  1,94 9,855 1,950,845   X  - 1,950,824   No nc fixed between Gs and Gc CMQ_7566  short chain dehydrogenase reductase  SC_173  2,141,595  2,145,557   X  - 2,142,360   Yes  confirmed fixed between Gs and Gc CMQ_7662  d-isomer specific 2-hydroxyacid dehydrogenase  SC_173  144,461  148,338   X  - 145,983   Yes  confirmed fixed between Gs and Gc CMQ_235  ankyrin repeat-containing protein  SC_179  557,417  562,149   X  - 560,393  Yes  confirmed fixed between Gs and Gc CMQ_319  duf221 domain protein  SC_179  940,757  945,050   X  - 943,74 7  Yes  nc shared CMQ_5561  methyltransferase type 11  SC_89  557,677 558,334   X  - 558,206  Yes  confirmed fixed between Gs and Gc CMQ_5634  ribonuclease p complex subunit  SC_89  810,058 817,060  X  - 811,177  Yes  confirmed exclusive to Gs      238   Scaffold    Stop codon     RNA evidence a Gene  Gene description    ID  from to   gain lost position   RN Aseq and/or EST  annotation SNP class  CMQ_5658  ABC transporter  SC_89  370,126  372,064   - X  372,064   Yes  confirmed exclusive to Gs  CMQ_5691  c6 zinc finger domain containing protein  SC_89  200,560  206,396  X  - 201,113  No  nc fixed between Gs and Gc CMQ_4272  aristolochene synthase  SC_97  2,165,605  2,166,601  - X  2,165,607   Yes  confirmed exclusive to Gs  CMQ_4308  tpr domain containing protein  SC_97  2,188,283 2,189,729   X  - 2,188,551   Yes  confirmed fixed between Gs and Gc CMQ_4308  tpr domain containing protein  SC_97  2,188,283 2,189,729   X  - 2,189,422  Y es confirmed fixed between Gs and Gc CMQ_4310  glycoside hydrolase family 3 domain containing protein SC_97  677,205  680,705   X  - 677,855   Yes  confirmed fixed between Gs and Gc CMQ_4346  duf636 domain containing protein  SC_97  2,014,860 2,016,111  X  - 2,015,359   Yes  confirmed exclusive to Gs  CMQ_4381  major facilitator superfamily transporter multidrug resistance  SC_97  1,383,778  1,385,732   X  - 1,384,324  Yes  nc exclusive to Gc from JP  CMQ_4594  kinesin light chain  SC_97  1,054,003  1,057,778   X  - 1,055,6 83  Yes  confirmed fixed between Gs and Gc CMQ_4594  kinesin light chain  SC_97  1,054,003  1,057,778   X  - 1,056,579   Yes  confirmed fixed between Gs and Gc CMQ_4620  cytochrome p450 monooxygenase   SC_97  2,147,761  2,149,239  - X  2,147,764   Yes  confirmed exclusive to Gs  CMQ_4642  cytochrome c oxidase assembly protein  SC_97  522,663  528,069   X  - 527,727   Yes  confirmed fixed between Gs and Gc CMQ_4686  nonribosomal peptide synthetase 11  SC_97  2,154,440  2,154,770   - X  2,154,442   Yes  confirmed exclusive to Gs  CMQ_4704  DNA mismatch repair protein   SC_97  1,765,673  1,767,902   X  - 1,767,188   Yes  confirmed fixed between Gs and Gc CMQ_4750  methyltransferase type 11 domain containing protein SC_97  2,156,931  2,157,678   - X  2,157,677   Yes  confirmed fixed between Gs and Gc CMQ_4813  alcohol dehydrogenase  SC_97  3,984 4,977   X  - 4,371   Yes  confirmed fixed between Gs and Gc CMQ_4844  ABC multidrug transporter mdr1  SC_97  2,158,740  2,162,491  X  - 2,160,614  Yes  confirmed exclusive to Gs  CMQ_4844  ABC multidrug transporter mdr1  SC_97  2,158,740  2,162,491  X  - 2,160,719   Yes  confirmed exclusive to Gs  CMQ_7031  hypothetical protein  SC_105  22,586  22,826  X  - 22,783   Yes  confirmed fixed between Gs and Gc CMQ_6504  hypothetical protein  SC_108  1,425,750  1,427,958   X  - 1,425,851   Yes  confirmed fixed between Gs and Gc CMQ_6899  hypothetical protein  SC_108  310,115  312,256   X  - 311,047   Yes  confirmed exclusive to Gs  CMQ_6899  hypothetical protein  SC_108  310,115  312,256   X  - 311,676   Yes  confirmed exclusive to Gc CMQ_6 945  hypothetical protein  SC_108  1,263,658  1,265,139   X  - 1,263,896  Yes  confirmed exclusive to Gs  CMQ_7010  hypothetical protein  SC_108  647,790  649,796   X  - 649,512   Yes  confirmed exclusive to Gs  CMQ_3377  hypothetical protein  SC_113  2,351,277  2,352 ,385   - X  2,351,280   Yes  confirmed fixed between Gs and Gc CMQ_3672  hypothetical protein  SC_113  2,178,566  2,181,847   X  - 2,179,332   Yes  confirmed exclusive to Gc CMQ_3744  hypothetical protein  SC_113  894,884 896,097   X  - 894,970   Yes  confirmed fixed between Gs and Gc CMQ_4129  hypothetical protein  SC_113  313,835  314,557   X  - 313,942  Yes  confirmed exclusive to Gs  CMQ_4151  hypothetical protein  SC_113  2,455,279  2,456,106   X  - 2,456,102   Yes  confirmed fixed between Gs and Gc CMQ_4166  hypothetical protein  SC_113  2,045,229  2,046,555   X  - 2,045,411   Yes  confirmed fixed between Gs and Gc CMQ_5187  hypothetical protein  SC_132  468,567  468,927   - X  468,570   Yes  confirmed fixed between Gs and Gc CMQ_5224  hypothetical protein  SC_132  313,184 313,927   X   - 313,734   Yes  confirmed fixed between Gs and Gc CMQ_5224  hypothetical protein  SC_132  313,184 313,927   X  - 313,735   Yes  confirmed fixed between Gs and Gc CMQ_5230  hypothetical protein  SC_132  179,393  180,065   X  - 179,732   Yes  confirmed fixed between Gs and Gc CMQ_5252  hypothetical protein  SC_132  427,645  429,724   X  - 429,116  Yes  confirmed fixed between Gs and Gc CMQ_5384  hypothetical protein  SC_132  450,192  451,290   X  - 450,859   Yes  confirmed fixed between Gs and Gc CMQ_1572  hypothetical protein  SC_140  352,991  357,307   X  - 354,072   Yes  confirmed exclusive to Gc from JP  CMQ_1658  hypothetical protein  SC_140  390,422 390,623  - X  390,425   Yes  confirmed fixed between Gs and Gc CMQ_1731  hypothetical protein  SC_140  4,121,328 4,121,775   X  - 4,121,733   Yes  confirmed fixed between Gs and Gc CMQ_711  hypothetical protein  SC_140  4,139,826 4,149,737   X  - 4,149,687   Yes  confirmed fixed between Gs and Gc CMQ_866  hypothetical protein  SC_140  2,426,461 2,427,679   X  - 2,427,191   Yes  confirmed fixed between Gs and Gc CMQ_2694  hypothetical protein  SC_144  2,093,750  2,095,583   X  - 2,094,251   Yes  confirmed exclusive to Gs  CMQ_2694  hypothetical protein  SC_144  2,093,750  2,095,583   X  - 2,095,268   Yes  confirmed exclusive to Gs      239   Scaffold    Stop codon     RNA evidence a Gene  Gene description    ID  from to   gain lost position   RN Aseq and/or EST  annotation SNP class  CMQ_2733  hypothetical protein  SC_144  1,342,629 1,343,439  X  - 1,343,213  Yes  confirmed fixed between Gs and Gc CMQ_2784  hypothetical protein  SC_144  2,286,211 2,290,184  - X  2,286,212  Yes  confirmed fixed between Gs and Gc CMQ_2875  hypothetical protein  SC_144  329,491 330,599  - X  329,494  Yes  confirmed fixed between Gs and Gc CMQ_2883  hypothetical protein  SC_144  2,061,319 2,062,634  X  - 2,061,788  Yes  confirmed fixed between Gs and Gc CMQ_2982  hypothetical protein  SC_144  2,056,629 2,058,582  X  - 2,057,866  Yes  confirmed fixed between Gs and Gc CMQ_2982  hypothetical protein  SC_144  2,056,629 2,058,582  X  - 2,057,867  Yes  confirmed fixed between Gs and Gc CMQ_2982  hypothetical protein  SC_144  2,056,629 2,058,582  X  - 2,058,446  Yes  confirmed fixed between Gs and Gc CM Q_3076  hypothetical protein  SC_144  780,569 781,924  X  - 781,919  Yes  confirmed fixed between Gs and Gc CMQ_3086  hypothetical protein  SC_144  382,354 382,837  - X  382,356  Yes  confirmed fixed between Gs and Gc CMQ_5913  hypothetical protein  SC_146  331,122 332,570  X  - 331,261  Yes  confirmed exclusive to Gc from JP  CMQ_2379  hypothetical protein  SC_156  723,985 724,628  X  - 724,035  Yes  confirmed fixed between Gs and Gc CMQ_2493  hypothetical protein  SC_156  1,033,692 1,034,459  X  - 1,034,016  Yes  confirmed exclusive to Gc from JP  CMQ_4924  hypothetical protein  SC_160  134,153 134,900  X  - 134,885  Yes  confirmed fixed between Gs and Gc CMQ_8178  hypothetical protein  SC_161  453,820 457,067  - X  457,065  Yes  confirmed fixed between Gs and Gc CMQ_8230  hypothetical protein  SC_161  661,156 661,746  X  - 661,650  Yes  confirmed shared CMQ_6397  hypothetical protein  SC_167  1,836,776 1,838,110  X  - 1,838,012  Yes  confirmed fixed between Gs and Gc CMQ_6467  hypothetical protein  SC_167  1,776,729 1,777,100  - X  1,776,732  Yes  confirmed fixed between Gs and Gc CMQ_1922  hypothetical protein  SC_168  446,150 447,728  X  - 446,547  Yes  confirmed exclusive to Gs  CMQ_1945  hypothetical protein  SC_168  51,526 52,066  X  - 51,893  Yes  confirmed fixed between Gs and Gc CMQ_7312  hypothetical protein  SC_173  165,197 166,754  X  - 165,928  Yes  confirmed exclusive to Gc from JP  CMQ_7312  hypothetical protein  SC_173  165,197 166,754  X  - 165,934  Yes  confirmed exclusive to Gc from JP  CMQ_7448  hypothetical protein  SC_1 73 1,451,403 1,452,607  X  - 1,451,846  Yes  confirmed exclusive to Gs  CMQ_7466  hypothetical protein  SC_173  1,851,470 1,853,295  X  - 1,852,771  Yes  confirmed exclusive to Gs  CMQ_7570  hypothetical protein  SC_173  1,691,446 1,694,174  X  - 1,693,931  Ye s confirmed fixed between Gs and Gc CMQ_7578  hypothetical protein  SC_173  1,644,717 1,645,174  X  - 1,645,140  Yes  confirmed fixed between Gs and Gc CMQ_7685  hypothetical protein  SC_173  1,556,619 1,556,994  - X  1,556,992  Yes  confirmed fixed between Gs and Gc CMQ_215  hypothetical protein  SC_179  179,159 180,043  X  - 179,801  Yes  confirmed exclusive to Gc from JP  CMQ_215  hypothetical protein  SC_179  179,159 180,043  X  - 180,025  Yes  confirmed exclusive to Gs  CMQ_1832  hypothetical protein  SC_180  62,260 65,724  - X  65,723  Yes  confirmed exclusive to Gs  CMQ_4677  hypothetical protein   SC_97  1,408,128 1,408,410   - X  1,408,131   Yes  confirmed fixed between Gs and Gc a To avoid stop-codon variants due to erroneous annotation, exon-intron boundaries were check using expressed sequence tag libraries (EST) and transcriptome data from slkw1407 reference strain; nc, indicate genes for which EST and/or RNA -seq data were missing or didnot provide enough coverage to confirm the gene annotations.   The stop-codon variant confirmed by direct sequencing is bolded     240 B. 11  GO functional enrichment analysis of potential pseudogenes in Gs and Gc strains  GO - ID  Term Number of annotated test set Number of annotated Reference set  Test set%  Referece set%  Category a FDR  P- Value #Test  #Ref  GO:0055114 oxidation-reduction process 71 5400 31.0 13.9 P 0.27 1.91E-04 22 750 GO:0016491 oxidoreductase activity 71 5400 31.0 14.4 F 0.27 3.11E-04 22 776 GO:0055085 transmembrane transport 71 5400 16.9 8.2 P 1 0.013374468 12 444 GO:0004553 hydrolase activity, hydrolyzing O-glycosyl compounds 71 5400 7.0 2.0 F 1 0.014654786 5 107 GO:0004497 monooxygenase activity 71 5400 7.0 2.0 F 1 0.015181917 5 108 GO:0000272 polysaccharide catabolic process 71 5400 4.2 0.7 P 1 0.017840778 3 40 GO:0016798 hydrolase activity, acting on glycosyl bonds 71 5400 7.0 2.1 F 1 0.019863043 5 116 GO:0042537 benzene-containing compound metabolic process 71 5400 8.5 0.3 P 1 0.024626681 2 17 GO:0000166 nucleotide binding 71 5400 18.3 20.3 F 1 0.041854134 21 1096 a F, Molecular Functions; P, Biological Process             241 B.12 Map of western North America representing the mountain pine beetle (MPB) distribution.   In the green region the MPB occurs with its sister species, the jeffrey pine beetle. Numbers indicate the fungal collection sites (Table 4.1). Colored dots represent the locations and host tree species from which the fungi were isolated.        242 B.13 Large-scale synteny between Grosmannia genomes  The MUMmer dotplots indicate matching sequences and show co-directional regions of synteny in red. The genome of the reference strain with a total length of 29.8 Mb is represented on the X-axis and those from other strains including Gc from P. ponderosa (a) and P. jeffreyi (b), and Gs from northern BC (c) and California (d) are shown on the Y-axis. The values above the graph show the percentage of the Gc and Gs assembled contig length that was syntenic with the reference genome (See appendix B.2).       243 B.14  G enetic distance between the twelve Grosmannia genomes.  The multidimensional scaling-plot represents the genetic distance between genomes calculated from 103,430 single nucleotide variants.             244  B.15  Haplotype networks and genealogies of nine gene regions sequenced in 28 Grosmannia strains.   The genealogy of each of the nine-gene regions was best described by a single, unrooted, most-parsimonious tree, except for the positioning of Gc ? 10 in the CMQ861 ?ABC.G transporter tree, which shows paraphyly between Gc strains from P. ponderosa and those from P. jeffreyi. Re lationships among the clades were topologically identical in the network method as well as the tree-based methods using maximum parsimony (MP) and Bayesian analyses. Here, the parsimony networks are displayed with the MP bootstrap support and Bayesian posterior probability given for the main branches. The purple circles represent the Gs haplotypes and the green represents the Gc haplotypes. A connecting line between haplotypes represents one mutation and small black dots represent missing (inferred) haplotypes. Blue-filled circles are haplotypes found only in Gc strains from P. ponderosa, and the gray-filled circles are haplotypes in Gs strains from the California localized populations.            245         246  B.16 Distributions of polymorphism-to-divergence ratios (NI) and the rate of protein-coding evolution (dN/dS) across 3,476 Gs-Gc orthologous gene models.   a) From the distribution of the neutrality index (NI), the 1,215 protein - coding genes with ? log10 NI >0 show evidence of adaptive evolution, but the majority of t he genes appear to be under weak purifying selection (? log10 NI<0 ). b) The distribution of the estimated average rate of protein- coding evolution in pairwise Gs ? Gc comparison s show that the majority of genes have dN/dS !1, indicating that most of these genes are either under purifying selection or positive selection at specific sites. For 228 protein - coding genes, dN/dS >1 suggests adaptive evolution or at least relaxed selecti on constraints.    

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:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0165679/manifest

Comment

Related Items