Bacterial Endosymbionts of Endophytic Fungi: Diversity, Phylogenetic Structure, and Biotic Interactions

Home , Ampelomyces quisqualis, Aphanoascus, Aphanoascus fulvescens, Ascobolus carbonarius, Barssia, Bimuria

Item Type text; Electronic Dissertation

Authors Hoffman, Michele Therese

Publisher The University of Arizona.

Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.

Download date 23/09/2021 12:40:10

Link to Item http://hdl.handle.net/10150/196079

BACTERIAL ENDOSYMBIONTS OF ENDOPHYTIC FUNGI: DIVERSITY, PHYLOGENETIC STRUCTURE, AND BIOTIC INTERACTIONS

Michele Therese Hoffman ______

A Dissertation Submitted to the Faculty of the

DEPARTMENT OF PLANT SCIENCES

In Partial Fulfillment of the Requirements For the Degree of

DOCTOR OF PHILOSOPHY WITH A MAJOR IN PLANT PATHOLOGY

In the Graduate College

THE UNIVERSITY OF ARIZONA

2010

THE UNIVERSITY OF ARIZONA GRADUATE COLLEGE

As members of the Dissertation Committee, we certify that we have read the dissertation prepared by Michele T. Hoffman entitled "Bacterial endosymbionts of endophytic fungi: diversity, phylogenetic structure, and biotic interactions.” and recommend that it be accepted as fulfilling the dissertation requirement for the

Degree of Doctor of Philosophy.

A. Elizabeth Arnold ______Date: 4/21/10

Judith L. Bronstein ______Date: 4/21/10

Marc J. Orbach ______Date: 4/21/10

Final approval and acceptance of this dissertation is contingent upon the candidate’s submission of the final copies of the dissertation to the Graduate College.

I hereby certify that I have read this dissertation prepared under my direction and recommend that it be accepted as fulfilling the dissertation requirement.

A. Elizabeth Arnold ______Date: 4/21/10 Dissertation Director 3

STATEMENT BY AUTHOR

This dissertation has been submitted in partial fulfillment of requirements for an advanced degree at the University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library.

Brief quotations from this dissertation are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his or her judgment the proposed use of the material is in the interests of scholarship. In all other instances, however, permission must be obtained from the author.

SIGNED: Michele T. Hoffman

ACKNOWLEDGMENTS

This research would not have been possible without the assistance of many wonderful individuals. I wish to first thank my advisor A. Elizabeth (Betsy) Arnold for her support and guidance these past years. I am forever grateful to have had this opportunity to participate in this outstanding research program. I appreciated that I was given the freedom to explore my own research questions while being rigorously challenged as a scientist. I am also most grateful to the members of my committee, Judie Bronstein for your thoughtful questions and feedback, Marc Orbach for helping me troubleshoot research problems and Sandy Pierson, for all things prokaryotic. I have appreciated all your time, guidance and patience. I want to thank the School of Plant Sciences and the IGERT program for funding and the staff members in Plant Pathology and Microbiology for your kindness and good will. Numerous friends deserve my thanks and gratitude for their support, cheerful spirit, and camaraderie. I wish to especially thank Cara Gibson and members of her phylogenetics lunch club, Jeff Oliver and Chris Schmidt, along with fellow graduate students and faculty: Fabiola Santos Rodriguez, Anne Estes, David Maddison, Randy Ryan, David Bentley, Periasamy (Ravi) Chitrampalam, Scott Kroken, Mary Jane Epps, Jana U’ren, Rhodesia Celoy, Mariana Del Olmo Ruiz, David Jarvis, Gerard White, Baomin Wang, Mary Shimabukuro, Dylan Grippi, Andrea Woodard, Jacob Russell, Eric McDowell, and Bridget Barker. I could not have accomplished this work without the help and guidance of Mali Gunatilaka. You are a special soul Mali and I owe you one million thanks! Finally, I wish to thank my incredible friends and family for their love and support. 5

DEDICATION

This journey represents the culmination of many years of good intentions, perseverance, and the unwavering belief that dreams can and do eventually come true. I dedicate this work to Edie, the person who encouraged me to dream.

TABLE OF CONTENTS

ABSTRACT...... 7

INTRODUCTION...... 9

1.1 Literature review.……………………………….…...... 9

1.2 Explanation of dissertation format………………………………………………...15

PRESENT STUDY...... 17

2.1 Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees……………………………………………………………………....17

2.2 Diverse bacterial endosymbionts inhabit living hyphae of phylogenetically diverse fungal endophytes …………………………………………………………...….18

2.3 IAA production by endophytic Pestalotiopsis neglecta is enhanced by an endohyphal bacterium .…………………………………..…………...……………....…18

REFERENCES…………………………………………………………………………..19

APPENDIX A: GEOGRAPHIC LOCALITY AND HOST INDENTITY SHAPE

FUNGAL ENDOPHYTE COMMUNITIES IN CUPRESSACEOUS TREES...... 24

APPENDIX B: DIVERSE BACTERIAL ENDOSYMBIONTS INHABIT LIVING

HYPHAE OF PHYLOGENETICALLY DIVERSE FUNGAL ENDOPHYTES ….….49

APPENDIX C: IAA PRODUCTION BY ENDOPHYTIC PESTALOTIOPSIS

NEGLECTA IS ENHANCED BY AN ENDOHYPHAL BACTERIUM..…………….99 7

ABSTRACT

This dissertation comprises a series of studies designed to explore the associations between plants and the endophytic fungi they harbor in their above-ground tissues. By viewing endophyte diversity in ecologically and economically important hosts through the lenses of phylogenetic biology, microbiology, and biotechnology, this body of work links plant ecology with newly discovered symbiotic units comprised of endophytic fungi and the bacteria that inhabit them.

This work begins with a large-scale survey of endophytic fungi from native and non-native Cupressaceae in Arizona and North Carolina. After isolating over 400 strains of endophytes, I inferred the evolutionary relationships among these fungi using both

Bayesian and parsimony analyses. In addition to showing that native and introduced plants contained different endophytes, I found that the endophytes themselves harbor additional microbial symbionts, recovering members of the beta- and gamma- proteobacterial orders Burkholderiales, Xanthomonadales, and Enterobacteriales and numerous novel, previously uncultured bacteria. This work finds that phylogenetically diverse bacterial endosymbionts occur within living hyphae of multiple major lineages of ascomycetous endophytes.

A focus on 29 fungal/bacterial associations revealed that bacterial and fungal phylogenies are incongruent with each other and did not reflect the phylogenetic relationships of host plants. Instead, both endophyte and bacterial assemblages were strongly structured by geography, consistent with local horizontal transmission. 8

Endophytes could be cured of their bacterial endosymbionts using antibiotics, providing a tractable experimental system for comparisons of growth and metabolite production under varying conditions. Studies of seven focal fungal/bacterial pairs showed that bacteria could significantly alter growth of fungi at different nutrient and temperature levels in vitro, and that different members of the same bacterial lineages interact with different fungi in different ways.

Focusing on one isolate, I then describe for the first time the production of indole-

3-acetic acid (IAA) by a non-pathogenic, foliar endophytic fungus (Pestalotiopsis neglecta), suggesting a potential benefit to the host plant harboring this fungus. I show that this fungus is inhabited by an endohyphal bacterium (Luteibacter sp.) and demonstrate that mycelium containing this bacterium produces significantly more IAA in vitro than the fungus alone. I predict that the general biochemical pathway used by the fungal-endohyphal complex is L-tryptophan-dependent and measure effects of IAA production in vivo, focusing on root and shoot growth in tomato seedlings.

INTRODUCTION

1.1 Literature review

1.1.1 Fungal endophytes

All lineages of land plants surveyed to date harbor living fungi within their healthy, aboveground tissues (e.g., Carroll, 1988; Petrini, 1996; Davis et al., 2003; Cohen, 2004;

Arnold & Lutzoni, 2007; Kauserud et al., 2008; Arnold et al., 2009). These endophytic fungi colonize and persist within living tissues such as leaves and stems without causing overt symptoms of disease (Petrini, 1991; Stone et al., 2000). Fungal endophytes are highly diverse at the species level (Carroll, 1995; Fröhlich & Hyde, 1999; Lu et al., 2004;

Rodrigues et al., 2004; Arnold, 2007), represent every major lineage of non-lichenized

Pezizomycotina (Higgins et al., 2007; Arnold, 2008), and have arisen multiple times in the evolution of Fungi (Spatafora et al., 2006; Arnold et al., 2009). In life history they range from the systemic, vertically transmitted clavicipitaceous endophytes that have co- evolved with their cool-season grass hosts (Class 1 endophytes, Rodriguez et al., 2009) to highly localized, horizontally transmitted endophytes that associate at various levels of specificity with all major lineages of plants (Class 3 endophytes, Rodriguez et al., 2009).

Among these, the horizontally transmitted, phylogenetically diverse Class 3 endophytes associated with foliage of long-lived perennial trees are of special interest: individual trees concurrently harbor dozens to hundreds of endophyte species within their asymptomatic photosynthetic tissues (Lodge et al., 1996; Müller & Hallaksela, 1998;

Arnold et al., 2003; Göre & Bucak, 2007), raising questions regarding the nature of their 10

interactions with hosts. These interactions appear to range from latent pathogenicity to latent saprotrophy, and encompass a variety of defensive mutualisms and physiological costs and benefits (Arnold, 2007). Such interactions change frequently over evolutionary time and may be variable and unpredictable in ecological time frames as well (Arnold,

2008; Arnold et al., 2009).

The mechanisms driving variation in such interactions provide the context for my dissertation research. Whether variable outcomes reflect flexible genomic architecture in endophytes or their plant hosts, sensitivity of the interaction or either partner to biotic or abiotic stress, or additional microbial participants in species interactions is not yet clear.

In this dissertation, I have focused on the third area, with special attention to the previously unexplored roles of endohyphal bacteria in shaping plant-endophyte interactions.

1.1.2 Endohyphal bacterial symbionts

In addition to interacting with environmental and ectosymbiotic bacteria, some plant- associated fungi harbor bacteria within their hyphae (first noted as ‘bacteria-like organisms’ of unknown function by MacDonald and Chandler, 1981). These bacteria, best known from living hyphae of several species of Glomeromycota and

Mucoromycotina, can alter the fungal phenotype (see Bianciotto et al., 1996; Bertaux et al., 2003; Bidartondo et al., 2004; de Boer et al., 2005; Waller et al., 2005; Shefferson et al., 2005). For example, Candidatus Glomeribacter gigasporarum colonizes spores and hyphae of the arbuscular mycorrhizal (AM) fungus Gigaspora gigasporarum (Bianciotto 11

et al., 2000; Levy et al., 2003; Lumini et al., 2007). When cured of the bacterium, fungal growth and development are suppressed, and the morphology of the cell wall, vacuoles, and lipid bodies changes noticeably. Recovery of a P-transporter operon in Burkholderia sp. from Gigaspora margarita (Ruiz-Lozano & Bonfante, 1999), and the discovery of phosphate-solubilizing bacteria within Glomus mossae spores (Mirabal-Alonso & Ortega-

Delgado, 2007), suggest a competitive role in phosphate acquisition and transport by these bacteria within the AM symbiosis. Recently, Partida-Martinez and Hertweck (2005) reported that a plant pathogen, Rhizopus microsporus (Mucoromycotina), harbors an endosymbiotic Burkholderia sp. that produces a phytotoxin (rhizoxin) responsible for the pathogenicity of the fungus. Each of these suggests a pervasive role of these endohyphal bacteria not only on their fungal hosts, but on plant-fungus interactions as well. To date, however, endohyphal bacteria have been studied only in rhizosphere-associated or plant- pathogenic fungi. These studies have not encompassed some of the most species-rich fungal lineages, nor the ubiquitous, avirulent endophytic fungi found in above-ground plant tissues.

1.1.3 Endohyphal bacteria: hidden drivers of plant-endophyte interactions?

In this dissertation, I demonstrate that foliar endophytes harbor diverse, apparently horizontally acquired, and facultative bacterial endosymbionts, and provide the first evidence for their effects on foliar endophytes and plant-endophyte interactions. I have focused on host plants representing the Cupressaceae (cedars and cypresses), an economically and ecologically important family of conifers. The family comprises 130- 12

140 species of trees and shrubs in approximately 27-30 genera and encompasses the widest geographic range of any conifer family, with species ranging from 70°N to 55°S.

Well-known species include coastal redwood (Sequoia sempervirens), giant sequoia

(Sequoiadendron giganteum), and bald cypress (Taxodium spp.). In the mid-elevation woodlands and forests of Arizona and much of the intermountain west, Cupressaceae form the dominant component of tree communities. Species of Juniperus and Cupressus are especially common both naturally and as ornamentals, with species such as

Platycladus orientalis (Oriental arbor-vitae) commonly planted as well. Endophytic fungi have been examined in a small number of cupressaceous hosts previously (Carroll 1978,

1995; Camacho et al., 1997; Deckert, 2000; Hata, 1996; Arnold et al., 2007), but never in an explicit geographic context, nor with the geographic breadth and depth of the present work.

I first address the degree to which endophytic fungal associates of trees in the

Cupressaceae differ across the geographic ranges of hosts. In this work I examine the role of native vs. non-native status in shaping endophyte communities as well. My goal is to understand the intrinsic variation in fungal communities encountered and acquired by plants in different geographic areas. Using both genotype-level and phylogenetic analyses, I show that (1) conspecific trees in widely separated geographic areas

(southwestern vs. southeastern US) harbor different endophyte communities, and (2) that native and introduced trees in the same geographic areas differ in their endophyte assemblages. That study yielded a tremendous diversity of previously unknown endophytic fungi and a new understanding of their phylogenetic relationships. 13

Next, I examined the potential for such endophytic fungi to harbor endohyphal bacteria. For this work, I concentrated on fungi associated with healthy foliage of

Cupressaceae at a smaller geographic scale, with a particular focus on exploring fungal communities associated with plants on multiple mountain ranges within Arizona. After extracting total genomic DNA from several axenic endophyte cultures, I discovered the presence of measurable bacterial DNA within those DNA samples. After ruling out contamination (Hoffman and Arnold, 2010), I sequenced the 16S ribosomal DNA region of representative strains and recovered members of the beta- and gamma-proteobacterial orders Burkholderiales, Xanthomonadales, and Enterobacteriales, including putative

Ralstonia pickettii, Variovorax paradoxus, Luteibacter rhizovicina, Pantoea spp., and numerous novel, previously uncultured bacteria. I localized these bacteria within hyphae and confirmed the viability of both the bacteria and the hyphae harboring them using

Molecular Probes Live-Dead stain. Phylogenetic analyses of fungi and their associated bacteria then provided strong evidence that bacteria can occur within endophytic members of four classes of Ascomycota. My work suggests that these are facultative associates of their fungal hosts, and highlights the lack of congruence among bacterial, fungal, and host plant phylogenies.

After developing methods to cure fungi of their endohyphal bacteria, I examined the effects of bacterial endosymbionts on endophytic growth rates in vitro. Specifically, by cultivating fungi on growth media containing the antibiotic ciprofloxacin, I produced fungal cultivars that were confirmed by microscopy and PCR to lack endohyphal bacteria

(Hoffman and Arnold, 2010). By comparing the growth of replicate clones with and 14

without bacteria on various media, at different temperatures and at different pH levels, I found that these bacteria had variable and sometimes unpredictable effects. One strain,

9143, proved especially interesting because (1) I could identify the both the fungus and bacterium with reasonable certainty using morphological and phylogenetic methods, (2) I could separate the fungal and bacterial partners in vitro, and (3) this particular genus

(Pestalotiopsis) is well known for its production of numerous medicinally important secondary metabolites.

The ability to synthesize phytohormones, including indole-3-acetic acid (IAA), gibberellins, ethylene, cytokinins, jasmonic acid and abscisic acid, have been documented in numerous bacteria and fungi (Costacurta and Vanderleyden, 1995; Glick, 1995;

Lindow et al., 1998; Robinson et al., 1998; Koga, 1995; Maor et al., 2004; Sirrenberg et al., 2007). Some of these microorganisms utilize a L-tryptophan-dependent pathway for

IAA production, whereas others utilize a L-tryptophan independent pathway (Tudzynski and Sharon, 2002; Buchanan et al., 2005). My investigation of IAA production in infected endophytes found that some isolates were capable of IAA production when supplied with L-tryptophan and that the infected endophytes produced significantly higher levels of IAA than the cured counterpart, under identical growth conditions.

Culture filtrate from the endophyte-bacterium complex significantly increased root growth in vitro in seedlings of tomato relative both to controls and to filtrate from the endophyte alone. These data suggest that this newly described microbial partnership can manipulate plant biochemistry with a positive effect.

1.2 Explanation of dissertation format

The goals of this dissertation are to explore the interactions of endophytic fungi with other microorganisms and their plant hosts, and to expand our knowledge of endophyte diversity, community structure, and ecological roles in nature. I introduce this dissertation research, methods and results as three appendices.

In Appendix A, I examine the abundance, diversity, and species composition of endophytic fungi in pairs of related species of Cupressaceae from a mesic forest (central

North Carolina, USA) and from a xeric desert environment (southern Arizona, USA),

Utilizing a culture-based approach, I examined the fungal communities of two native species (Cupressus arizonica, AZ and Juniperus virginiana, NC) and a co-occurring, non-native species (Platycladus orientalis in both AZ and NC). My results lay the groundwork for addressing relative importance of locality, host species, and native/non- native status in endophyte community structure.

In Appendix B, I examine evidence of endohyphal bacteria associated with these fungal endophytes and began to elucidate the infection frequency, taxonomy, and phylogenetic relationships of these cryptic microorganisms. This work reports, for the first time, the occurrence of endohyphal bacteria in foliar endophytes and highlights their occurrence in four classes of Pezizomycotina. I recover novel lineages of endohyphal bacteria relative to previous studies and suggest that the association of these bacteria and their endophytic hosts is facultative, reflecting (1) loss of bacterial endosymbionts in culture over time, and (2) incongruent fungal- and bacterial phylogenies. These analyses 16

set the stage for experimental work addressing the ecological importance of these previously unknown symbionts.

The data presented in Appendix C are the results of in vitro and in planta experiments, conducted in collaboration with Malkanthi K. Gunatilaka, E. M. Kithsiri

Wijeratne, and A. A. Leslie Gunatilaka, measuring phytohormone production in a focal endophytic fungus. This work demonstrates, for the first time, the ability of a foliar endophyte to produce indole-3-acetic-acid (IAA), a phytohormone with strong effects on plant cell growth and elongation. Moreover, we provide the first documentation of a role of an endohyphal bacterium in enhancing IAA production, and demonstrate a strong effect on seedling growth. In general, this work begins to answer the question: how do bacterial endosymbionts influence the outcome of plant-fungal symbioses?

PRESENT STUDY

The methods, results and conclusions from this research are presented in the appended manuscripts. The following is a summary of the most important findings from each.

2.1 Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees

Fungal endophytes, found in every major lineage of the most species-rich group of fungi

(Ascomycota), are believed to represent a significant proportion of Earth’s fungal diversity. Using 95% sequence similarity as a proxy for species recognition, I report a high level of species richness in three species of Cupressaceae, an economically and ecologically important family of conifers. In addition to isolating over 400 strains of endophytes, I inferred the evolutionary relationships among these fungi for the first time using both Bayesian and parsimony analyses. Analyses of the nuclear ribosomal internal transcribed spacer regions (ITS rDNA) for 109 representative isolates showed that diversity was two times greater in plants from North Carolina vs. Arizona, and that endophytes recovered in Arizona were more likely to be host-generalists relative to those in North Carolina. Overall abundance, diversity, and taxonomic composition of these endophyte communities differed as a function of host identity, native vs. non-native status, and locality. 18

2.2 Diverse bacterial endosymbionts inhabit living hyphae of phylogenetically diverse fungal endophytes

Here I report that phylogenetically diverse bacteria occur within living hyphae of endophytic fungi, including members of multiple major lineages of Ascomycota.

Drawing from surveys of endophytes in healthy photosynthetic tissues of ecologically and economically important trees in five biogeographic provinces, I show that bacterial and fungal phylogenies are not congruent and do not reflect the evolutionary relationships of the host plants they both inhabit. Instead, both endophyte and bacterial endosymbiont species appear to be horizontally transmitted and are structured more by geography than by host identity.

2.3 IAA production by endophytic Pestalotiopsis neglecta is enhanced by an endohyphal bacterium

In this section, I demonstrate the effects of a representative endohyphal bacterium, identified using phylogenetic analyses as Luteibacter sp., on growth rates of endophytic

Pestalotiopsis neglecta under various nutrient, temperature and pH conditions. I show that this fungal endophyte is capable of producing the phytohormone indole-3-acetic acid

(IAA) in vitro. I report a significant difference in production of IAA for the infected vs. cured and that this same compound could positively influence the growth rate of tomato seedlings. 19

REFERENCES

Arnold, A. E., Miadlikowska, J., Higgins, K. L., Sarvate, S. D., Gugger, P., Way, A., Hofstetter, V., Kauff, F. & Lutzoni, F. (2009). Hyperdiverse fungal endophytes and endolichenic fungi elucidate the evolution of major ecological modes in the Ascomycota. Systematic Biology 58: 283-297.

Arnold, A.E. (2008). Endophytic fungi: Hidden components of tropical community ecology. Tropical Forest Community Ecology eds. Walter P. Carson & Stefan A. Schnitzer, Wiley-Blackwell West Sussex, UK.

Arnold, A. E. (2007). Understanding the diversity of foliar endophytic fungi: Progress, challenges, and frontiers. Fungal Biology Reviews 21: 51-66.

Arnold, A. E. & Lutzoni, F. (2007). Diversity and host range of foliar fungal endophytes: Are tropical leaves biodiversity hotspots? Ecology 88: 541-549.

Arnold, A. E., Henk, D. A., Eells, R., Lutzoni, F. & Vilgalys, R. (2007). Diversity and phylogenic affinities of foliar fungal endophytes in loblolly pine inferred by culturing and environmental PCR. Mycologia 99:185-206.

Arnold, A. E., Mejía, L., Kyllo, D., Rojas, E., Maynard, Z., Robbins, N. A. & Herre, E. A. (2003). Fungal endophytes limit pathogen damage in a tropical tree. Proceedings of the National Academy of Sciences USA 100: 15649–15654.

Bertaux, J. et al. (2003). In situ identification of intracellular bacteria related to Paenibacillus spp. in the mycelium of the ectomycorrhizal fungus Laccarria bicolor S238N. Applied and Environmental Microbiology 69 (7): 4243-4248.

Bianciotto, V., Bandi, C., Minerdi, D., Sironi, M., Tichy, H.V., and Bonfante, P. (1996). An obligate endosymbiont mycorrhizal fungus itself harbors oligately intracellular bacteria. Applied and Environmental Microbiology 62(8): 3005-3010.

Bianciotto, V. at al. (2000). Detection and identification of bacterial endosymbionts in arbuscular mycorrhizal fungi belonging to the family Gigasporaceae. Applied and Environmental Microbiology 66 (10): 4503-4509.

Bianciotto, V., Lumini, E., Bonfante, P. and Vandamme, P. (2003). ‘Candidatus Glomeribacter gigasporarum’ gen. nov., sp. nov., an endosymbiont of arbuscular mycorrhizal fungi. Int J Syst Evol Microbiol 53:121-124.

Bidartondo, M. I. et al. (2004). Changing partners in the dark: isotopic and molecular evidence of ectomycorrhizal liaisons between forest orchids and trees. Proc. R. Soc. London B 271: 1799-1806.

Buchanan, B. B., Gruissem, W., and Jones, R. L. (2005). Biochemistry and Molecular Biology of Plants. American Society of Plant Physiologists, Rockville, MD.

Camacho, F. J., Gernandt, D. S., Liston, A., Stone, J.K., and Klein, A. S. (1997). Endophytic fungal DNA, the source of contamination in spruce needle DNA. Molecular Ecology 6:983-987.

Carroll, G. (1988). Fungal endophytes in stem and leaves: Latent pathogens to mutualistic symbionts. Ecology 69: 2-9.

Carroll, G. (1995). Forest endophytes: Pattern and process. Canadian Journal of Botany 73: S1316–S1324.

Carroll, G.C. (1978). Studies on the incidence of coniferous needle endophytes in the pacific northwest. Canadian Journal of Botany 56: 3034.

Cohen, S. D. (2004). Endophytic-host selectivity of Discula umbrinella on Quercus alba and Quercus rubra characterized by infection, pathogenicity and mycelial compatibility. European Journal of Plant Pathology 110(7): 713-721.

Costacurta, A. and Vanderleyden, J. (1995). Synthesis of phytohormones by plant- associated bacteria. Critical Rev. Microbiol. 21:1-18.

Davis, E.C. et al. (2003). Endophytic Xylaria (Xylariaceae) among liverworts and angiosperms: Phylogenetics, distribution, and symbiosis. American Journal of Botany 90(11): 1661-1667.

Deckert, R. J. (2000). Distribution of foliar fungal endophytes of Pinus strobus between and within host trees. Canadian Journal of Forestry 30:1436.

De Boer, W. et al. (2005). Living in a fungal world: impact of fungi on soil bacterial niche development. FEMS Microbiology Reviews 29: 795.

Fisher, P.J. and Petrini, O. (1992). Fungal saprobes and pathogens as endophytes of rice (Oryza-sativa L.) New Phytologist 120: 137-143.

Fralish, J. S. & Franklin, S. B. (2002). Taxonomy and Ecology of Woody Plants in North American Forests. New York: Wiley & Sons.

Freeman, S. and Rodriguez, R.J. (1993). Genetic conversion of a fungal plant pathogen to a nonpathogenic, endophytic mutualist. Science 260(5104): 75-78.

Fröhlich, J. & Hyde, K. D. (1999). Biodiversity of palm fungi in the tropics: Are global fungal diversity estimates realistic? Biodiversity and Conservation 8: 977–1004.

Gadek, P.A., et al (2000). Relationships within Cupressaceae sensu lato: A combined morphological and molecular approach. American Journal of Botany. 87(7): 1044-1057 21

Garbaye, J. (1994). Helper bacteria - a new dimension to the mycorrhizal symbiosis. New Phytologist 128(2): 197-210.

Gardes, M. & Bruns, T. D. (1993). ITS primers with enhanced specificity of basidiomycetes: application to the identification of mycorrhizae and rusts. Molecular Ecology 2: 113–118.

Glick, B. R. (1995). The enhancement of plant growth by freeliving bacteria. Can J Microbiol 41:109–117.

Gore, M.E. and Bucak, C. (2007). Geographical and seasonal influences on the distribution of fungal endophytes in Laurus nobilis. Forest Pathology 37: 281-288.

Hata, K. (1996). Variation in fungal endophyte populations in needles of the genus Pinus Canadian Journal of Botany 74: 103.

Higgins, K. L., Arnold, A. E., Miadlikowska, J., Sarvate, S. D. & Lutzoni, F. (2007). Host affinity and geographic structure among boreal and arctic endophytes from three major plant lineages. Molecular Phylogenetics and Evolution 42: 543–555.

Hoffman, M.T. & Arnold, A.E. (2008). Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees. Mycological Research 112: 331- 344.

Hoffman, M. T. and A. E. Arnold. (2010). Diverse bacteria inhabit living hyphae of phylogenetically diverse fungal endophytes. Applied and Environmental Microbiology in press.

Kauserud, H., Mathiesen, C. and Ohlson, M. (2008). High diversity of fungi associated with living parts of boreal forest bryophytes. NRC Research Press Botany 86: 1326- 1333.

Koga, J. (1995). Structure and function of indolepyruvate decarboxylase, a key enzyme in indole-3-acetic acid biosynthesis. Biochim Biophys Acta 1249:1–13

Levy, A., Chang B.J., Abbott, L.K., Kuo, J.,Harnett, G., & Inglis, T.J.J. (2003). Invasion of spores of the arbuscular mycorrhizal fungus Gigaspora decipiens by Burkholderia spp. Applied and Environmental Microbiology 69(10): 6250-6256.

Lindow, S.E., Desurmont, C., Elkins, R., McGourty, G., Clark, E. and Brandl, M.T. (1998). Occurrence of indole-3-acetic acid-producing bacteria on pear trees and their association with fruit russet. Phytopathology 88:1149-1157.

Little, D. P. (2006). Evolution and circumscription of the true cypresses (Cupressaceae: Cupressus). Systematic Botany 31(3): 461-480. 22

Lodge D.J., Fisher, P.J. and Sutton, B.C. (1996). Endophytic fungi of Manikara bidentata leaves in Puerto Rico. Mycologia 88(5): 733-738. Lu, G.Z, Cannon, P.F., Reid A., and Simmons, C.M. (2004). Diversity and molecular relationships of endophytic Colletotrichum isolates from the Iwokrama Forest Reserve, Guyana. Mycological Research 108: 53-63. Lumini, E. et al. (2007). Presymbiotic growth and sporal morphology are affected in the arbuscular mycorrhizal fungus Gigaspora margarita cured of its endobacteria. Cellular Microbiology 9(7): 1716-1729. MacDonald, R.M. and Chandler, M.R. (1981). Bacterium-like organelles in the vesicular- arbuscular mycorrhizal fungus Glomus caledonius. New Phytologist 89: 241-246.

Maor, R., Haskin, S., Levi-Kedmi, H., and Sharon, A. (2004). In planta production of indole-3-acetic acid by Colletotrichum gloeosporioides f. sp. aeschynomene. Applied and Environmental Microbiology 70(3):1852-1854.

Mayo, K., Davis, R.E., and Motta, J. (1986). Stimulation of germination of spores of Glomus versiforme by spore-associated bacteria. Mycologia 78(3): 426-431. Minerdi, D. et al., (2008). Bacterial ectosymbionts and virulence silencing in a Fusarium oxysporum strain. Environmental Microbiology 10(7):1725-1741.

Mirabal-Alonso, L. & Ortega-Delgado, E. (2007). Phosphate solubilizing bacteria isolated from the inside of Glomus mosseae spores from Cuba. First International Meeting on Microbial Phosphate Solubilization. Series: Developments in Plant and Soil Sciences , Vol. 102 Velazquez, E.; Rodriguez-Barrueco, C. (Eds.) Reprinted from Plant and Soil, Vol. 287, 1-84

Müller, M.M. and Hallaksela, A-M. (1998). Diversity of Norway spruce needle endophytes in various mixed and pure Norway spruce stands. Mycological Research 102(10): 1183-1189.

Partida-Martinez, L.P. & Hertweck, C. (2005). Pathogenic fungus harbours endosymbiotic bacteria for toxin production. Nature 437:884-888.

Petrini, O. (1996). Ecological and physiological aspects of host-specificity in endophytic fungi. In: Endophytic Fungi in Grasses and Woody Plants (S. C. Redlin & L. M. Carris, eds): 87–100. APS Press, St. Paul, Mn.

Petrini, O. (1991). In: Microbial Ecology of Leaves, eds. Andrews, J. H. & Hirano, S. S. (Springer, New York), pp. 179–197.

Robinson, M., Riov, J., and Sharon, A. (1998). Indole-3-acetic acid biosynthesis in Colletotrichum gloeosporioides f. sp. aeschynomene. Applied and Environmental Microbiology 64(12):5030-5032.

Rodrigues, K. F., Sieber, T. N., Gruenig, C. R. & Holdenrieder, O. (2004). Characterization of Guignardia mangiferae isolated from tropical plants based on morphology, ISSR-PCR amplifications and ITS1-5.8S-ITS2 sequences. Mycological Research 108: 45–52.

Rodriguez, R.J., White, J.F., Arnold, A.E., and Redman, R.S. (2009). Fungal endophytes: Diversity and functional roles. New Phytologist 182:314-330.

Ruiz-Lozano, J.M. and Bonfante, P. (1999). Identification of a putative P-transporter operon in the genome of a Burkholderia strain living inside the arbuscular mycorrhizal fungus Gigaspora margarita. Journal of Bacteriology 181:4106-4109.

Shefferson, R. P., et al., (2005). High Specificity generally characterizes mycorrhizal associations in rare lady’s slipper orchids, genus Cypripedium. Molecular Ecology 14: 613-626.

Sirrenberg, A. Göbel, C., Grond, S., Czempinski, N., Ratzinger, A., Karlovsky, P., Santos, P., Feussner, I. and Pawlowski, K. (2007). Piriformospora indica affects plant growth by auxin. Physiologia Plantarum 131:581-589.

Spatafora, J. W., Sung, G. H., Johnson, D., et al. (2007). A five-gene phylogeny of Pezizomycotina. Mycologia 98(6): 1018-1028.

Stanosz, G.R., Blodgett, J.T., Smith, D.R., and Kruger, E.L. (2001). Water stress and Sphaeropsis sapinea as a latent pathogen of red pine seedlings. New Phytologist 149: 531-538.

Stone, J. K., Bacon, C. W. & White, J. F., Jr. (2000). An overview of endophytic microbes: Endophytism defined. In: Bacon, C. W., White, J. F., Jr. (Eds.) Microbial Endophytes. Marcel Dekker, New York. 3–29.

Tudzynski, B. and Sharon, A. (2002). Biosynthesis, biological role and application of fungal phytohormones. The Mycota X. Industrial Applications. Springer-Verlag Berlin Heidelberg.

Waller, F., B. Achatz, H. Baltruschat, J. Fodor, K. Becker, M. Fischer, T. Heier, R. Hückelhoven, C. Neumann, D. von Wettstein, P. Franken, and K-H. Kogel (2005). The endophytic fungus Piriformospora indica reprograms barley to salt-stress tolerance, disease resistance, and higher yield. Proc. Natl. Acad. Sci. USA 102:13386–13391.

APPENDIX A

GEOGRAPHIC LOCALITY AND HOST INDENTITY SHAPE FUNGAL

ENDOPHYTE COMMUNITIES IN CUPRESSACEOUS TREES

Dear Ms Hoffman,

Thank you for your email. Please see our author pages http://www.elsevier.com/wps/find/authorsview.authors/copyright#whatrights , on author rights, which explains that you are able to use your paper as part of a dissertation.

Best wishes

Hannah Forder Journal Manager

Pp Caroline Peters, Journal Manager

-----Original Message----- From: [email protected] [mailto:[email protected]] Sent: 29 March 2010 21:33 To: Mycres (ELS) Cc: [email protected] Subject: permission to publish

Please do not reply to this automated email.

First Name Sender: Michele Last Name Sender: Hoffman Email Address: [email protected] Subject: permission to publish Journal ID: 707043 Journal Title: Mycological Research

Text of Email: Dear Sir or Madam, I wish to obtain permission from your journal to reprint the following publication as part of my forthcoming dissertation. Thank you for your time and consideration. Michele Hoffman Hoffman, M. T. and A. E. Arnold. 2008. Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees. Myco. Res. 112:331â344. mycological research 112 (2008) 331–344 26

journal homepage: www.elsevier.com/locate/mycres

Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees

Michele T. HOFFMAN, A. Elizabeth ARNOLD*

Division of Plant Pathology and Microbiology, Department of Plant Sciences, 1140 East South Campus Drive, University of Arizona, Tucson, AZ 85721, USA article info abstract

Article history: Understanding how fungal endophyte communities differ in abundance, diversity, Received 13 March 2006 taxonomic composition, and host afﬁnity over the geographic ranges of their hosts is key Received in revised form to understanding the ecology and evolutionary context of endophyte–plant associations. 28 August 2007 We examined endophytes associated with healthy photosynthetic tissues of three closely Accepted 24 October 2007 related tree species in the Cupressaceae (Coniferales): two native species within their natural Corresponding Editor: Barbara Schulz ranges [Juniperus virginiana in a mesic semideciduous forest, North Carolina (NC); Cupressus arizonica, under xeric conditions, Arizona (AZ)], and a non-native species planted in each Keywords: site (Platycladus orientalis). Endophytes were recovered from 229 of 960 tissue segments Ascomycota and represented at least 35 species of Ascomycota. Isolation frequency was more than three- Biodiversity fold greater for plants in NC than in AZ, and was 2.5 (AZ) to four (NC) times greater for Cupressaceae non-native Platycladus than for Cupressus or Juniperus. Analyses of ITS rDNA for 109 repre- Molecular phylogeny sentative isolates showed that endophyte diversity was more than twofold greater in NC Species richness than in AZ, and that endophytes recovered in AZ were more likely to be host-generalists Symbiosis relative to those in NC. Different endophyte genera dominated the assemblages of each host species/locality combination, but in both localities, Platycladus harboured less diverse and more cosmopolitan endophytes than did either native host. Parsimony and Bayesian analyses for four classes of Ascomycota (Dothideomycetes, Sordariomycetes, Pezizomycetes, Eurotiomycetes) based on LSU rDNA data (ca 1.2 kb) showed that well-supported clades of endophytes frequently contained representatives of a single locality or host species, under- scoring the importance of both geography and host identity in shaping a given plant’s endophyte community. Together, our data show that not only do the abundance, diversity, and taxonomic composition of endophyte communities differ as a function of host identity and locality, but that host afﬁnities of those communities are variable as well. ª 2007 The British Mycological Society. Published by Elsevier Ltd. All rights reserved.

Introduction Despite a strong foundation laid by a growing number of surveys, the degree to which endophytes differ in abundance, Recent studies have begun to elucidate the diversity and ubiq- diversity, taxonomic composition, and host afﬁnity over the uity of fungal endophytes associated with terrestrial plants, geographic ranges of particular host lineages remains unclear. providing a framework for understanding the evolutionary Resolving the importance of locality and host identity in shap- context of these plant–fungus associations (reviewed by Stone ing these aspects of endophyte communities is important for et al. 2000; Schulz & Boyle 2005; Arnold & Lutzoni 2007). understanding fundamental aspects of endophyte symbioses.

* Corresponding author. E-mail address: [email protected] 0953-7562/$ – see front matter ª 2007 The British Mycological Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.mycres.2007.10.014 332 M. T. Hoffman, A. E.27 Arnold

Previous studies have indicated that the abundance, diver- endophyte communities. Yet Cannon & Simmons (2002) re- sity, and species composition of endophytes can be strongly covered similar endophyte communities from phylogeneti- influenced by the locality in which a given plant occurs (e.g. cally diverse angiosperm trees in a tropical forest reserve. Carroll & Carroll 1978; Petrini et al. 1982; Bills & Polishook Suryanarayanan et al. (2004) recovered the same endophyte 1992; Fisher et al. 1994; Hata & Futai 1996; Bayman et al. 1998; species from diverse plant species in mangroves, dry decidu- Arnold et al. 2001; Higgins et al. 2007). For example, different ous forest, and other tropical forest types (Phyllosticta capitalen- microhabitats vary in inoculum potential, with higher infec- sis), and Mohali et al. (2005) found that Lasiodiplodia theobromae tion frequencies observed beneath forest canopies versus in occurs in multiple hosts and sites at a global scale. Suryanar- forest clearings (Arnold & Herre 2003) or in more mesic micro- ayanan et al. (2005) found no evidence for host specificity habitats relative to proximate sites that are more xeric (Petrini among endophytes of cacti in the southwestern USA, yet Hig- et al. 1982; Petrini 1996). At larger geographic scales, endo- gins et al. (2007) found that most genotypes of endophytes re- phytes are frequently more abundant in tropical regions covered from arctic and boreal plants were associated with than in boreal, arctic, or some temperate regions (Arnold & only one host species. Thus, research to date remains in con- Lutzoni 2007), although local climate, short-term weather pat- flict with regard to the prevalence of host specificity among terns, and some disturbance regimes can override latitudinal endophytes, and the potential for host specificity of endo- gradients (see Suryanarayanan et al. 2000; Arnold et al. 2007; phyte communities to differ over the range of a plant taxon, Higgins et al. 2007). Similarly, diversity of endophytes differs or among species within a particular site, is not clear. at small scales as a function of land use history, vegetation To our knowledge, no study to date has concurrently cover, and other factors. For example, Gamboa & Bayman assessed the importance of host identity and locality in shap- (2001) found higher endophyte diversity in leaves of Guarea ing endophyte abundance, diversity, taxon composition, and guidonia in a forest preserve relative to a disturbed forest relative host specificity by simultaneously sampling pairs of area in Puerto Rico. At larger scales, endophyte diversity closely related plants in geographically distinct sites. Molecu- varies as a function of latitude and annual rainfall (Arnold & lar tools are especially useful in this regard, providing a frame- Lutzoni 2007), although the contributions of co-varying fac- work for explicitly comparing sterile endophytes and tors, such as plant diversity, remain to be explored. Finally, providing a tool to disentangle potentially cryptic species for the species composition of endophyte assemblages also dif- which morphological characters are of limited use (Lacap fers as a function of localities. For example, Arnold et al. et al. 2003; Arnold et al. 2007). In turn, phylogenetic analyses (2003) found distinctive endophyte communities associated provide insight into the evolutionary associations of fungal with Theobroma cacao at multiple sites in Panama. In a study lineages with hosts and localities, and can provide much- of endophytes and saprotrophs associated with closely related needed insight into the degree to which particular clades of palms in Australia and Brunei Darussalam, Fro¨ hlich & Hyde fungi are more or less likely to form host-specific associations (1999) found that only 27 of 189 fungal species were shared be- with plants. Unfortunately, endophyte surveys are often re- tween localities. Similarly, Fisher et al. (1995) found that only stricted in terms of molecular phylogenetics due to the cost two of 25 species of endophytes recovered from Dryas octope- associated with sequencing informative loci for numerous tala (Rosaceae) in Switzerland also occurred in the same host and often phylogenetically diverse isolates (see Arnold et al. in Norway. 2007; Higgins et al. 2007). In particular, there is a need to exam- By comparing endophyte assemblages in a single host spe- ine the ability of particular loci to infer topologies that are con- cies or genus in multiple sites, these studies have shown that sistent with reconstructions provided by multi-locus analyses plants interact with distinctive endophyte communities (e.g., Lutzoni et al. 2004; James et al. 2006) and to maximize the across the geographic areas in which they grow. However, it phylogenetic power provided by loci that can be sequenced is unclear whether plants growing in different localities expe- rapidly and at reasonable cost. rience not only a distinctive abundance, diversity, and species Using pairs of related species of the conifer family Cupres- assemblage of endophytes, but also a different degree of rela- saceae from a mesic forest [central North Carolina (NC)] and tive host specificity. Do sympatric plant species differ in terms from a xeric desert environment [southern Arizona (AZ)], of the relative host specificity or host-generalism of the endo- we used a culture-based approach to examine the foliar phytes they harbour? Do different regions or environmental endophyte communities of two native species (Cupressus conditions select for more or less host-specific endophytes? arizonica, AZ and Juniperus virginiana, NC) and a co-occurring, To our knowledge, these questions have not been addressed non-native species (Platycladus orientalis in both AZ and previously. NC). Our objectives were to: (1) examine the abundance, Most studies examining host affinities of endophytes have diversity, taxonomic composition, and host preferences of focused on distantly related host species that co-occur within endophytic fungi of these cupressaceous hosts; (2) infer particular geographic areas, and have yielded conflicting re- the relationships of endophytes associated with these trees sults. For example, a study involving endophytes of mistletoe in a broad phylogenetic context; (3) evaluate an accelerated and fir established that despite close physical proximity of phylogenetic search strategy for rapidly and robustly diag- these two plant taxa, endophyte communities remained ex- nosing these phylogenetic relationships using the LROR- clusively associated with their particular host (Petrini 1996). LR7 region of the LSU rDNA; and (4) generate hypotheses Similarly, Suryanarayanan et al. (2000) found distinctive endo- regarding the relative importance of geographic locality phytes in Cuscuta reflexa relative to its host plants, and Arnold and host identity in structuring the endophyte communi- et al. (2000) found in lowland Panama that distantly related ties associated with these ecologically and economically hosts, growing within metres of one another, bore distinctive important trees. Endophytes of Cupressaceae 28 333

plating on 2 % malt extract agar (MEA). To conﬁrm the efﬁcacy Materials and methods of surface-sterilization, a subset of tissue segments was pressed against the medium under sterile conditions for Host species 30 s, and then removed. No mycelial growth was observed from these surface impressions. A total of 384 tissue segments The cypress family, Cupressaceae (Coniferales), is widely distrib- (96 per individual tree) were plated for each species from AZ, uted in warm-temperate regions and contains 110–130 species and 96 tissue segments (24 per individual tree) were plated of evergreen trees and shrubs in 25–30 genera (Fralish & for each species from NC, corresponding to pilot data regard- Franklin 2002). Within the Cupressoideae, the genera Cupressus ing differences in infection frequency (Arnold & Lutzoni 2007). and Juniperus comprise 70–86 species, whereas Platycladus is Tissue segments were incubated on sealed plates under ambi- monotypic (Platycladus orientalis). Cupressus arizonica (AZ ent light/dark conditions at room temperature (ca 21.5 C). cypress) isindigenousinthe western UnitedStatesand northern Hyphal growth was monitored over an eight-week period. Mexico (Little 1950; USDA, NRCS 2004). Juniperus virginiana (East- Using aseptic technique, fungi were transferred to axenic cul- ern red cedar) is distributed from Nova Scotia to Texas, and is the ture on 2 % MEA in 60 mm Petri plates and photographed after most widespread conifer in the eastern USA (Fralish & Franklin one week of growth. All samples were archived as living 2002). Platycladus orientalis is native to Korea, eastern Russia, and vouchers in sterile water at the Robert L. Gilbertson Mycolog- China. It has been extensively cultivated in southeastern Asia ical Herbarium at the University of AZ. Based on mycelial and has become naturalized in Florida (Little 1980; USDA, characteristics (Arnold et al. 2000), representatives of all NRCS 2004). It is commonly planted as an ornamental in both unique morphotypes were selected for molecular analysis. AZ and NC.

Genomic DNA extraction and PCR Sample collection

Total genomic DNA was extracted directly from pure cultures In March and April 2005, fresh, asymptomatic photosynthetic following Arnold et al. (2007). PCR was used to amplify the ITS tissue was collected from healthy trees at the campus arbore- rDNA and 5.8s gene (ca 600 bp) and LSU rDNA (ca 1.2 kb) fol- tum at the University of AZ (Tucson: Platycladus and Cupressus; lowing Higgins et al. (2007). Primers included ITS1F or ITS5 four individuals/species) and at Duke Forest, Duke University and ITS4 for ITS rDNA (White et al. 1990; Vilgalys & Hester (Durham, NC: Platycladus and Juniperus; four individuals/spe- 1990; Gardes & Bruns 1993), and LROR and LR7 for LSU rDNA cies). The University of AZ (32.231N, 110.952W, elevation (Vilgalys, unpubl.; Vilgalys & Hester 1990). Sigma Readymix 787 m; mean annual temperature ¼ 20.2 C) is located within REDTaq PCR reaction mix with MgCl (St. Louis, MO) was the Sonoran Desert bioregion. This site is relatively xeric, re- 2 used for all PCR reactions. The 25 ml reaction mixture included ceiving an average of 30.5 cm of precipitation annually. Dur- 12.5 ml REDTaq,1ml of each primer (10 mM), 1 ml DNA tem- ham, NC (36.001N, 78.940W, elevation 97 m; mean annual plate, and 9.5 ml PCR-quality water. Cycling reactions were temperature ¼ 15.5 C) is relatively mesic, receiving an aver- run on an MJ Research PTC200 thermocycler (Waltham, age of 109.2 cm of precipitation annually. At the University MA) with the following protocol: 94 C for 3 min; 36 cycles of of AZ, all focal plants were within 250 m of one another, and 94 C for 30 s, 54 C for 30 s, 72 C for 1 min; and 72 C for were cultivated in mixed, open stands with a variety of native 10 min. SYBR Green I stain (Molecular Probes, Invitrogen; and non-native ornamental plants. Focal individuals were Carlsbad, CA) was used to detect DNA bands on a 1 % agarose fully established and mature, and received minimal supple- gel. All products demonstrated single bands. mental water through infrequent irrigation. At Duke Univer- sity, all Juniperus individuals were located beneath an open canopy of Pinus taeda interspersed with Liquidambar styraciflua, Sequencing and analyses Carya spp., Quercus spp., Platanus occidentalis, and Acer spp. At that site, all Platycladus individuals were located within PCR products were cleaned, quantified, and normalized at the 200 m of native J. virginiana and were planted adjacent to the GATC sequencing facility at the University of AZ. Sequencing forest edge. was performed on the Applied Biosystems 3730XL DNA From each individual, we harvested mature photosyn- Analyser (Foster City, CA). The software applications phred thetic tissues from three branches extending in three ran- and phrap (Ewing & Green 1998; Ewing et al. 1998) were used domly chosen cardinal directions. Material was bagged for to call bases and assemble contigs, with automation provided transport to the laboratory and was used in endophyte isola- by the ChromaSeq package (Maddison & Maddison: http:// tions within 4 h of collection. mesquiteproject.org) implemented in Mesquite v. 1.06 (Maddison & Maddison: http://mesquiteproject.org). All base Endophyte isolations calls were verified by inspection of chromatograms in Sequencher version 4.5 (Gene Codes, Ann Arbor, MI). Photosynthetic tissue from each sample was rinsed in run- ITS rDNA sequences were obtained for 109 representative ning tap water for 30 s and then cut into 2 mm pieces before isolates (Supplementary Material Appendix 1). LSU rDNA se- surface-sterilization. Tissue fragments were agitated in 95 % quences were obtained for 64 representative isolates (Supple- ethanol for 30 s, 10 % bleach (0.5 % NaOCl) for 2 min, and mentary Material Appendix 2) based on ITS rDNA genotype 70 % ethanol for 2 min (Arnold et al. 2007). Under sterile condi- groups (described below). Sequence data generated for this tions, tissue segments were allowed to surface-dry before study are available from GenBank under accession numbers 334 M. T. Hoffman, A. E.29 Arnold

EF419892–EF420019 (ITS rDNA sequences) and EF420020– Manual alignment was performed using MacClade 4.08 EF420083 (LSU rDNA sequences). (Maddison & Maddison 2005). Each alignment was based on LSU rDNA secondary structure as defined by Saccharomyces cer- Endophyte richness, diversity, and similarity evisiae (Cannone et al. 2002). All ambiguously aligned regions among hosts and sites were excluded. The alignment for the Dothideomycetes consisted of 4595 characters, of which 1208 were included; for ITS rDNA sequence data were used in BLAST searches of the Sordariomycetes, 4605 characters, of which 1218 were in- GenBank to provide preliminary identification at higher taxo- cluded; for the Eurotiomycetes, 5443 characters, of which 2056 nomic levels and to guide taxon sampling for LSU rDNA anal- were included; for the Pezizomycetes, 4595 characters, of which yses. To designate operational taxonomic units (OTU) based 1208 were included. on ITS rDNA data, we used Sequencher to delimit groups cor- For each dataset, initial heuristic searches using parsi- responding to 90, 95, and 99 % ITS rDNA similarity without mony as the optimality criterion were implemented in PAUP considering differences in sequence length (Arnold et al. 4b10 (Swofford 2002), but these searches were ineffective in 2007). Previous studies have used 90–97 % ITS rDNA sequence recovering optimal trees after up to five days. To improve similarity as a proxy for species boundaries in Fungi (e.g. 97 %: the quality of searches, we initiated 30 sets of 200 searches O’Brien et al. 2005). One study has explicitly shown that groups for each dataset using PAUPRat (Sikes & Lewis 2001), which based on 90 % ITS rDNA similarity are highly congruent with implements the ‘parsimony ratchet’ method (Nixon 1999)to phylotypes based on a second locus (Arnold et al. 2007). Based efficiently recover shortest trees in datasets that are challeng- on these studies, we used 90 % ITS rDNA genotype groups as ing for traditional heuristic searches (Sikes & Lewis 2001). The a highly conservative proxy for species in all subsequent ‘filter trees’ command in PAUP 4b10 (Swofford 2002) was used analyses. to select all shortest trees from the resulting pools of trees Species accumulation curves and bootstrap estimates of (6001 trees per data set). The resulting set of shortest trees total species richness were inferred using EstimateS v. 7.5 was then used to construct a strict consensus tree. Support (Colwell 2005: http://viceroy.eeb.uconn.edu/EstimateS). In was measured using a neighbor-joining bootstrap (1K addition to assessing richness (number of species) for each replicates). partition of the data, we also calculated diversity, which takes To verify the quality of the topology given the PAUPRat into account the abundance of different species. Diversity was approach, we conducted an additional set of analyses for measured using Fisher’s a (Fisher et al. 1943), which is robust each dataset using Bayesian MCMCMC. Based on compari- for small sample sizes and thus accommodates a wide range sons of negative log likelihood values in Modeltest 3.7 of sampling effort (Higgins et al. 2007). (Posada & Crandall 1998), GTR þ I þ G was chosen as the All ITS rDNA genotypes (based on 90 % sequence similarity) best-fitting model for each dataset (Supplementary Material that were recovered more than once were compared for all Appendix 4). Searches were implemented in MrBayes v. host-locality combinations. Jaccard’s index was used to assess 3.1.1 (Huelsenbeck & Ronquist 2001) for 5M generations, initi- similarity on the basis of presence/absence data only (see ated with random trees, four chains, and sampling every Arnold et al. 2000). The Morisita–Horn index was used to as- 500th tree. Extension of each analysis by up to 5M generations sess similarity on the basis of isolation frequency (see Gamboa had no significant effect on negative log likelihood values. Af- & Bayman 2001; Arnold et al. 2003). Similarity indices were ter elimination of the burn-in for each analysis, a majority inferred using EstimateS v.7.5 (Colwell 2005: http://viceroy. rule consensus based on remaining trees was inferred. eeb.uconn.edu/EstimateS).

Phylogenetic analyses Results

Sixty-four LSU rDNA sequences for representative endo- From 960 tissue samples, 229 isolates of endophytic fungi phytes were organized at the class level on the basis of BLAST were recovered in culture. Isolation frequency, deﬁned as results and integrated into existing core alignments for the the percent of tissue segments bearing cultivable endophytes, Dothideomycetes, Sordariomycetes, Eurotiomycetes, and Pezizomy- was more than threefold greater for hosts from NC (56.3 %) cetes. Core alignments included representative endophytes than AZ (15.8 %), and was 2.5 to four times greater in Platycla- from boreal, temperate, and tropical sites (Arnold et al., in dus (AZ, 25.5 %; NC, 80.2 %) than in Cupressus (AZ; 6 %) or review; Arnold & Lutzoni 2007; Higgins et al. 2007; Supple- Juniperus (NC; 32.4 %; Table 1). mentary Material Appendix 2) as well as named species of Among 109 isolates sequenced for ITS rDNA, we recov- Ascomycota obtained from GenBank (Supplementary Material ered 35 unique ITS rDNA genotypes (90 % ITS rDNA sequence Appendix 3), and were subjected to preliminary analyses to similarity; Fisher’s a ¼ 17.9). A total of 43 and 64 ITS rDNA ensure that the initial BLAST results were accurate (Higgins genotypes based on 95 % and 99 % sequence similarity was et al. 2007). Taxon sampling for the Dothideomycetes consisted recovered (Fisher’s a ¼ 26.2, and 65.1, respectively). of 26 named sequences and 78 endophytes (46 from this Diversity of endophytes differed as a function of locality study); for the Sordariomycetes, 59 named sequences and 34 and host identity. Diversity of endophytes from hosts in endophytes (15 from this study); for the Eurotiomycetes,19 NC was more than twofold greater than for hosts from AZ named sequences and three endophytes (one from this (Table 1). The total diversity of endophytes recovered from study); and the Pezizomycetes, 23 named sequences and four Juniperus and Cupressus was 1.4 times greater than from Platy- endophytes (two from this study). cladus (Table 1). In NC, diversity was ca twofold greater in Endophytes of Cupressaceae 30 335

Table 1 – Number of tissue segments examined, isolates recovered, and isolates sequenced for ITS rDNA; diversity (Fisher’s alpha) of endophytes using functional taxonomic units based on 90, 95, and 99 % ITS rDNA sequence similarity; and number of singletons for each group based on 90 % ITS rDNA sequence similarity Host taxon/location Segments Isolates Isolates Fisher’s alpha Singletons plated recovered sequenced (90 % ITS rDNA) 90 % 95 % ITS 99 % ITS ITS rDNA rDNA rDNA

Cupressus/AZ 384 23 13 3.0 13.0 33.8 1 Juniperus/NC 96 31 18 21.1 42.4 42.4 6 Platycladus/AZ 384 98 28 4.6 6.7 18.4 4 Platycladus/NC 96 77 50 8.2 9.2 22.1 9

Cupressus & Juniperus 480 54 31 17.9 48.9 75.9 7 Platycladus 480 175 78 12.7 15.6 39.5 13 Cupressus & Platycladus (AZ) 768 121 41 7.5 11.2 38.8 5 Juniperus & Platycladus (NC) 192 108 68 19.1 23.6 43.7 15

Hosts include Cupressus arizonica, Arizona (AZ); Juniperus virginiana, North Carolina (NC), and Platycladus orientalis in AZ and NC.

Juniperus than in Platycladus regardless of the stringency of ITS Despite a large number of shared species between hosts rDNA groups. In AZ, diversity in Cupressus was 1.6 times less within each site, the dominant endophyte taxon associated than in Platycladus when species boundaries were based on with each host and locality was distinct (Supplementary 90 % ITS rDNA similarity. However, when species boundaries Material Appendix 1). In all cases, the most abundant taxa were defined at greater levels of ITS rDNA similarity (i.e. 95 were Dothideomycetes. Cupressus (AZ) was dominated by an and 99 % ITS rDNA congruence), diversity was ca twofold unidentified dothideomycete (90 % ITS rDNA genotype I). greater in Cupressus than in Platycladus (Table 1). The highly diverse community of endophytes associated Endophyte species richness also differed as a function of with Juniperus (NC) was not characterized by a dominant locality and host identity (Fig 1A–F). Greater species richness species, although three isolates (of 18 sequenced) represented per sampling effort was recovered in NC versus AZ (Fig 1A– a dothideomycete with phylogenetic affinity to Letendraea B), in Juniperus and Cupressus versus Platycladus (Fig 1C–D), (90 % ITS rDNA genotype E; see below). Platycladus in NC was and in Juniperus in NC relative to any other species/locality dominated by Phyllosticta sp. (90 % ITS rDNA genotype C; 21 combination (Fig 1E–F). Overall, observed species richness of 68 isolates), and in AZ by Aureobasidium sp. and Phoma sp. fell within the 95 % confidence intervals for estimated (90 % ITS rDNA genotypes F and A, respectively, comprising richness, suggesting that our sample was effective in captur- ten and nine isolates of 41 sequenced). ing the species richness of these endophyte communities (Fig 1G). However, the relatively steep slope of the accumula- Phylogenetic relationships of endophytes tion curve indicates that numerous species of endophytes are yet to be recovered from these hosts as a whole. Results of phylogenetic analyses of endophytes from Cupressa- When genotype accumulation curves were compared for ceae, representative endophytes from other host species and ITS rDNA groups based on different levels of sequence similar- sites, and representative Ascomycota for each class are shown ity, we found the largest differentiation between the 95 and in Figs 2–5. The majority of endophytes recovered in this study 99 % groupings, with the most marked difference occurring were placed with support in the Dothideomycetes (46 isolates) at ca 98 % similarity (Fig 1H). This reflects that most of the en- and Sordariomycetes (15 isolates). Only one endophyte (9147 dophytes recovered in this study are clustered in phylogenet- from Platycladus in NC) was recovered as a member of the ically distinct groups that share high (98 %) ITS rDNA Eurotiomycetes. No endophytes were recovered among the Leo- sequence similarity (see also Figs 2–5). tiomycetes, or from the lichen-dominated clades Lecanoromy- cetes, Lichinomycetes, and Arthoniomycetes. Two endophytes Relative importance of locality versus host identity from Platycladus in AZ formed a clade within a well-supported lineage of Pezizomycetes. Similarity indices based on both presence/absence data (Jac- Phylogenetic analyses of the Dothideomycetes (Fig 2) using card’s index) and isolation frequencies (Morisita–Horn index) the parsimony ratchet yielded 780 best trees with tree lengths and using 90 % ITS rDNA groupings suggest that endophyte of 731 steps. BS support 70 % was observed for 12 nodes. communities in these hosts are structured more by locality Bayesian PP values 95 % were observed for 65 nodes, includ- than by host identity (Table 2). Highest similarity was ob- ing all nodes with strong BS support. Many endophytes recov- served among samples from different host species in the ered in this study were reconstructed as part of larger clades same site (i.e. Platycladus and Cupressus in AZ; Platycladus of endophytic fungi from diverse hosts and sites that were and Juniperus in NC). In contrast, samples from the same strongly supported as sister to Botryosphaeria ribis (clade A), host species in different sites had low similarity values (i.e. reconstructed in a clade containing Discosphaerina fagi and Platycladus in NC and AZ). When compared across study sites, a variety of boreal forest endophytes (clade B), or strongly sup- Juniperus and Cupressus did not share more endophytes with ported as part of a clade containing Phoma glomerata and one another than with Platycladus. a conifer endophyte from boreal forest (clades E and F). The 336 M. T. Hoffman, A. E.31 Arnold

35 35 A Native & non-native B Native & non-native 30 Arizona 30 North Carolina 25 25

20 20

15 15

10 10

5 5

0 0 0 10 20 30 40 50 60 70 80 90 0 10 20 30 40 50 60 70 80 90

35 35 C All native D All non-native 30 AZ & NC 30 AZ & NC

25 25

20 20

15 15

10 10

5 5

0 0 0 102030405060708090 0 102030405060708090

35 35 E F 30 30 ITS rDNA genotypes 25 25

20 20 Platycladus NC 15 15 Juniperus NC

10 Platycladus AZ 10

5 Cupressus AZ 5

0 0 05 101520253035404550 05 10 15 20 25 30 35 40 45 50

50 70 GH99% ITS rDNA genotype 60 40 50 30 95% ITS rDNA genotype 40 90% ITS rDNA genotype 20 30 20 10 10

0 0 0 25 50 75 100 125 0 25 50 75 100 125 Isolates

90% ITS rDNA genotype 95% confidence intervals Bootstrap

Fig 1 – Genotype accumulation curves for fungal endophytes of Cupressus arizonica (AZ), Juniperus virginiana (NC), and Platycladus orientalis (AZ and NC). (A–G) Genotype groups are based on 90 % ITS rDNA sequence similarity, which is used here as a conservative proxy for species boundaries (following Arnold et al. 2007). (A) Species accumulation for endophytes of hosts in AZ. (B) Species accumulation for endophytes of hosts in NC. (C) Species accumulation of endophytes of native hosts: Cupressus (AZ) and Juniperus (NC). (D) Species accumulation of endophytes from non-native Platycladus (AZ and NC). (E) Comparison of species accumulation curves for each host species in AZ. (F) Comparison of species accumulation curves for each host species in NC. (G) Accumulation of ITS rDNA genotypes over the entire study. (H) Species accumulation and richness inferred using groups based on 90, 95, and 99 % ITS rDNA similarity. Endophytes of Cupressaceae 32 337

Dendrographa minor Botryosphaeria ribis 100/ 96 469 Laetia thamnia - Panama 96/ - 2834 Dryas integrifolia - Québec 98/ - 100/ - 2779 Dryas integrifolia - Québec 3300 Dryas integrifolia - Québec 1182 Magnolia grandiflora - North Carolina 100/ - 6730 Swartzia simplex - Panama 6731 Swartzia simplex - Panama 435b Laetia thamnia - Panama 100/ - 9149b Platycladus orientalis - NC Δ C 100/ - 9130 Platycladus orientalis - NC Δ C 100/ - 9144 Platycladus orientalis - NC Δ C 9135 Platycladus orientalis - NC Δ C A 9134 Platycladus orientalis - NC Δ C 9129 Platycladus orientalis - NC Δ C Microxyphium citri 100/ - Raciborskiomyces longisetosum 100/ - 9128 Platycladus orientalis - NC Δ U H 100/ 100 2712 Huperzia selago - Québec 100/ - 2722 Huperzia selago - Québec 100/98 462 Laetia thamnia - Panama 100/ - 4140 Dryas integrifolia -Québec 100/ - 3358A Dryas integrifolia - Québec 4221 Dryas integrifolia - Québec 100/ 99 100/ - 3326 Picea mariana - Québec 5007 Picea mariana - Québec 100/ - 95/ - Sydowia polyspora 1029 Magnolia grandiflora - North Carolina 100/ - 4109 Huperzia selago - Québec 100/ - Delphinella strobiligena 100/ - 4947A Picea mariana - Québec 4089 Picea mariana - Québec 96/ - 100/ - Dothidea insculpta Dothidea sambuci 98/ - Dothidea ribesia Stylodothis puccinioides 9096 Platycladus orientalis - AZ Δ n/a Discosphaerina fagi 3275 Dryas integrifolia - Québec 100/ - 3335 Dryas integrifolia - Québec 3353 Dryas integrifolia - Québec 100/ - 9042 Platycladus orientalis - AZ Δ F 9060 Platycladus orientalis - AZ Δ F 100/ - 9079 Platycladus orientalis - AZ Δ n/a 9084b Platycladus orientalis - AZ Δ n/a B 100/ - 9002 Platycladus orientalis - AZ Δ F 100/ - 9009b Platycladus orientalis - AZ Δ F 9028 Platycladus orientalis - AZ Δ F 9031 Platycladus orientalis - AZ Δ F 9036 Platycladus orientalis - AZ Δ F 100/ - 3329 Picea mariana - Québec 3377 Picea mariana - Québec 100/ - Lojkania enalia 5718 Dryas integrifolia - Iqaluit 100/ 100 Trematosphaeria heterospora 100/ - 95/ - 5095 Picea mariana - Québec 99/ - Westerdykella cylindrica Preussia terricola 100/ - 100/87 9106 Cupressus arizonica - AZ M 100/ - 9116 Cupressus arizonica - AZ M 99/ - 9120 Cupressus arizonica - AZ M C 100/ - 9104 Cupressus arizonica - AZ M 9051 Platycladus orientalis - AZ Δ M Byssothecium circinans 100/ - 3323 Dryas integrifolia - Québec 100/ - Bimuria novae-zelandiae Letendraea helminthicola 100/ - 9203 Juniperus virginiana - NC E 100/90 9200 Juniperus virginiana - NC E 9149a Platycladus orientalis - AZ Δ E 100/70 9137 Platycladus orientalis - NC Δ E D 9145 Platycladus orientalis - NC Δ E 9140 Platycladus orientalis - NC Δ E Phoma glomerata 97/ - 100/ - 9158 Platycladus orientalis - NC Δ A 100/ - 96/ - 9054 Platycladus orientalis - NC Δ A 9107 Cupressus arizonica - AZ A E 9065 Platycladus orientalis - AZ Δ A 5654 Picea mariana - Québec 100/ - 9005 Cupressus arizonica - AZ A 9008 Cupressus arizonica - AZ A 96/ - 9007 Platycladus orientalis - AZ Δ A F 9018 Platycladus orientalis - AZ Δ A 98/ - 9015 Platycladus orientalis - AZ Δ F 9027 Platycladus orientalis - AZ Δ A 9030 Platycladus orientalis - AZ Δ A 100/ 95 Curvularia brachyspora 99/96 Cochliobolus heterostrophus 100/ - Setosphaeria monoceras Pyrenophora tritici-repentis 98/ - 9055 Platycladus orientalis - AZ Δ B 100/83 100/ - 9026 Platycladus orientalis - AZ Δ B 100/ - 9021 Platycladus orientalis - AZ Δ B 9009a Platycladus orientalis - AZ Δ B G 100/ - 9057 Cupressus arizonica - AZ I 98/97 9059 Cupressus arizonica - AZ I 9058 Cupressus arizonica - AZ I 100/ - Setomelanomma holmii Ampelomyces quisqualis 98/ - Phaeosphaeria avenaria 2851 Huperzia selago - Québec 100/ - 2706 Huperzia selago - Québec 5246 Huperzia selago - Québec

Fig 2 – Majority rule consensus tree based on Bayesian analyses of LSU rDNA data for the Dothideomycetes. Numbers above branches indicate branch support: Bayesian PP values 95 % are shown before the slash; NJ BS values 70 % are shown after the slash. Branches recovered in strict consensus tree from parsimony analyses are shown with bold black lines. Circles after endophyte culture numbers indicate native host species; triangles indicate non-native hosts. Letters indicate 90 % ITS rDNA genotype groups based on sequence similarity (Supplementary Material Appendix 1). Endophytic fungi from other studies are listed in Supplementary Material Appendix 2. All named taxa and GenBank identiﬁcation numbers are shown in Sup- plementary Material Appendix 3. Dendrographa minor was included as an outgroup. 33

Leotia lubrica 100/100 Lulworthia fucicola Lulworthia grandispora Oxydothis frondicola 95/ - Microdochium nivale Hyponectria buxi 100/ - 100/77 4653 Picea mariana - Québec 97/ - 96/100 Arthrinium phaeospermum 100/ - Apiospora sinensis 100/93 6722 Faramea occidentalis - Panama 100/ - 9143 Platycladus orientalis - NC J 97/ - 95/71 9089 Platycladus orientalis - AZ R - /100 9126 Platycladus orientalis - NC J A 9121 Platycladus orientalis - NC J Cryptosphaeria eunomia Fasciatispora petrakii 99/ - Cainia graminis Seynesia erumpens 100/ - Hypoxylon fragiforme 95/ - 100/ - 256 Laetia thamnia - Panama 2204 Pinus taeda - North Carolina 100/ - 6717 Gustavia superba - Panama 100/100 9202 Juniperus virginiana - NC L 9201 Juniperus virginiana - NC L B 100/ - Astrocystis cocoes 99/ - Rosellinia necatrix 99/ - Xylaria acuta Xylaria hypoxylon 6728 Faramea occidentalis - Panama Halorosellinia oceanica Thyridium vestitum Gaeumannomyces graminis var graminis 100/100 6702 Trichilia tuberculata - Panama Diaporthe phaseolorum 100/ - 100/100 Valsa ambiens 78/ - 98/90 Cryptodiaporthe corni 100/70 Cryphonectria cubensis 100/ - Cryphonectria havanensis 97/ - 100/100 Melanconis marginalis 100/97 Discula destructiva Gnomonia setacea 100/80 Gnomoniella fraxini Plagiostoma euphorbiae Cephalotheca sulfurea Menispora tortuosa 100/ - Camarops microspora 100/88 Farrowia longicollea Farrowia seminuda 97/ - 100/95 Neurospora crassa 100/100 Sordaria macrospora Sordaria fimicola Chaetomium globosum 97/ - 9038 Platycladus orientalis - AZ P 96/ - 9097 Cupressus arizonica - AZ T 100/ - 9069 Platycladus orientalis - AZ H 9093 Platycladus orientalis - AZ H C 9094 Platycladus orientalis - AZ H 9092 Platycladus orientalis - AZ S 9085 Platycladus orientalis - AZ H Pleurothecium recurvatum 6733 Swartzia simplex - Panama 100/100 412 Laetia thamnia - Panama 6756 Gustavia superba - Panama 6741 Theobroma cacao - Panama 6749 Theobroma cacao - Panama 6714 Gustavia superba - Panama 151 Laetia thamnia - Panama 100/ - Microascus trigonosporus 100/73 Ascosalsum cincinnatulum Magnisphaera stevemossago 100/ - 100/83 Ascosacculus heteroguttulatus 99/96 Natantispora lotica Halosarpheia marina Sagaaromyces abonnis 100/100 Corollospora maritima Varicosporina ramulosa 100/- Torrubiella luteorostrata Hydropisphaera erubescens Cordyceps khaoyaiensis 100/99 425 Laetia thamnia - Panama 4939 Picea mariana - Québec 100/77 416 Laetia thamnia - Panama 100/ - 6708 Gustavia superba - Panama Balansia henningsiana 100/ - Hypocrea citrina Verticillium epiphytum Emericellopsis terricola 100/99 9154 Platycladus orientalis -NC K 9168 Platycladus orientalis - NC K D 4780 Dryas integrifolia - Iqaluit 100/ - Hypomyces polyporinus Cordyceps irangiensis Cordyceps sinensis

Fig 3 – Majority rule consensus tree based on Bayesian analyses of LSU rDNA data for the Sordariomycetes. Numbers above branches indicate branch support: Bayesian PP values 95 % are shown before the slash; NJ BS values 70 % are shown after the slash. Branches recovered in strict consensus tree from parsimony analyses are shown with bold black lines. Circles after endophyte culture numbers indicate native host species; triangles indicate non-native hosts. Letters indicate 90 % ITS rDNA genotype groups based on sequence similarity (Supplementary Material Appendix 1). Endophytic fungi from other studies are listed in Supplementary Material Appendix 2. All named taxa and GenBank identiﬁcation numbers are shown in Sup- plementary Material Appendix 3. Leotia lubrica was included as an outgroup. 34

Peltula umbilicata

Arthroderma curreyi

100/98 Auxarthron zuffianum

Aphanoascus fulvescens 100/88

Coccidioides posadasii - / 93

Paracoccidioides brasiliensis

100/90 Ascosphaera apis 100/ -

Eremascus albus

Byssochlamys nivea

100/ - 100/96 Penicillium freii 100/87

9147 Platycladus orientalis - NC Δ J A

Pyrenula cruenta

Pyrenula pseudobufonia

100/97 4466 Picea mariana - Québec

Verrucaria pachyderma 100/ -

Ceramothyrium carniolicum

100/ - Capronia mansonii

99/ - 422 Laetia thamnia - Panama

99/ - Exophiala jeanselmei

Glyphium elatum

Capronia pilosella 99/ -

Dermatocarpon luridum

Fig 4 – Majority rule consensus tree based on Bayesian analyses of LSU rDNA data for the Eurotiomycetes. Numbers above branches indicate branch support: Bayesian PP values 95 % are shown before the slash; NJ BS values 70 % are shown after the slash. Branches recovered in strict consensus tree from parsimony analyses are shown with bold black lines. Triangle indicates non-native host. Letters indicate 90 % ITS rDNA genotype groups based on sequence similarity (Supplementary Material Appendix 1). Endophytic fungi from other studies are listed in Supplementary Material Appendix 2. All named taxa and GenBank identiﬁcation numbers are shown in Supplementary Material Appendix 3. Peltula umbilicata was included as an outgroup. 35

Candida albicans

Saccharomyces cerevisiae

Schizosaccharomyces pombe

Neolecta vitellina

Taphrina communis

Ascobolus carbonarius 100/100

Ascobolus crenulatus

9098 Platycladus orientalis - AZ Δ N/A 100/91 100/ - A 9084a Platycladus orientalis - AZ Δ Q

100/100 Sarcosphaera crassa

Peziza succosa

4250 Dryas integrifolia - Iqaluit

Peziza proteana 100/72 Peziza quelepidotia

Sarcoscypha coccinea

95/ - 4245 Dryas integrifolia - Iqaluit

100/ - Otidea onotica 97/ - Aleuria aurantia 99/ -

97/94 Cheilymenia stercorea

Barssia oregonensis 100/100

Helvella compressa

100/70 Gyromitra californica 100/93

Gyromitra esculenta

100/100 Disciotis venosa -/99

Verpa conica 100/ -

Morchella elata 100/100

Morchella esculenta

Fig 5 – Majority rule consensus tree based on Bayesian analyses of LSU rDNA data for the Pezizomycetes. Numbers above branches indicate branch support: Bayesian PP values 95 % are shown before the slash; NJ BS values 70 % are shown after the slash. Branches recovered in strict consensus tree from parsimony analyses are shown with bold black lines. Triangles indicate non-native hosts. Letters indicate 90 % ITS rDNA genotype groups based on sequence similarity (Supplementary Material Appendix 1). Endophytic fungi from other studies are listed in Supplementary Material Appendix 2. All named taxa and GenBank identiﬁcation numbers are shown in Supplementary Material Appendix 3. The ﬁve taxa at the base of the tree were included as outgroups. Endophytes of Cupressaceae 36 341

by both measures. Endophyte 9147 was reconstructed with Table 2 – Similarity of endophyte assemblages with regard to locality and host identity based on non- strong support as sister to Penicillium freii within the well- singleton ITS rDNA genotypes (90 % sequence similarity) supported Eurotiales. For the Pezizomycetes (Fig 5), the parsimony ratchet ap- Platycladus Platycladus Juniperus Cupressus NC AZ NC AZ proach yielded 827 best trees with tree lengths of 561 steps. BS support 70 % was observed for 11 nodes; Bayesian PP Platycladus NC 0.05 0.24 0.04 95 % was observed for 16 nodes. Ten nodes received signifi- Platycladus AZ 0.09 0.12 0.42 cant support from both methods. Endophytes 9098 and Juniperus NC 0.40 0.09 0.09 Cupressus AZ 0.10 0.50 0.10 9084a from Platycladus in AZ were reconstructed as sister to one another with strong support, and are nested within Similarity values reflect the Morisita–Horn index (above diagonal: a strongly supported clade within the Pezizales. index based on isolation frequencies) and Jaccard’s index (below diagonal: index based on presence/absence data only). Both indices range from 0 (no overlap between communities) to 1 (total overlap between communities). Highest similarities for each comparison Discussion are shown in bold. AZ, Arizona; NC, North Carolina. We used a culture-based approach, paired with sequencing of both a fast-evolving locus (ITS rDNA) and a phylogenetically informative locus (LSU rDNA), to assess the diversity and majority of endophytes within the Dothideomycetes were taxon composition of fungal endophytes among native and reconstructed with strong support as members of the well- non-native conifers in two geographically distinct localities. supported Pleosporales (clades C–G). One additional endophyte We found that the abundance, diversity, species composition, (H) was reconstructed with strong support as the sister taxon and relative host affinity of endophyte communities differed of Raciborskiomyces longisetosum. Clade C was reconstructed as as a function of locality and host species. Cultivable endo- sister to Preussia terricola, albeit without strong support. Clade phytes were present in up to 80.2 % of tissue segments exam- D was reconstructed as sister to Letendraea helminthicola. ined, but their incidence differed more than 13-fold as Clades E and F were strongly supported as a clade with Phoma a function of locality and host identity. Despite very conserva- glomerata and included an endophyte from Picea mariana in tive estimates of species boundaries, diversity ranged from Que´bec. Clade G was well supported as sister to Pyrenophora low values (Fisher’s a ¼ 3: Cupressus, AZ) to values consistent tritici-repentis. Throughout the Dothideomycetes, ITS rDNA ge- with those of angiosperm trees in tropical forests (Fisher’s notype groups (based on 90 % ITS rDNA similarity) were a ¼ 21.1: Juniperus, NC; Arnold & Lutzoni 2007). Most endo- largely congruent with LSU rDNA phylotypes. Phylotypes con- phytes were placed in phylogenetic analyses within the Dothi- taining endophytes from cupressaceous hosts often reflected deomycetes and, to a lesser extent, the Sordariomycetes, but locality (AZ versus NC; e.g. clades A, B), but within-clade struc- members of both the Eurotiomycetes and Pezizomycetes were ture based on locality and host identity was suggested in sev- also recovered. Isolates representing the Dothideomycetes eral clades (e.g. clades D–G). were proportionally more common in AZ than NC (ca 88 versus Analyses of the Sordariomycetes (Fig 3) using the parsimony ca 75 % of isolates with definitively identified ITS rDNA geno- ratchet yielded 626 best trees with tree lengths of 1289 steps. types, respectively; Supplementary Material Appendix 1, Figs BS support 70 % was observed for 25 nodes, and Bayesian 2–5). The Sordariomycetes were proportionally more common PP 95 % was observed for 53 nodes. Twenty-four nodes in NC than AZ (ca 21 versus ca 10 %, respectively). Overall and were supported by both methods. Endophytes from cupressa- for three host/locality combinations (Cupressus AZ, Platycladus ceous hosts were reconstructed within the well-supported AZ, and Platycladus NC), the majority of definitively identified Xylariales (clades A–B), the Sordariales (clade C), and the Hypo- isolates were placed in the Dothideomycetes. However, Juniperus creales (clade D). Clades A and B were reconstructed with sup- from NC was characterized by a higher incidence of Sordario- port as sister to endophytes from Panama. Clade C was mycetes (ca 56 % of identified isolates; Supplementary Material reconstructed as sister to the ubiquitous fungus Chaetomium Appendix 1). globosum, and clade D as sister to Emericellopsis, albeit without Although endophyte genotypes often occurred in more strong support. ITS rDNA genotype groups were consistent than one host species (Supplementary Material Appendix 1) with LSU rDNA phylotypes for the hypocrealean endo- and clades often contained representatives of multiple hosts phytes (clade D) and xylarialean endophytes in clade B. and sites (Figs 2–5), the occurrence of numerous singleton spe- However, clades A (Xylariales) and C (Sordariales) were char- cies (Table 1, Fig 1) and relatively low similarity among hosts acterized by poor matches between ITS rDNA genotype and localities (Table 2) suggest that each tree species in each groups and LSU rDNA phylotypes. Phylotypes reflected clus- locality has a signature community of endophytic symbionts. tering of endophytes from different host taxa (clade C) and This was underscored by the observation that the dominant sites (clade A) although two well-supported phylotypes endophyte species associated with each host was distinct (clades B, D) contained endophytes from a single host spe- (Supplementary Material Appendix 1). The conservative species in a single site. cies concept applied here likely underestimates richness and For the Eurotiomycetes (Fig 4), the parsimony ratchet ap- diversity (Table 1), and may overestimate similarity of proach yielded 1005 best trees with tree lengths of 387 steps. communities among hosts and sites (Table 2). Diversity of BS support 70 % was observed for seven nodes, and Bayesian endophytes in these temperate hosts thus appears to be sur- PP 95 % was observed for 13 nodes. Six nodes were supported prisingly high, with a high degree of geographic turnover 342 M. T. Hoffman, A. E.37 Arnold

and distinctively structured communities in each host and a search strategy for parsimony analyses involving en- locality. vironmental samples (e.g. unknown endophyte cultures). Fisher et al. (1993, 1994) found that endophyte assemblages That study concluded that additional data were needed to of trees planted outside their native range are depauperate in more adequately infer phylogenetic relationships for large terms of infection frequency, species richness, and prevalence datasets. In this study, we found that ca 1200 bp of LSU of host-specific endophytes. Although our study did not ex- rDNA, coupled with the PAUPRat method, provided the tools amine endophytes of Platycladus within its native range, we for expediently recovering topologies within four classes of did compare endophytes of Platycladus with those of two Ascomycota. These topologies were not in conflict with topolo- closely related, native hosts in a mesic forest and an arid des- gies resulting from Bayesian analysis (Figs 2–5). However, ert. In both sites, isolation frequency was higher in Platycladus Bayesian methods consistently yielded greater numbers of than in either native host. Moreover, Platycladus in NC significantly supported nodes than did the BS method harboured cultivable endophytes in a significantly higher per- employed here, which relied on NJ to rapidly assess branch centage of tissue segments (80.2 %) than did seven representa- support. Well-supported phylogenies are key to diagnosing tive, native species studied previously in Duke Forest, NC the taxonomic affinity of unknown endophytes and are neces- (mean S.D. ¼ 36.4 % 27 of tissue segments; Wilcoxon sary for adequately addressing ecological questions for these signed-rank test, P ¼ 0.031; Arnold & Lutzoni 2007). In contrast, sterile isolates. Future studies will benefit from multiple, con- Juniperus fell within the range of other native taxa in that current analysis methods, especially with regard to linking forest (32.3 %; Wilcoxon signed-rank test, P ¼ 0.578). traditional morphological species concepts to extensive mo- Elevated frequencies of endophyte infection in Platycladus lecular datasets. relative to closely related, co-occurring plants could reflect The occurrence of clades composed only of endophytes three non-exclusive factors: (1) the presence of both local limits our ability to diagnose species boundaries using the and introduced endophytes within these non-native trees; phylogenetic criteria proposed in other studies (e.g. Arnold (2) host-specific rates of infection that are independent of et al. 2007). However, well-supported, endophyte-containing the native/non-native status of these plants; and/or (3) the clades provide a tool for assessing the frequency with which presence of more cosmopolitan/less host-specific endo- particular fungal lineages, rather than genotypes, are repre- phytes, which might be more readily cultivated than those sented among different host species or sites. The prevalence with greater host specificity. of endophyte-only clades suggests that cupressaceous hosts Further sampling is needed to address the first two scenar- harbour interesting and perhaps novel taxa of Ascomycota, ios, such that firm conclusions on this topic cannot be drawn but such statements can only be made with caution pending from this study. However, support for the third scenario comes the population of public databases with additional sequences from previous studies indicating that opportunistic, host- for known but unsequenced fungi, and ongoing morphologi- generalist endophytes are especially common in plants out- cal delimitation of the isolates recovered here. Such clades side their native ranges (e.g. Fisher et al. 1994). Under this also are consistent with the diversification of fungi through scenario, endophytes of Platycladus should be characterized the endophyte symbiosis. Intensive sampling of multiple by a greater proportion of host-generalists than Juniperus and host species within a given plant family represents an impor- Cupressus. Even though all species harboured putatively oppor- tant but under-explored approach for investigating potential tunistic, widespread taxa (e.g. Phoma sp., Aureobasidium sp., co-cladogenesis of endophytes and their plant hosts. Alternaria sp., Cladosporium sp., Xylaria sp.), these cosmopolitan Although the designation of ITS rDNA genotype groups genera were nearly five times more common in Platycladus based on percent similarity is a straightforward and rapid than either native host (Supplementary Material Appendix 1). method for designating species boundaries, there is no We also found that these cosmopolitan genera were three threshold value of sequence similarity that is universally use- times more common in AZ than in NC. We hypothesize that ful for distinguishing species of fungi. Percent divergence be- desert endophytes may be under particularly strong selection tween species is likely to range widely among and within to act as host-generalists: the benefit of opportunistically major lineages, different biological species of fungi frequently infecting hosts is likely high relative to the cost of prolonged have identical ITS rDNA sequences (Lieckfeldt & Seifert 2000), exposure to intense heat, UV radiation, and desiccation typical and cryptic species may sometimes only be revealed through of this region. High rates of host generalism among desert en- a phylogenetic approach based on other loci (Taylor et al. dophytes are further supported by similarity indices, which 2000). We found that our conclusions regarding the relative showed greater similarity in endophyte communities for dif- diversity of endophytes associated with Cupressus versus Platy- ferent host species in AZ than in NC (Table 2). The most com- cladus were reversed when different ITS rDNA genotype simi- mon endophytes recovered from our AZ site showed affinity larities were used to designate species boundaries (Table 1). to Aureobasidium sp., Phoma sp., and Alternaria sp. (Supplemen- This observation, coupled with clade-specific differences in tary Material Appendix 1), consistent with a recent survey of concordance between ITS rDNA genotypes and phylotypes non-host-specific cactus endophytes in AZ (Suryanarayanan (Figs 2–5), warrants continued conservatism in using se- et al. 2005). Thus our data suggest that relative host affinity of quence similarity as a proxy for species boundaries. endophyte communities has the potential to change over the With the exception of native fescue grass in AZ, the com- geographic range of host plant lineages, and may differ among munity structure, ecological roles, and evolution of fungal en- host plant taxa as well. dophytes have not been well defined in desert environments, Arnold et al. (2007) investigated the use of a 600 bp region of particularly with molecular techniques (Faeth & Sullivan LSU rDNA, paired with a phylogenetic backbone constraint, as 2003; but also see Suryanarayanan et al. 2005). How such Endophytes of Cupressaceae 38 343

endophytes might benefit or negatively influence their host Arnold AE, Maynard Z, Gilbert GS, Coley PD, Kursar TA, 2000. Are plant, particularly in times of severe stress, has not been de- tropical fungal endophytes hyperdiverse? Ecology Letters 3: termined. Similarly, the biotically rich forests of the mesic 267–274. Bayman P, Angulo-Sandoval P, Baéz-Ortiz Z, Lodge DJ, 1998. Dis- southeastern USA remain mostly unexplored in terms of their tribution and dispersal of Xylaria endophytes in two tree endophytic symbionts, and may harbour a diversity of endo- species in Puerto Rico. Mycological Research 102: 944–948. phytes similar in scale to that in tropical forests (Arnold & Bills GF, Polishook JD, 1992. Recovery of endophytic fungi from Lutzoni 2007). The relative costs and benefits of endophytic Chamaecyparis thyoides. Sydowia 44: 1–12. symbioses for particular plant species remain largely unre- Carroll GC, Carroll FE, 1978. Studies on the incidence of coniferous solved, and are particularly interesting in an evolutionary con- needle endophytes in the Pacific Northwest. Canadian Journal of text given the ways in which endophyte communities differ in Botany 56: 3034–3043. Cannon PF, Simmons CM, 2002. Diversity and host preference abundance, diversity, composition, and relative host specific- of leaf endophytic fungi in the Iwokrama Forest Reserve, ity over the ranges of their host plants. Guyana. Mycologia 94: 210–220. Cannone JJ, Subramanian S, Schnare MN, Collet JR, D’Souza LM, Du Y, Feng B, Lin N, Madabusi LV, Muller KM, Pande N, Shang Z, Yu N, Gutell RR, 2002. The Comparative RNA Web Acknowledgements (CRW) Site: an online database of comparative sequences and structure information for ribosomal, intron, and other RNAs. BMC Bioinformatics We thank the University of Arizona’s Division of Plant 3: 15. Ewing B, Green P, 1998. Base-calling of automated sequencer Pathology and Microbiology and College of Agriculture and traces using Phred. II. Error probabilities. Genome Research 8: Life Sciences for supporting this work. Additional support 186–194. from the National Science Foundation [DEB-0343953 to Ewing B, Hillier L, Wendl M, Green P, 1998. Base–calling of auto- A.E.A. (and James W. Dalling); DEB-0200413 to A.E.A.] is grate- mated sequencer traces using Phred. I. Accuracy assessment. fully acknowledged. We thank David Maddison for sharing Genome Research 8: 175–185. pre-release versions of MacClade and Mesquite, and for ad- Faeth SH, Sullivan TJ, 2003. Mutualistic asexual endophytes in a native grass are usually parasitic. American Naturalist 161: vice with regard to phylogenetic analyses. We are grateful to 310–325. Rebecca Porter, Mary Shimabukuro, Eric Janson, and Lindsay Fisher PJ, Petrini O, Sutton BC, 1993. A comparative study of Higgins for technical assistance, and to Cara Gibson, Mali fungal endophytes in xylem and bark of Eucalyptus nitens in Gunatilaka, Ming-Min Lee, and especially François Lutzoni for Australia and England. Sydowia 45: 1–14. helpful discussion. Fisher PJ, Petrini O, Petrini LE, Sutton BC, 1994. Fungal endophytes from the leaves and twigs of Quercus ilex L. from England, Majorca, and Switzerland. New Phytologist 127: 133–137. Fisher PJ, Graf F, Petrini LE, Sutton BC, Wookey PA, 1995. Fungal endophytes of Dryas octopetala from a high arctic polar semi- Supplementary data desert and from the Swiss Alps. Mycologia 87: 319–323. Fisher RA, Corbet AS, Williams CB, 1943. The relation between the number of species and the number of individuals in a random Supplementary data associated with this article can be found sample of an animal population. Journal of Animal Ecology 12: in the online version, at doi:10.1016/j.mycres.2007.10.014. 42–58. Fralish JS, Franklin SB, 2002. Taxonomy and Ecology of Woody Plants references in North American Forests. Wiley & Sons, New York. Fro¨ hlich J, Hyde KD, 1999. Biodiversity of palm fungi in the tropics: are global fungal diversity estimates realistic? Biodiversity and Conservation 8: 977–1004. Arnold AE, Lutzoni F, 2007. Diversity and host range of foliar Gamboa MA, Bayman P, 2001. Communities of endophytic fungi fungal endophytes: are tropical leaves biodiversity hotspots? in leaves of a tropical timber tree (Guarea guidonia: Meliaceae). Ecology 88: 541–549. Biotropica 33: 352–360. Arnold AE, Henk DA, Eells R, Lutzoni F, Vilgalys R, 2007. Diversity Gardes M, Bruns TD, 1993. ITS primers with enhanced specificity and phylogenic affinities of foliar fungal endophytes in lob- of basidiomycetes: application to the identification of mycor- lolly pine inferred by culturing and environmental PCR. rhizae and rusts. Molecular Ecology 2: 113–118. Mycologia 99: 185–206. Hata K, Futai K, 1996. Variation in fungal endophyte populations Arnold AE, Miadlikowska J, Higgins KL, Sarvate SD, Gugger P, Way in needles of the genus Pinus. Canadian Journal of Botany 74: A, Hofstetter V, Kauff F, Lutzoni F, in review. Leaves and 103–114. lichens are cradles of fungal diversification. Systematic Biology. Higgins KL, Arnold AE, Miadlikowska J, Sarvate SD, Lutzoni F, Arnold AE, Herre EA, 2003. Canopy cover and leaf age affect 2007. Host affinity and geographic structure among boreal and colonization by tropical fungal endophytes: ecological pat- arctic endophytes from three major plant lineages. Molecular tern and process in Theobroma cacao (Malvaceae). Mycologia 95: Phylogenetics and Evolution 42: 543–555. 388–398. Huelsenbeck JP, Ronquist F, 2001. MRBAYES: Bayesian inference Arnold AE, MejıáL, Kyllo D, Rojas E, Maynard Z, Robbins NA, of phylogenetic trees. Bioinformatics 17: 754–755. Herre EA, 2003. Fungal endophytes limit pathogen damage in James TY, Kauff F, Schoch C, Matheny PB, Hofstetter V, Cox CJ, a tropical tree. Proceedings of the National Academy of Sciences, Celio G, Guiedan C, Fraker E, Miadlikowska J, Lumbsch T, USA 100: 15649–15654. Rauhut A, Reeb V, Arnold AE, Amtoft A, Stajich JE, Hosaka K, Arnold AE, Maynard Z, Gilbert GS, 2001. Fungal endophytes in Sung G, Johnson D, O’Rourke B, Binder M, Curtis JM, Slot C, dicotyledonous neotropical trees: patterns of abundance and Wang Z, Wilson AW, Schu¨ ßler A, Longcore JE, O’Donnell K, diversity. Mycological Research 105: 1502–1507. Mozley-Standridge S, Porter D, Letcher PM, Powell MJ, Taylor JW, 344 M. T. Hoffman, A. E.39 Arnold

White MM, Griffith GW, Davies DR, Sugiyama J, Rossman AY, closely related species of the lichen fungus Peltigera. European Rogers JD, Pfister DH, Hewitt D, Hansen K, Hambleton S, Journal of Phycology 40: 363–378. Shoemaker RA, Kohlmeyer J, Volkmann-Kohlmeyer B, Spotts RA, Petrini O, Stone J, Carroll FE, 1982. Endophytic fungi in evergreen Serdani M, Crous PW, Hughes KW, Matsuura K, Langer E, shrubs in western Oregon – a preliminary study. Canadian Langer G, Untereiner WA, Lu¨ cking R, Bu¨ del B, Geiser DM, Journal of Botany 60: 789–796. Aptroot A, Buck WR, Cole MS, Diederich P, Printzen C, Schmitt I, Petrini O, 1996. Ecological and physiological aspects of host- Schultz M, Yahr R, Zavarzin AI, Hibbett DM, Lutzoni F, specificity in endophytic fungi. In: Redlin SC, Carris LM (eds), McLaughlin DJ, Spatafora JW, Vilgalys R, 2006. Reconstructing the Endophytic Fungi in Grasses and Woody Plants. APS Press, St Paul, early evolution of the Fungi using a six-gene phylogeny. Nature MN, pp. 87–100. 443: 818–822. Posada D, Crandall KA, 1998. MODELTEST: testing the model of Lacap DC, Hyde KD, Liew ECY, 2003. An evaluation of the fungal DNA substitution. Bioinformatics 14: 817–818. ‘morphotype’ concept based on ribosomal DNA sequences. Schulz B, Boyle C, 2005. The endophytic continuum. Mycological Fungal Diversity 12: 53–66. Research 109: 661–686. Lieckfeldt E, Seifert KA, 2000. An evaluation of the use of ITS se- Sikes DS, Lewis PO, 2001. Beta Software, Version 1. PAUPRat: PAUP quences in the taxonomy of the Hypocreales. Studies in Mycology implementation of the parsimony ratchet Distributed by the 45: 35–44. authors. Department of Ecology and Evolutionary Biology, Little jr EL, 1980. The Audubon Society Field Guide to North American University of Connecticut, Storrs. Trees. Alfred A. Knopf, New York. Stone JK, Bacon CW, White jr JF, 2000. An overview of endophytic Little jr EL, 1950. Southwestern Trees: a guide to the native species of microbes: endophytism defined. In: Bacon CW, White jr JF New Mexico and Arizona [U.S Department of Agriculture (eds), Microbial Endophytes. Marcel Dekker, New York, pp. 3–29. Handbook no 9.] United States Government Printing Office, Suryanarayanan TS, Senthilarasu G, Muruganandam V, 2000. Washington, DC. Endophytic fungi from Cuscuta reflexa and its host plants. Lutzoni F, Kauff F, Cox CJ, McLaughlin D, Celio G, Dentinger B, Fungal Diversity 4: 117–123. Padamsee M, Hibbett D, James TY, Baloch E, Grube M, Reeb V, Suryanarayanan TS, Ravishankar JP, Venkatesan G, Murali TS, Hofstetter V, Schoch C, Arnold AE, Miadlikowska JM, 2004. Characterization of the melanin pigment of a cosmo- Spatafora J, Johnson D, Hambleton S, Crockett M, politan fungal endophyte. Mycological Research 108: 974–978. Shoemaker R, Sung GH, Lu¨ cking R, Lumbsch T, O’Donnell K, Suryanarayanan TS, Wittlinger SK, Faeth SH, 2005. Endophytic Binder M, Diederich P, Ertz D, Gueidan C, Hansen K, Harris RK, fungi associated with cacti in Arizona. Mycological Research Hosaka K, Lim YW, Matheny B, Nishida H, Pfister D, Rogers J, 109: 635–639. Rossman A, Schmitt I, Sipman H, Stone J, Sugiyama J, Yahr R, Swofford DL, 2002. PAUP*: phylogenetic analysis using parsimony (*and Vilgalys R, 2004. Assembling the fungal tree of life: progress, other methods). Version 4b10. Sinauer Associates, Sunderland, MA. classification, and evolution of subcellular traits. American Taylor JW, Jacobson DJ, Kroken S, Kasuga T, Geiser DM, Hibbett DS, Journal of Botany 91: 1446–1480. Fisher MC, 2000. Phylogenetic species recognition and species Maddison DR, Maddison WP, 2005. MacClade: analysis of phylogeny concepts in fungi. Fungal Genetics and Biology 31: 21–32. and character evolution. Version 4.08. Sinauer Associates, USDA, NRCS, 2004. The PLANTS Database. Version 3.5. National Sunderland, MA. Plant Data Center, Baton Rouge, LA. http://plants.usda.gov. Mohali S, Burgess TI, Wingfield MJ, 2005. Diversity and host as- Vilgalys R, Hester M, 1990. Rapid genetic identification and map- sociation of the tropical tree endophyte Lasiodiplodia theobro- ping of enzymatically amplified ribosomal DNA from several mae revealed using simple sequence repeat markers. Forest Cryptococcus species. Journal of Bacteriology 172: 4238–4246. Pathology 35: 385–396. White TJ, Bruns T, Lee S, Taylor J, 1990. Amplification and direct Nixon KC, 1999. The parsimony ratchet, a new method for rapid sequencing of fungal ribosomal RNA genes for phylogenetics. parsimony analysis. Cladistics 15: 407–414. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ (eds), PCR O’Brien HE, Miadlikowska J, Lutzoni F, 2005. Assessing host spe- Protocols: a guide to methods and applications. Academic Press, cialization in symbiotic cyanobacteria associated with four San Diego, pp. 315–322. 40

Appendix 1. Endophyte isolates (identification numbers, host taxa, and origin), ITS rDNA genotype groups based on 90%, 95%, and 99% sequence similarity, and top BLAST matches based on analyses conducted in August-

October, 2005. Isolates used for phylogenetic analyses of LSU rDNA data are marked with asterisks. Four isolates were included in LSUrDNA analyses but are not listed below because ITS rDNA sequences were not obtained

(9079, 9084b, 9096, and 9098). Identification numbers correspond to vouchers maintained as living cultures at the

University of Arizona.

ITS rDNA ITS rDNA ITS rDNA Isolate Host Location genotype genotype genotype Blast top match (ITS):GenBank accession number, ID, and e-value (90%) (95%) (99%)

*9002 Platycladus AZ F G 3L gi|30026473|gb|AY225167.1| Aureobasidium pullulans isolate LB... 1053 0.0 *9005 Cupressus AZ A 2Q R gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 993 0.0 *9007 Platycladus AZ A A J gi|59783265|gb|AY831558.1| Leptosphaerulina trifolii strain W... 805 0.0 *9008 Cupressus AZ A A A gi|33302361|gb|AY337712.1| Phoma herbarum 18S ribosomal RNA g... 862 0.0 *9009a Platycladus AZ B B B gi|32394890|gb|AY154712.1| Alternaria tenuissima strain IA287... 1063 0.0 *9009b Platycladus AZ F G 3L gi|30026473|gb|AY225167.1| Aureobasidium pullulans isolate LB... 1047 0.0 *9015 Platycladus AZ F G S gi|37786120|gb|AY213639.1| Aureobasidium pullulans strain UWF... 991 0.0 *9018 Platycladus AZ A A A gi|33302361|gb|AY337712.1| Phoma herbarum 18S ribosomal RNA g... 860 0.0 *9021 Platycladus AZ B B B gi|32394890|gb|AY154712.1| Alternaria tenuissima strain IA287... 1082 0.0 *9026 Platycladus AZ B B B gi|32394890|gb|AY154712.1| Alternaria tenuissima strain IA287... 1112 0.0 *9027 Platycladus AZ A A J gi|59783265|gb|AY831558.1| Leptosphaerulina trifolii strain W... 837 0.0 *9028 Platycladus AZ F G K gi|30026472|gb|AY225166.1| Aureobasidium pullulans isolate SK... 993 0.0 *9030 Platycladus AZ A A J gi|59783265|gb|AY831558.1| Leptosphaerulina trifolii strain W... 839 0.0 *9031 Platycladus AZ F G K gi|30026472|gb|AY225166.1| Aureobasidium pullulans isolate SK... 1017 0.0 *9036 Platycladus AZ F G K gi|30026472|gb|AY225166.1| Aureobasidium pullulans isolate SK... 918 0.0 41

*9038 Platycladus AZ P P T gi|30089120|emb|AJ458185.1|CNI458185 Chaetomium nigricolor 5.... 714 0.0 *9042 Platycladus AZ F G K gi|30026472|gb|AY225166.1| Aureobasidium pullulans isolate SK... 993 0.0 *9051 Platycladus AZ M M U gi|41058714|gb|AY465446.1| Dothideales sp. GS2N1c 18S ribosom... 646 0.0 *9054 Platycladus AZ A 2Q M gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 950 0.0 *9055 Platycladus AZ B B B gi|32394890|gb|AY154712.1| Alternaria tenuissima strain IA287... 1114 0.0 *9057 Cupressus AZ I H L gi|47420271|gb|AY546017.1| Fungal endophyte WMS23 internal tr... 813 0.0 *9058 Cupressus AZ I H L gi|47420271|gb|AY546017.1| Fungal endophyte WMS23 internal tr... 813 0.0 *9059 Cupressus AZ I H L gi|47420271|gb|AY546017.1| Fungal endophyte WMS23 internal tr... 813 0.0 *9060 Platycladus AZ F G V gi|30026472|gb|AY225166.1| Aureobasidium pullulans isolate SK... 995 0.0 9064 Platycladus AZ A 2Q 3K gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 997 0.0 *9065 Platycladus AZ A 2Q M gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 993 0.0 *9069 Platycladus AZ H F H gi|29165262|gb|AY219880.1| Lecythophora sp. UBCtra1453C 18S r... 868 0.0 *9084a Platycladus AZ Q Q W gi|25045988|gb|AF491556.1| Peziza varia KH-97-88 (C) 18S ribo... 1183 0.0 *9085 Platycladus AZ H R X gi|28202120|gb|AY141180.1| Aureobasidium pullulans 18S riboso... 821 0.0 *9089 Platycladus AZ R S Y gi|22773864|gb|AF405301.1| Bartalinia robillardoides 18S ribo... 854 0.0 *9092 Platycladus AZ S T Z gi|71725083|dbj|AB231012.1| Lecythophora hoffmannii genes for... 880 0.0 *9093 Platycladus AZ H F H gi|29165262|gb|AY219880.1| Lecythophora sp. UBCtra1453C 18S r... 868 0.0 *9094 Platycladus AZ H F 2A gi|29165262|gb|AY219880.1| Lecythophora sp. UBCtra1453C 18S r... 872 0.0 *9097 Cupressus AZ T U 2B gi|12583572|emb|AJ271575.1|TSU271575 Thielavia subthermophila... 1053 0.0 *9104 Cupressus AZ M N 2C gi|46243857|gb|AY510419.1| Preussia similis strain s19 intern... 942 0.0 *9106 Cupressus AZ M N 2D gi|41058713|gb|AY465445.1| Dothideales sp. GS5N1b 18S ribosom... 890 0.0 *9107 Cupressus AZ A 2Q 3K gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 971 0.0 9115 Cupressus AZ I V 2E gi|47420271|gb|AY546017.1| Fungal endophyte WMS23 internal tr... 722 0.0 *9116 Cupressus AZ M W 2F emb|AJ972795.1| Monodictys sp. MA 4647 18S rRNA gene, 5.8S rR... 751 0.0 *9120 Cupressus AZ M M 2G gb|AY465446.1| Dothideales sp. GS2N1c 18S ribosomal RNA gene,... 607 1e-170 *9121 Platycladus NC J J 2H gb|AF377289.1| Pestalotiopsis funereoides 18S ribosomal RNA g... 1035 0.0 9122 Platycladus NC J J O gb|AF377289.1| Pestalotiopsis funereoides 18S ribosomal RNA g... 1039 0.0 *9126 Platycladus NC J J O gb|AF377289.1| Pestalotiopsis funereoides 18S ribosomal RNA g... 1037 0.0 *9128 Platycladus NC U X 2I gb|AF393724.2| Cladosporium tenuissimum strain ATCC 38027 18S... 961 0.0 *9129 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1154 0.0 gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... *9130 Platycladus NC C C D 1160 0.0 9133 Platycladus NC D D E gb|AY786322.1| Botryosphaeria dothidea strain CBS 116743 18S ... 1041 0.0 42

*9134 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1158 0.0 *9135 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1154 0.0 9136 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1116 0.0 *9137 Platycladus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1063 0.0 9138 Platycladus NC V Y 2J gb|DQ094165.1| Fungal sp. YX-41 internal transcribed spacer 1... 999 0.0 *9140 Platycladus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1072 0.0 9142 Platycladus NC C C C gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1154 0.0 *9143 Platycladus NC J J 2K gb|AF377289.1| Pestalotiopsis funereoides 18S ribosomal RNA g... 1037 0.0 *9144 Platycladus NC C C D gi|12082793|gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ... 1217 0.0 *9145 Platycladus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1076 0.0 9146 Platycladus NC E E I gb|DQ094168.1| Fungal sp. YX-47 internal transcribed spacer 1... 1041 0.0 *9147 Platycladus NC W Z 2L gb|AF125942.1| Penicillium sp. NRRL 28148 internal transcribe... 1033 0.0 9148 Platycladus NC E E I gb|DQ094168.1| Fungal sp. YX-47 internal transcribed spacer 1... 1017 0.0 *9149a Platycladus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1055 0.0 *9149b Platycladus NC C I N gi|12082793|gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ... 908 0.0 9150 Platycladus NC C C G gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1122 0.0 9151 Platycladus NC C C G gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1120 0.0 9153 Platycladus NC X 2A 2M gb|AY063310.1| Fungal endophyte WMS14 internal transcribed sp... 648 0.0 *9154 Platycladus NC K K 2N gb|AY833034.1| Uncultured mycorrhizal ascomycete isolate 1 in... 387 2e-104 9157 Platycladus NC A A 2O gb|AY337712.1| Phoma herbarum 18S ribosomal RNA gene, partial... 825 0.0 *9158 Platycladus NC A A J gb|AY131201.1| Ascochyta lentis isolate MU AL1 18S ribosomal ... 811 0.0 9160 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1164 0.0 9162 Platycladus NC C C G gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1120 0.0 9163 Platycladus NC C I 2P gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 771 0.0 9165 Platycladus NC Y 2B 2Q gb|AY265334.1| Scolecobasidium humicola strain B3343 internal... 315 8e-83 9167 Platycladus NC E E I gb|DQ094168.1| Fungal sp. YX-47 internal transcribed spacer 1... 1033 0.0 *9168 Platycladus NC K K 2R gb|AY833034.1| Uncultured mycorrhizal ascomycete isolate 1 in... 387 2e-104 9169 Platycladus NC K K 2S dbj|AB067720.1| Cordyceps sinensis genes for 18S rRNA, ITS1, ... 464 1e-127 9171 Platycladus NC Z 2C 2T emb|AJ390436.1|HAN390436 Nemania serpens 18S rRNA gene (parti... 969 0.0 9172 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1158 0.0 9173 Platycladus NC C C D gi|12082793|gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ... 1078 0.0 9174 Cupressus AZ F G 2U gb|AY225166.1| Aureobasidium pullulans isolate SK3 18S riboso... 934 0. 43

9176 Platycladus NC C C C gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1156 0.0 9177 Platycladus NC C C C gi|12082793|gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ... 1245 0.0 9178 Platycladus NC C I N gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 908 0.0 9179 Platycladus NC G O Q gb|AF502622.1| Leaf litter ascomycete strain its031 isolate 1... 820 0.0 9183 Juniperus NC A 2Q 2V gi|28974197|gb|AY183371.1| Phoma glomerata 18S ribosomal RNA ... 1049 0.0 9184 Platycladus NC C C G gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1076 0.0 9187 Juniperus NC O 2D 2W dbj|AB107890.1| Phomopsis sp. MAFF 665006 genes for ITS1, 5.8... 1021 0.0 9188 Juniperus NC O 2E 2X gb|AY577815.1| Diaporthe phaseolorum strain E99382 18S riboso... 892 0.0 9189 Juniperus NC 2A 2F 2Y gb|AY208791.1| Stagonospora sp. Po41 internal transcribed spa... 839 0.0 9190 Juniperus NC D D E gb|AY786322.1| Botryosphaeria dothidea strain CBS 116743 18S ... 987 0.0 9191 Platycladus NC 2B 2G 2Z gb|AY213667.1| Paecilomyces lilacinus strain UWFP 674 18S rib... 969 0.0 9193 Juniperus NC 2C 2H 3A gb|AY063297.1| Fungal endophyte WMS1 internal transcribed spa... 593 2e-166 9194 Platycladus NC G O Q gb|AF502622.1| Leaf litter ascomycete strain its031 isolate 1... 813 0.0 9195 Platycladus NC C C D gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ribosomal RN... 1154 0.0 9198 Juniperus NC 2A 2I 3B gb|AY183366.1| Rhizosphaera kalkhoffii 18S ribosomal RNA gene... 829 0.0 9199 Juniperus NC 2D 2J 3C emb|AJ539482.1|MCR539482 Morchella crassipes 18S rRNA gene (p... 646 0.0 *9200 Juniperus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1067 0.0 *9201 Juniperus NC L L P emb|AJ390411.1|HAN390411 Biscogniauxia atropunctata 18S rRNA ... 1057 0.0 *9202 Juniperus NC L L P emb|AJ390411.1|HAN390411 Biscogniauxia atropunctata 18S rRNA ... 1017 0.0 *9203 Juniperus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1061 0.0 9206 Platycladus NC 2E 2K 3D gb|AY971711.1| Fungal sp. 4.32 18S ribosomal RNA gene, partia... 607 1e-170 9208 Juniperus NC 2F 2L 3E gb|AY315399.1| Xylariaceae sp. C8 internal transcribed spacer... 410 1e-111 9209 Juniperus NC 2G 2M 3F gb|AF409982.1| Pestalotiopsis sp. EN 7 18S ribosomal RNA gene... 442 4e-121 9211 Juniperus NC 2H 2N 3G gb|AY315396.1| Xylaria sp. F23 internal transcribed spacer 1,... 1322 0.0 9217 Juniperus NC E E F gb|AY561200.1| Foliar endophyte of Picea glauca sp. P3 intern... 1037 0.0 9220 Juniperus NC J J 3H gb|AY687871.1| Pestalotia lawsoniae voucher PSH2000I-057 18S ... 1070 0.0 9247 Juniperus NC N 2O 3I gb|AF260224.1|AF260224 Kabatina juniperi 18S ribosomal RNA, p... 1155 0.0 9248 Platycladus NC G O Q gb|AF502622.1| Leaf litter ascomycete strain its031 isolate 1... 816 0.0 9249 Platycladus NC C C C gi|12082793|gb|AF312009.1|AF312009 Phyllosticta spinarum 18S ... 1245 0.0 9283 Platycladus NC 2I 2P 3J dbj|AB041241.1| Guignardia mangiferae genes for 18S rRNA, ITS... 1274 0.0

Appendix 2. GenBank accession numbers for LSU rDNA sequences of endophytic fungi obtained from other studies for phylogenetic analyses.

Endophyte isolate Accession no.

4245 DQ979442 4250 DQ979443 4466 DQ979444 422 EF420091 4653 DQ979449 6722 EF420100 256 EF420088 2204 EF420087 6717 EF420099 6728 EF420101 6702 EF420096 6733 EF420104 412 EF420089 6756 EF420107 6741 EF420105 6749 EF420106 6714 EF420098 151 EF420086 425 EF420092 4939 DQ979453 416 EF420090 6708 EF420097 4780 DQ979452 469 EF420095 2834 DQ979423 2779 DQ979422 3300 DQ979427 1182 EF420085 6730 EF420102 6731 EF420103 435b EF420093 2712 DQ979419 2722 DQ979420 462 EF420094 4140 DQ979440 3358A DQ979433 4221 DQ979441 3326 DQ979429 5007 DQ979455 1029 EF420084 45

4109 DQ979438 4947A DQ979454 4089 DQ979437 3275 DQ979425 3335 DQ979431 3353 DQ979432 3329 DQ979430 3377 DQ979434 5718 DQ979463 5095 DQ979457 3323 DQ979428 5654 DQ979462 2851 DQ979424 2706 DQ979418 5246 DQ979458

Appendix 3. Representative Ascomycota species included in phylogenetic analyses and

GenBank identification numbers. Sequences for Coccidioides posadasii were obtained from

TIGR (http://www.tigr.org).

Taxon GenBank ID

Aleuria aurantia 45775583 Ampelomyces quisqualis 31415592 Aphanoascus fulvescens 33113349 Apiospora sinensis 27447246 Arthrinium phaeospermum 27447247 Arthroderma curreyi 33113367 Ascobolus carbonarius 45775606 Ascobolus crenulatus 45775607 Ascosacculus heteroguttulatus 34453082 Ascosalsum cincinnatulum 34453080 Ascosphaera apis 11120650 Astrocystis cocoes 27447238 Auxarthron zuffianum 33113353 Balansia henningsiana 45155172 Barssia oregonensis 45775581 Bimuria novae-zelandiae 15808399 Botryosphaeria ribis 11120642 Byssochlamys nivea 33113391 Byssothecium circinans 15808400 Cainia graminis 20334356 Camarops microspora 27447236 Candida albicans 671812 Capronia mansonii 11120644 Capronia pilosella 12025061 Cephalotheca sulfurea 20334357 Ceramothyrium carniolicum 11120645 Chaetomium globosum 13469777 Cheilymenia stercorea 45775590 Coccidioides posadasii TIGR222929 Cochliobolus heterostrophus 45775574 Cordyceps irangiensis 13241779 Cordyceps khaoyaiensis 13241765 Cordyceps sinensis 29466997 Corollospora maritima 22203655 Cryphonectria cubensis 22671402 Cryphonectria havanensis 22671403 Cryptodiaporthe corni 22671407 Cryptosphaeria eunomia var. eunomia 27447241 Curvularia brachyspora 12025063 Delphinella strobiligena 15808401 Dendrographa leucophaea f. minor 47499217 Dermatocarpon luridum 52699696 Diaporthe phaseolorum 1399144 Disciotis venosa 45775596 Discosphaerina fagi 15808402 Discula destructiva 22671422 Dothidea insculpta 52699697 Dothidea ribesia 15808403

Dothidea sambuci 45775610 Emericellopsis terricola 1698507 Eremascus albus 11120651 Exophiala jeanselmei 4206347 Fasciatispora petrakii 27447243 Farrowia longicollea 13469782 Farrowia seminuda 13469784 Gaeumannomyces graminis var. graminis 16565889 Glyphium elatum 17104831 Gnomonia setacea 16565895 Gnomoniella fraxini 16565884 Gyromitra esculenta 52699700 Gyromitra californica 45775602 Halorosellinia oceanica 27447237 Halosarpheia marina 34453085 Helvella compressa 45775584 Hydropisphaera erubescens 45155165 Hypocrea citrina 45775578 Hypomyces polyporinus 32261014 Hyponectria buxi 27447249 Hypoxylon fragiforme 27447244 Leotia lubrica 45775573 Letendraea helminthicola 15808405 Lojkania enalia 15808406 Lulworthia fucicola 22203665 Lulworthia grandispora 22203666 Magnisphaera stevemossago 34453095 Melanconis marginalis 22671436 Menispora tortuosa 45775611 Microascus trigonosporus 1399149 Microdochium nivale 2565289 Microxyphium citri 11120643 Morchella elata 45775594 Morchella esculenta 12025081 Natantispora lotica 34453084 Neolecta vitellina 12025084 Neurospora crassa 13469785 Otidea onotica 17154899 Oxydothis frondicola 27447250 Paracoccidioides brasiliensis 2271524 Peltula umbilicata 15216700 Penicillium freii 52699706 Peziza proteana 45775588 Peziza quelepidotia 52699707 Peziza succosa 17154944 Phaeosphaeria avenaria 45775613 Phoma glomerata 31415593 Plagiostoma euphorbiae 22671445 Pleurothecium recurvatum 45775614 Preussia terricola 45775615 Pyrenophora tritici-repentis 45775601 Pyrenula cruenta 12025090 Pyrenula pseudobufonia 52699710 Raciborskiomyces longisetosum 15808410 Rosellinia necatrix 27447239 Saccharomyces cerevisiae 172409 Sagaaromyces abonnis 34453078 Sarcoscypha coccinea 45775576 Sarcosphaera crassa 45775597 Schizosaccharomyces pombe 288696 Setomelanomma holmii 28144315 Setosphaeria monoceras 15808411

Seynesia erumpens 12025093 Sordaria fimicola 45155203 Sordaria macrospora 37992293 Stylodothis puccinioides 11120648 Sydowia polyspora 45775604 Taphrina communis 52699720 Thyridium vestitum 45775600 Torrubiella luteorostrata 13241778 Trematosphaeria heterospora 15808412 Varicosporina ramulosa 22203671 Verpa conica 45775595 Verrucaria pachyderma 15216679 Verticillium epiphytum 8777749 Valsa ambiens 16565896 Westerdykella cylindrica 11120649 Xylaria acuta 45775605 Xylaria hypoxylon 45775577

Appendix 4. Model parameters for Bayesian analyses for each focal class of Ascomycota.

Class Selected model - Ln likelihood Proportion of invariant Gamma shape sites parameter Dothideomycetes GTR+I+G 8463.17251 0.420231 0.526946 Sordariomycetes GTR+I+G 5924.5000 0.565827 0.4267 Eurotiomycetes GTR+I+G 3613.9267 0.634904 0.55459 Pezizomycetes GTR+I+G 4519.40130 0.522561 0.639625

Estimated Dothideomycetes base frequencies = A:0.266879 C:0.217424 G:0.307280 T:0.208417 Estimated Pezizomycetes base frequencies = A:0.287876 C:0.185319 G:0.288132 T:0.238673 Estimated Eurotiomecetes base frequencies = A:0.273327 C:0.208930 G:0.305662 T:0.212081 Estimated Sordariomycetes base frequencies = A:0.276040 C:0.210782 G:0.300050 T:0.213129

APPENDIX B

DIVERSE BACTERIAL ENDOSYMBIONTS INHABIT LIVING HYPHAE OF

PHYLOGENETICALLY DIVERSE FUNGAL ENDOPHYTES 50 51

Diverse bacteria inhabit living hyphae of phylogenetically diverse fungal endophytes

Michele T. Hoffman and A. Elizabeth Arnold

Division of Plant Pathology and Microbiology

School of Plant Sciences

1140 E. South Campus Drive

University of Arizona

Tucson, AZ 85721 USA

For correspondence: AE Arnold, Tel: 520.621.7212; Fax: 520.621.9290; Email:

[email protected]

Running head: Endohyphal bacteria of endophytic fungi

Contains 3 tables, 5 figures, 4 appendices 52

Abstract

Both the establishment and outcomes of plant-fungal symbioses can be influenced by abiotic factors, the interplay of fungal and plant genotypes, and additional microbes associated with fungal mycelia. Recently, bacterial endosymbionts have been documented in soilborne Glomeromycota and Mucoromycotina, and in at least one species each of mycorrhizal Basidiomycota and Ascomycota. Here, we show for the first time that phylogenetically diverse endohyphal bacteria occur in living hyphae of diverse foliar endophytes, including representatives of four classes of Ascomycota. We examined 414 isolates of endophytic fungi, isolated from photosynthetic tissues of six species of cupressaceous trees in five biogeographic provinces, for endohyphal bacteria using microscopy and molecular techniques. Viable bacteria were observed within living hyphae of endophytic Pezizomycetes, Dothideomycetes, Eurotiomycetes, and

Sordariomycetes from all tree species and biotic regions surveyed. A focus on 29 fungal/bacterial associations revealed that bacterial and fungal phylogenies were incongruent with each other and with taxonomic relationships of host plants. Overall, eight families and 15 distinct genotypes of endohyphal bacteria were recovered; most were Proteobacteria, but a small number of Bacillaceae also were found, including one that appears to occur as an endophyte of plants. Frequent loss of bacteria following subculturing suggests a facultative association. Our study recovered distinct lineages of endohyphal bacteria relative to previous studies, is the first to document their occurrence in foliar endophytes representing four of the most species-rich classes of fungi, and highlights for the first time their diversity and phylogenetic relationships 53

with regard both to the endophytes they inhabit and the plants in which these endophyte-bacterial symbiota occur.

Keywords: Ascomycota, endohyphal bacteria, Bayesian analysis, Burkholderiales,

Cupressaceae, diversity, fungal endophytes, phylogeny, FISH 54

Introduction

Traits related to the establishment and outcome of plant-fungal symbioses can reflect not only abiotic conditions and the unique interactions of particular fungal and plant genotypes (50, 51, 57, 60, 63, 68), but also additional microbes that interact intimately with fungal mycelia (4, 12, 43). For example, mycorrhizosphere-associated actinomycetes release volatile compounds that influence spore germination in the arbuscular mycorrhizal (AM) fungus Gigaspora margarita (Glomeromycota) (14).

Levy et al. (35) describe Burkholderia spp. that colonize spores and hyphae of the AM fungus Gigaspora decipiens and are associated with decreased spore germination.

Diverse "helper" bacteria have been implicated in promoting hyphal growth and the establishment of ectomycorrhizal symbioses (23, 26, 58, 71). Minerdi et al. (44) found that a consortium of ectosymbiotic bacteria limited the ability of the pathogen Fusarium oxysporum to infect and cause vascular wilts in lettuce, with virulence restored to the pathogen when ectosymbionts were removed.

In addition to interacting with environmental and ectosymbiotic bacteria, some plant-associated fungi harbor bacteria within their hyphae (first noted as 'bacteria-like organisms' of unknown function) (39). These bacteria, best known from living hyphae of several species of Glomeromycota and Mucoromycotina, can alter fungal interactions with host plants in diverse ways (see 12, 32, 52). For example, the vertically transmitted bacterium Candidatus Glomeribacter gigasporarum colonizes spores and hyphae of the arbuscular mycorrhizal (AM) fungus Gigaspora gigasporarum (9, 10). Removal of the bacterial partner from the fungal spores suppresses fungal growth and development, 55

altering the morphology of the fungal cell wall, vacuoles, and lipid bodies. (38). In turn, the discovery of phosphate-solubilizing bacteria within Glomus mossae spores (45), coupled with the recovery of a P-transporter operon in Burkholderia sp. from Gigaspora margarita (55), suggest a competitive role in phosphate acquisition and transport by these bacteria within the AM symbiosis. Within the Mucoromycotina, Partida-Martinez and

Hertweck (52) reported that a soilborne plant pathogen, Rhizopus microsporus, harbors endosymbiotic Burkholderia that produce a phytotoxin (rhizoxin) responsible for the pathogenicity of the fungus.

These examples, coupled with the discovery of bacteria within hyphae of ectomycorrhizal Dikarya (Tuber borchii; Ascomycota; Laccaria bicolor and

Piriformospora indica; Basidiomycota) (5, 6, 7, 59), suggest that the capacity to harbor endohyphal bacteria is widespread among fungi. To date, however, endocellular bacteria have been documented only from fungi that occur in the soil and rhizosphere

(12, 32). Here we report for the first time that phylogenetically diverse bacteria occur within living hyphae of foliar endophytic fungi, including members of four classes of filamentous Ascomycota. We use a combination of light and fluorescent microscopy to visualize bacterial infections within living hyphae of representative strains. Then, drawing from surveys of endophytes from asymptomatic foliage of cupressaceous trees in five biogeographic provinces, we provide a first characterization of the phylogenetic relationships, host associations, and geographic distributions of endohyphal bacteria associated with focal fungal endophytes.

Methods

Conifers (Pinophyta or Coniferae), comprising ca. 630 species, form the dominant tree component in major biomes such as boreal forests, temperate tree savannas, and temperate rainforests (20, 21, 22). The most widely distributed family of conifers, the

Cupressaceae (cypresses and relatives), includes 30 genera and ca. 142 species in seven proposed subfamilies (25). Eleven genera are placed in the subfamily Cupressoideae, including three examined in this study (Cupressus, Juniperus, and Platycladus).

We collected foliar endophytes from six species of Cupressoideae, as available, in five localities (Table 1): semi-deciduous forest in the piedmont of North Carolina,

USA (NC), and four distinct regions of Arizona, USA (AZ) selected on the basis of distinctive biogeographic histories and the occurrence of desired plant species. These sites included the Chuska Mountains (CHU) of northeastern AZ, with petran montane coniferous forest featuring significant elements of the Rocky Mountain flora; the

Mogollon Rim (MOG) and central mountains, including Mingus Mountain and the

Bradshaw Mountains, which lie at the interface of the Great Basin coniferous woodlands, interior chaparral, and petran montane forest; the Sky Island archipelago

(SKY) of southeastern AZ, including the Santa Catalina and Chiricahua Mountains, characterized by madrean evergreen forest; and the Campus Arboretum at the

University of Arizona (UA), a semi-urban, cultivated setting within the Arizona upland province of the Sonoran Desert.

Endophyte collection

Fresh, asymptomatic photosynthetic tissue was collected from at least three healthy, mature individuals of each focal species in each locality during the growing seasons of

2004-2007. From each tree, healthy, mature photosynthetic tissue was collected from three haphazardly selected branches at the outer canopy ca. 1-2 m above ground.

Material was transferred to the laboratory for processing within 6-12 hours of collection.

Tissue samples were washed in running tap water and then cut into 2mm segments, corresponding to infection domains for individual fungi (3). Segments were surface-sterilized by rinsing in 95% ethanol for 30 seconds, 10% Clorox® (0.6% sodium hypochlorite) for two minutes, and 70% ethanol for two minutes (1), allowed to surface- dry under sterile conditions, and plated on 2% malt extract agar (MEA), which encourages growth by a diversity of endophytes (24). In sum, 384 tissue segments (96 per individual) were plated per species at each location, with the exceptions of

Juniperus deppeana in the Chiricahua Mountains (144 segments), and Platycladus orientalis and Juniperus virginiana from North Carolina (96 segments/species).

Plates were incubated at room temperature and inspected for hyphal growth daily for 18 weeks. Emerging fungi were isolated on 2% MEA, archived as living vouchers in sterile water, and deposited at the Robert L. Gilbertson Mycological

Herbarium at the University of Arizona (ARIZ; accession numbers available from MH).

Seven isolates of bacteria obtained directly from surface-sterilized plant tissue (i.e., 58

bacterial endophytes) were accessioned as sterile cultures and maintained in sterile 80% glycerol at -80°C.

PCR amplification and sequencing

Total genomic DNA was extracted directly from axenic fungal cultures following

Arnold et al. (2). In addition, DNA was extracted a second time from mycelium of five isolates using the QIAGEN® DNeasy® Plant Mini kit to ensure that none of our records of endohyphal bacteria represented contamination of reagents in our standard DNA extraction protocol. Our two extraction methods were consistent in terms of DNA extraction quality, sequence quality, and resulting diagnoses of bacterial infection.

For each fungal isolate, we used PCR to amplify the nuclear ribosomal internal transcribed spacers and 5.8s gene (ITSrDNA), and when possible, the first 600bp of the large subunit (LSUrDNA) as a single fragment (ca. 1000-1200bps in length) using primers ITS1F and ITS4 or LR3 (27, 56, 67, 70). Each 25µl reaction included 12.5 µl of Sigma ReadymixTM REDTaqTM, 1 µl of each primer (10 µM), 9.5 µl of PCR-quality water, and 1 µl of DNA template. Cycling reactions were run with MJ Research

PTC200TM thermocyclers and consisted of 94°C for 3 minutes; 36 cycles of 94°C for 30 seconds, 54°C for 30 seconds, 72°C for 1 minute; and 72°C for 10 minutes.

The presence or absence of bacteria within the surrounding matrix was determined initially using light microscopy. Fungal isolates were examined after one week of growth in pure culture on 2% MEA using a Leica DM4000B with bright field imaging (400X; NA = 0.75). Once visual examination ruled out non-endohyphal 59

bacteria (i.e., contaminants in the medium or microbes on hyphal surfaces), total genomic DNA extracted from fresh mycelia was examined using PCR primers specific to bacterial 16S rDNA: 10F and 1507R (1497 bp) or 27F and 1429R (1402 bp) (34, 46).

PCR reaction mixes and cycling parameters were as described above, except that annealing temperatures were 58°C (10F and 1507R) or 55°C (27F and 1429R).

SYBR® Green I stain (Molecular Probes, Invitrogen) was used to detect DNA bands on 1.5% agarose gels. Positive PCR products were cleaned, quantified, and normalized at the University of Arizona Genomic Analysis and Technology Core facility (GATC), followed by bidirectional sequencing with PCR primers (5 µM) using an Applied Biosystems 3730XL DNA Analyzer. PCR products of insufficient concentration for sequencing were cloned using Agilent Technologies StrataClone™ cloning kits following the manufacturer’s instructions. Water was used in place of template for negative controls. Positive clones were amplified and sequenced using primers M13F and M13R.

The software applications phred and phrap (18, 19) were used to call bases and assemble bidirectional reads into contiguous consensus sequences, with automation provided by ChromaSeq (40) implemented in Mesquite v. 2.6 (41). Base calls were verified by inspection of chromatograms in Sequencher™ v 4.5 (Gene Codes Corp.,

Ann Arbor, MI). ITSrDNA and partial LSUrDNA sequence data have been submitted to

GenBank under accession numbers GQ153054-GQ153264 or were published previously by the authors (28), and bacterial 16S rDNA sequences under accession numbers HM046622-HM046627 and HM117722-117749. 60

Diversity and taxonomic placement of endophytes

Previous studies have used 90-97% ITS rDNA sequence similarity to estimate species boundaries for environmental samples of fungi (97%, O’Brien et al. (49); 90%,

Hoffman & Arnold (28); 95%, Arnold et al. (1). Empirical estimates of percent sequence divergence between sister species for four endophyte-containing ascomycetous genera (Botryosphaeria, Colletotrichum, Mycosphaerella, and Xylaria) indicated that groups based on 95% ITS rDNA sequence similarity conservatively estimate species (66). We used Sequencher™ v 4.5 (Gene Codes Corp., Ann Arbor, MI.; default settings except for the requirement of 20 character minimum overlap; see

Arnold et al. (3)) to estimate fungal OTU at 95% ITSrDNA sequence similarity, and bacterial OTU at 97% 16S sequence similarity (62).

Taxonomic placement for fungi at the level of order and above was estimated by

BLAST comparisons with the curated ITSrDNA database for fungi maintained by the

Alaska Fungal Metagenomics Project (FMP; http://www.borealfungi.uaf.edu/) and phylogenetic analysis (below). Bacterial taxonomy was estimated by BLAST comparisons with GenBank and the Ribosomal Database Project (RDP; release 9) classifier program (Naive Bayesian rRNA Classifier Version 2.0, July 2007) (15, 69), combined with phylogenetic analyses (below). All statistical analyses were performed in JMP 7.0 (SAS Institute Inc. Cary, NC).

Live-Dead stain

Live-Dead stain was used to confirm the presence and viability of endohyphal bacteria for fungal isolates identified as containing bacteria on the basis of light microscopy and

PCR (above). We treated fresh mycelia in sterile water with 2µl of Molecular Probe

Live-Dead fluorescent stain for 15 min and prepared slides with VECTASHIELD®

HardSet™ 1400 medium (Vector Laboratories, Inc.) to prevent photobleaching. A Leica

DM4000B microscope with Luminera camera and 100W mercury arc lamp was used for fluorescent imaging with a Chroma Technology filter set 35002 (480nm excitation/520nm emission) and 1000X APO oil objective (NA=1.40-0.70; 0.13mm working distance) at room temperature. Visible fluorescence of bacterial nucleic acids within living hyphae that was not consistent with fungal mitochondrial or nuclear DNA, coupled with positive PCR results and a lack of extrahyphal or ectosymbiotic bacteria, provided evidence of viable endohyphal bacteria (47, 64).

Fluorescent in situ hybridization (FISH)

To rule out misinterpretation of fluorescence results from Live-Dead analyses we used

FISH with a probe specific for eubacteria to confirm endocellular infection in two focal isolates (9084b and 9143). In each case, we examined hyphae from cultures grown on

2% MEA and on 2% MEA amended with antibiotics, which removed evident bacterial infections (Hoffman and Arnold, in prep.).

Fresh mycelium was harvested and suspended in 1000ul 1X phosphate buffered saline (PBS). After centrifugation at 8000rpm for 5 minutes, PBS buffer was replaced 62

with fresh 4% formaldehyde solution and incubated for 1.5 hours at 4°C. Samples then were centrifuged at 8000rpm for 5 minutes and the pellet resuspended in PBS buffer twice prior to storage in absolute ethanol at -20°C. Fixed mycelium was collected to

0.5 ml tubes and dehydrated in a series of ethanol/PBS solutions beginning with 50%,

70%, and finally 95% ethanol. Dehydrated mycelium was incubated for 90 minutes at

46°C in 10ul of hybridization buffer (HB) using 40% formamide hybridization stringency (800ul formamide, 800ul DEPC water, 500ul EDTA) with 2ul of EUB338 probe (10uM), a universal 16S rDNA oligonucleotide probe (labeled with 5' TAMRA fluorochrome tag; 5' GCTGCCTCCCGTAGGAGT 3'; EX559nm/EM583nm; Integrated

DNA Technologies, Inc.). Each sample was rinsed in 100ul wash buffer (460ul 5M

NaCl, 1ml 1M Tris, 50ul SDS, DEPC water, fill to 50ml) and warmed to 46°C twice.

This method was repeated using [1] antibiotic-cured isolates of each strain, which showed no evidence of endohyphal bacteria; and [2] hybridization buffer and PBS only, but no EUB338 probe, to control for autoflorescence. Mycelium was mounted on gelatin-coated glass slides using DAPI as a counter-stain (42) and examined with either an Olympus BX61 with a mercury arc lamp or a Leica DMI 6000 confocal system equipped with a 543nm laser (63X oil; 1024x1024 format; xyz acquisition; line average

=1; frame average = 4). Leica software LAS-AF v 1.8.2 was used to capture images.

Phylogenetic analyses

Based on positive 16S rDNA PCR results and the lack of any extrahyphal bacteria in our axenic cultures, endohyphal bacteria were observed in 75 of 414 fungal isolates. Twenty- 63

one infected isolates representing a broad diversity of Pezizomycotina, and 49 isolates in which infections were never observed and that represented a broad array of ITS rDNA genotype groups, were sequenced for 1200 bps of LSU rDNA following Arnold et al. (3)

(primers LROR or LR3R and LR7; protocols as described above). Additional LSU rDNA sequence data have been submitted to GenBank under accession numbers HM117750-

HM117756.

These 70 sequences were integrated into core alignments of 157 representative

Ascomycota, partitioned by class into Sordariomycetes, Dothideomycetes, and

Eurotiomycetes (28) using Mesquite v 2.6 (41). Details of taxon sampling are given in

Table 2 and Appendix 1. Because our sample of Pezizomycetes contained only one isolate with strong evidence of endohyphal bacteria, we did not address phylogenetic relationships of pezizomycetous endophytes here.

Alignment for each class, based on LSU rDNA secondary structure defined by

Saccharomyces cerevisiae (13), was performed using ClustalW with manual adjustment in Mesquite. All ambiguously aligned regions were excluded, with matrix details given in Appendix 2. Alignments are available online at http://arnoldlab.net/alignments.html.

Phylogenetic relationships for members of each class were inferred using parsimony and Bayesian Metropolis-coupled Markov chain Monte Carlo (MCMCMC) analyses. For the former, five sets of 200 heuristic searches with random stepwise addition and tree bisection-reconnection (TBR) were implemented using PAUPRat, which incorporates the ‘parsimony ratchet’ method (28, 48, 61) in PAUP* 4b10 (65).

The ‘filter trees’ command in PAUP* 4b10 was used to select all shortest trees, which 64

were used to assemble strict consensuses. Bootstrap support was determined using 1000 replicates and the default settings for nonparametric NJ analysis as implemented in

PAUP* 4b10. Clades with ≥70% bootstrap support were considered strongly supported.

Details of parsimony searches are given in Appendix 2.

For Bayesian analyses, Modeltest 3.7 (54) was used to select the appropriate model of evolution; in each case, GTR+I+G was selected using the Akaike Information

Criterion. Analyses were executed in Mr. Bayes 3.1.2 (31) for four runs of up to 6 million generations each, initiated with random trees, four chains, and sampling every

1000th generation. Likelihoods converged to a stable range for each data set and all trees prior to convergence were discarded as burn-in. Clades with > 95% posterior probability values were considered to have significant support.

Twenty-nine bacterial 16S DNA sequences, derived from total genomic DNA extractions of fungal endophyte mycelia with no evidence of extrahyphal bacteria

(Table 2), and seven 16S sequences from bacterial endophytes obtained directly from plant material (Table 3), were incorporated into a core alignment of 30 sequences of named bacterial taxa obtained from the Ribosomal Database Project 9 website (Cole et al. (15); http://rdp.cme.msu.edu) using seqmatch (type strains, isolates, >1200 bp, good quality; ca. 1400bp), together with eight additional sequences from GenBank that represented endosymbionts of fungi isolated in previous studies (Appendix 3) (9, 10,

53). Data were aligned using NAST and screened for chimeric sequences (minimum sequence length = 300 bp) (http://greengenes.lbl.gov/) (17). Ambiguous regions were excluded, resulting in a matrix of 6300 characters (http://arnoldlab.net/alignments.html). 65

For the bacterial data set, a heuristic search using maximum parsimony with random step-wise additions and TBR branch swapping was implemented in PAUP*

4b10, resulting in 167 optimal trees of 3668 steps. Support was assessed using a nonparametric neighbor-joining bootstrap (1000 replicates). Bayesian MCMCMC analysis was implemented in MrBayes v. 3.1.2 for 2.5 million generations, initiated with random trees, four chains, and sampling every 1000th tree, using GTR+I+G based on evaluation in Modeltest 3.7 (54). After elimination of the first 1,400 trees as burn-in, the remaining 1,100 trees were used to infer a majority rule consensus. These results were complemented by phylogenetic inference in ARB v. 7.12.07 (37) using the AxML maximum likelihood (ML) method (FastDNAML; olsen/felsenstein (F84); no filter: all positions included in phylogenetic analysis; -ln likelihood = 22430.51). Branch support was assessed using parametric ML bootstrapping (100 replicates) and Bayesian posterior probabilities.

Results

Cultivable endophytes were isolated from healthy foliage of all plant species and from every study site (Table 1), yielding 414 isolates from 4,560 tissue segments. ITS rDNA data for all isolates were assembled at 95% sequence similarity to yield 113 OTU

(putative species; Fisher’s alpha = 51.2). Representatives of 5 classes and approximately 10 orders, 13 families, and 28 genera were recovered. The most common orders differed among host species and sites, but the entire dataset comprised a high frequency of Pezizales (Pezizomycetes), Capnodiales, Pleosporales, Botryosphaeriales, 66

Dothideales (Dothideomycetes), Helotiales (Leotiomycetes), and Xylariales

(Sordariomycetes) (Appendix 4). Host specificity, diversity, and geographic distributions of these fungi will be addressed in a forthcoming paper (Hoffman &

Arnold, in prep.).

Endohyphal bacteria

Endohyphal bacteria initially were observed in 75 of 414 endophyte isolates (18%; determined by positive PCR results using 16S primers and no evidence of extrahyphal bacteria), including isolates from all four classes of Ascomycota recovered in our surveys (Pezizomycetes, Dothideomycetes, Eurotiomycetes, and Sordariomycetes), isolates from all three host genera (Cupressus, Juniperus, and Platycladus), and isolates from plants in all biogeographic regions (Table 2). After 47 of these isolates were subcultured, however, they later were screened as negative for bacteria. In general, length of time in culture (after ca. 2 weeks) appeared to adversely affect the detectability or persistence of bacterial infections.

Based on examination with light microscopy and Live-Dead stain, no isolate scored as positive for endohyphal bacteria had evident extrahyphal or ectosymbiotic bacteria. Live-Dead stain indicated that bacteria within hyphae were viable and confirmed viability of the hyphae that housed them (data not shown). Fluorescence in situ hybridization (FISH) for isolates 9084b and 9143 further confirmed the presence of endohyphal bacteria within living mycelia. Isolate 9084b is shown with the EUB338 fluorescent probe mounted in anti-fading mount medium only (Fig. 1A). Isolate 9143 is 67

shown with the EUB338 probe and DAPI counter-stain (Fig. 1B), illustrating the presence of small numbers of bacterial cells in comparison to significant amounts of total genomic DNA contained in the hyphal structure.

Phylogenetic inferences and taxonomic placement

Phylogenetic analyses confirmed BLAST-level taxonomy at the class level for fungal endophytes in the Dothideomycetes, Eurotiomycetes, and Sordariomycetes (Figs. 3-5).

No pattern regarding the structure of endophyte lineages as a function of host plant taxonomy was evident (i.e., the sister relationship of Cupressus and Juniperus, and their sister relationship to Platycladus, is not reflected in the phylogenetic relationships of their fungal associates). Isolates containing bacterial associates were spread broadly across endophyte-containing clades in each class, indicating a phylogenetically widespread capacity to harbor bacteria.

Taxonomic placement of endohyphal bacteria initially was assessed using

BLAST comparisons in GenBank, typically yielding matches to unidentified or nameless environmental samples. The RDP classifier (69) placed these bacteria into two phyla, Proteobacteria (including Alphaproteobacteria, Betaproteobacteria, and

Gammaproteobacteria) and Firmicutes. Overall, we recovered five orders and 10 families, putatively identified as Sphingomonadaceae (Alphaproteobacteria),

Burkholderiaceae, Comamonadaceae, and Oxalobacteraceae (Betaproteobacteria),

Moraxellaceae, Xanthomonadaceae, Pasteurellaceae, and Enterobacteriaceae

(Gammaproteobacteria), and Bacillaceae and Paenibacillaceae (Firmicutes) (Table 2). 68

Based on 97% 16S rDNA identity, these bacteria represented 15 OTU. Of these, nine were found only once. Three of the remaining six OTU were found in fungi from multiple genera of Cupressaceae, four were found in fungi from more than one biogeographic region, and all were found in fungi representing multiple genera. Four genotypes each were found in fungi representing different classes of Pezizomycotina

(genotypes A, B, C, E) (Table 2).

Proteobacteria were observed in association with fungi of the Dothideomycetes,

Eurotiomycetes, and Sordariomycetes, and in endophytes from all three plant genera representing hosts in AZ and NC. Firmicutes were found in Eurotiomycetes and one member of the Pezizomycetes. Nominal logistic analysis provided no evidence for a significant effect of host genus (Cupressus, Juniperus, or Platycladus), region (AZ vs.

NC), or fungal class on the incidence of major bacterial groups (Alpha-, Beta-, and

Gammaproteobacteria; Firmicutes) (P >0.05 for all effects).

Phylogenetic relationships of endohyphal bacteria from fungal endophytes were not congruent with those of the endophytes they inhabited (Figs. 2-5), nor with relationships among host plants (data not shown). The bacterial lineages recovered here were distinct from those recognized previously as endosymbionts of fungi (Fig. 2).

Comparison of bacterial endophytes and endohyphal bacteria

Only one bacterial genotype was represented both among bacterial endophytes (isolated from surface-sterilized plant tissue only) and putative endohyphal bacteria (genotype A,

Bacillales; Tables 2 and 3). This genotype was isolated directly from foliage of J. 69

osteosperma (Chuska Mountains) and C. arizonica and J. deppeana (Sky Islands), and was amplified from genomic DNA from a fungal culture recovered from J. deppeana

(Sky Islands). Two additional endophytes recovered here also were Bacillaceae, representing genotypes that never were found as endohyphal bacteria (genotypes H, J)

(Tables 2, 3). One bacterial endophyte represented the Enterobacteriaceae (putative

Erwinia) and was not found as a bacterial endosymbiont within fungal hyphae.

Discussion

Recent studies have highlighted the potential of endosymbiotic microbes, including bacteria and viruses, to shape the ecological roles of fungi (e.g., 43, 52). Endohyphal bacteria have been found previously in living hyphae of plant-associated

Glomeromycota, Mucoromycotina, and up to three taxa of mycorrhizal Dikarya. Our study is the first to compare their phylogenetic relationships with diverse fungal hosts, and the first to document their occurrence in ascomycetous endophytes of foliage representing four of the most species-rich classes of Ascomycota (Pezizomycetes,

Sordariomycetes, Dothideomycetes, and Eurotiomycetes), which together include numerous plant pathogens, saprotrophs, and pathogens and parasites of animals (31).

Coupled with previous studies demonstrating the presence of bacteria in living mycelia of mycorrhizal or soil-borne fungi (6, 7, 9, 52), our data provide strong evidence that the ability of fungi to harbor endohyphal bacteria is phylogenetically widespread.

Our survey data show that endohyphal bacteria-endophyte associations occur in regions with divergent biogeographic histories and markedly different environmental 70

conditions (e.g., diverse, aridland montane systems vs. piedmont forest in southeastern

North America) and in fungi associated with three genera of plants. Screening of a small number of isolates obtained in previous studies (Figs. 3-5) confirmed that endohyphal bacteria occur in endophytes from additional biogeographic provinces and host plant lineages: infections were observed in endophytes isolated from Pinaceae in boreal forest (endophyte 4466 from Picea mariana in Canada) and two families of angiosperms from a tropical forest (endophyte 6722 from Faramea occidentalis,

Rubiaceae, and endophyte 6731 from Swartzia simplex, Fabaceae, recovered at Barro

Colorado Island, Panama). Because infections either were lost or became more difficult to detect following subculturing, and because of potential limitations of our microscopy, screening methods and primer selection, our results likely underestimate the incidence of endohyphal bacteria in these plant-associated fungi.

Examination with Live-Dead stain provided evidence that bacteria within fungal hyphae were viable, and showed that such bacteria occur within living fungal tissues.

FISH confirmed these results by diagnosing the presence of bacteria using 16S rDNA specific tags (Fig. 1). In turn, our phylogenetic inferences show that endohyphal bacteria include a greater phylogenetic diversity – both within and beyond the

Proteobacteria – than was previously known (23, 33, 53; Fig. 2).

In contrast to previous studies, which have focused primarily on bacteria associated with a particular fungal species or closely related group of species, we addressed the diversity of bacteria associated with a phylogenetically broad sample of ecologically similar fungi. Our analyses provide no evidence of co-cladogenesis: fungal 71

and bacterial trees were obviously incongruent, and neither matched the phylogenetic relationships among the plant species surveyed here. Our phylogenetic results are consistent with the hypothesis of horizontal transmission both for fungal endophytes, as expected (1, 28), and for the bacterial symbionts examined here (Figs. 2-5).

Using cultivation media infused with antibiotics we have found that many of these fungi can be cured of their endosymbionts (Hoffman & Arnold, in prep.), and we currently are investigating methods to re-infect cured mycelia. These attempts will shed light on our hypotheses regarding the facultative nature of these associations, their transmission modes, and their costs and benefits with regard to the fungi they inhabit.

Notably, some endophytes were unable to grow when transferred to media containing antibiotics (Hoffman & Arnold, in prep.), suggesting that the relationship may be more intimate or tend toward greater obligacy in some pairs of fungi/bacteria than others. For one isolate (9143), we have successfully isolated the bacterial partner from the fungus and confirmed its identity using 16S rDNA sequencing (Hoffman et al., in prep.).

Attempts to reinfect the endophyte with this bacterium have been ineffective to date.

Discovering the diversity and ecological roles of fungal endophytes encompasses a trove of future research that will be important for understanding plant ecology and evolution. So too does elucidation of the incidence, diversity, and ecological importance of their endohyphal bacteria, which appear to be common but previously overlooked inhabitants of plant-symbiotic fungi. Evidence from a variety of studies (reviewed by 12, 32) indicates that endohyphal bacteria have the capacity to alter the outcome of plant-fungal interactions in a phylogenetically diverse array of 72

symbioses. Experimental assessment of such effects will provide key evidence to understanding the degree to which bacterial associates influence the nature of endophytic symbioses. In turn, incorporating ancestral state reconstructions and further sampling of both endohyphal bacteria (of fungi) and bacterial endophytes (of plants) will elucidate the degree to which bacteria have transitioned from endophytic to fungal- endosymbiotic, or the reverse – or have followed a distinctive evolutionary trajectory.

Acknowledgments

We thank the Division of Plant Pathology and Microbiology, School of Plant Sciences, and College of Agriculture and Life Sciences at the University of Arizona, and the

Arizona-Nevada Academy of Sciences, for supporting this work. Additional support from the National Science Foundation (NSF-0626520 and 0702825 to AEA; NSF-IGERT to

MTH) is gratefully acknowledged. We thank D.R. Maddison for sharing pre-release versions of Mesquite and Chromaseq; M. Gunatilaka, M. Shimabukuro, D. Grippi, M. J.

Epps, and C. Weeks-Galindo for lab assistance and helpful discussion; J. U’Ren, F.

Lutzoni, E. Gaya, J. Miadlikowska, and F. Santos-Rodriguez for isolation of samples from the Chiricahua Mountains; and B. Klein and undergraduate students at Diné College for isolation of samples from the Chuska Mountains on the Navajo Nation. We are especially grateful to H. Van Etten and R. Palanivelu for access to microscopy facilities,

A. Estes for EUB338 probes and helpful discussion about FISH, and J. L. Bronstein for comments on the manuscript. This paper represents a portion of the doctoral dissertation research of M.H. in Plant Pathology and Microbiology at the University of Arizona. 73

References

1. Arnold, A. E., J. Miadlikowska, K. L. Higgins, S. D. Sarvate, P. Gugger, A. Way, V. Hofstetter, F. Kauff, and F. Lutzoni. 2009. A phylogenetic estimation of trophic transition networks for ascomycetous fungi: are lichens cradles of symbiotrophic fungal diversification? Syst. Bio. 58:283–297.

2. Arnold, A. E., D. A. Henk, R. Eells, F. Lutzoni, & R. Vilgalys. 2007. Diversity and phylogenic affinities of foliar fungal endophytes in loblolly pine inferred by culturing and environmental PCR. Mycologia 99:185–206.

3. Arnold, A. E., L. Mejía, D. Kyllo, E. Rojas, Z. Maynard, N. A. Robbins, and E. A. Herre. 2003. Fungal endophytes limit pathogen damage in a tropical tree. Proc. Natl. Acad. Sci. USA 100:15649–15654.

4. Artursson, V., R. D. Finlay, and J. K. Jansson. 2005. Combined bromodeoxyuridine immunocapture and terminal-restriction fragment length polymorphism analysis highlights differences in the active soil bacterial metagenome due to Glomus mosseae inoculation or plant species. Environ. Microbiol. 7:1952–1966.

5. Barbieri. E., L. Potenza, I. Rossi, D. Sisti, G. Giomaro, S. Rossetti, C. Beimfohr, V. Stocchi. 2000. Phylogenetic characterization and in situ detection of a Cytophaga- Flexibacter-Bacteroides phylogroup bacterium in Tuber borchii Vittad. ectomycorrhizal mycelium. Appl. Environ. Microbiol. 66:5035–5042.

6. Barbieri, E., C. Guidi, J. Bertaux, P. Frey-Klett, J. Garbaye, P. Ceccaroli, R. Saltarelli, A. Zambonelli and V. Srocchi. 2007. Occurrence and diversity of bacterial communities in Tuber magnatum during truffle maturation. Environ. Microbiol. 9(9):2234–2246.

7. Bertaux J., M. Schmid, N. C. Prevost-Boure, J. L. Churin, A. Hartmann, J. Garbaye, and P. Frey-Klett. 2003. In situ identification of intracellular bacteria related to Paenibacillus spp. in the mycelium of the ectomycorrhizal fungus Laccarria bicolor S238N. Appl. Environ. Microbiol. 69:4243–4248.

8. Bertaux J., M. Schmid, P. Hutzler, A. Hartmann, J. Garbaye, and P. Frey-Klett. 2005. Occurrence and distribution of endobacteria in the plant-associated mycelium of the ectomycorrhizal fungus Laccaria bicolor S238N. Environ. Microbiol. 7:1786–1795.

9. Bianciotto, V., C. Bandi, D. Minerdi, M. Sironi, H. V. Tichy, and P. Bonfante. 1996. An obligately endosymbiont mycorrhizal fungus itself harbors obligately intracellular bacteria. Appl. Environ. Microbiol. 62:3005–3010.

10. Bianciotto V, E. Lumini, L. Lanfranco, D. Minerdi, P. Bonfante, and S. Perotto. 2000. Detection and identification of bacterial endosymbionts in arbuscular mycorrhizal fungi belonging to the family Gigasporaceae. Appl. Environ. Microbiol. 66:4503–4509.

11. Bianciotto, V., E. Lumini, P. Bonfante, and P. Vandamme. 2003. ‘Candidatus Glomeribacter gigasporarum’ gen. nov., sp. nov., an endosymbiont of arbuscular mycorrhizal fungi. Int. J. Syst. Evol. Microbiol. 53:121–124.

12. Bonfante, P. and I-A. Anca. 2009. Plants, mycorrhizal fungi, and bacteria: A network of interactions. Annu. Rev. Microbiol. 63:363–383.

13. Cannone, J. J., S. Subramanian, M. N. Schnare, J. R. Collet, L/ M. D’Souza, Y. Du, B. Feng, N. Lin, L. V. Madabusi, K. M. Muller, N. Pande, Z. Shang, N. Yu, and R. R. Gutell. 2002. The Comparative RNA Website (CRW): An online database of comparative sequence and structure information for ribosomal, intron, and other RNAs. BioMed. Cent. Bioinformatics 3:15.

14. Carpenter-Boggs, L., T. E. Loynachan, and P. D. Stahl. 1995. Spore germination of Gigaspora margarita stimulated by volatiles of soil-isolated actinomycetes. Soil Biol. Biochem. 27:1445–1451.

15. Cole, J. R., B. Chai, R. J. Farris, Q. Wang, A. S. Kulam-Syed-Mohideen, D. M. McGarrell, A. M. Bandela, E. Cardenas, G. M. Garrity, and J. M. Tiedje. 2007. The ribosomal database project (RDP-II): introducing myRDP space and quality controlled public data. Nuc. Acids Res. 35:D169–D172.

16. De Boer W., L. B. Folman, R. C. Summerbell, and L. Boddy. 2005. Living in a fungal world: Impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29:795–811.

17. DeSantis, T. Z., P. Hugenholtz, K. Keller, E. L. Brodie, N. Larsen, Y. M. Piceno, R. Phan, and G. L. Andersen. 2006. NAST: A multiple sequence alignment server for comparative analysis of 16S rRNA genes. Nucl. Acids Res. 34:W394–399.

18. Ewing, B., and P. Green. 1998. Base-calling of automated sequencer traces using Phred. II. Error probabilities. Genom. Res. 8:186–194.

19. Ewing, B., L. Hillier, M. Wendl, and P. Green. 1998. Base–calling of automated sequencer traces using Phred. I. Accuracy assessment. Genom. Res. 8:175–185.

20. Farjon, A., C. N. Page, and N. Schellevis. 1993. A preliminary world list of threatened conifer taxa. Biodiv. Conserv. 2:304–326.

21. Farjon, Aljos. 2008. A natural history of conifers. Timber Press. Portland, OR. 75

22. Fralish, J. S. and S. B. Franklin. 2002. Taxonomy and Ecology of Woody Plants in North American Forests. John Wiley and Sons, Inc., New York, N.Y.

23. Frey-Klett P., J. Garbaye, and M. Tarkka. 2007. The mycorrhiza helper bacteria revisited. New Phyto. 176:22–36.

24. Fröhlich, J., and K. D. Hyde. 1999. Biodiversity of palm fungi in the tropics: Are global fungal diversity estimates realistic? Biodiver. Conserv. 8:977–1004.

25. Gadek P. A., D. L. Alpers, M. M. Heslewood, and C. J. Quinn. 2000. Relationships within Cupressaceae sensu lato: A combined morphological and molecular approach. Amer. J. Bot. 87:1044–1057.

26. Garbaye, J. 1994. Helper bacteria - a new dimension to the mycorrhizal symbiosis. New Phyto. 128:197–210.

27. Gardes, M. and T. D. Bruns. 1993. ITS primers with enhanced specificity of basidiomycetes: application to the identification of mycorrhizae and rusts. Mol. Ecol. 2:113–118.

28. Hoffman, M. T. and A. E. Arnold. 2008. Geographic locality and host identity shape fungal endophyte communities in cupressaceous trees. Myco. Res. 112:331–344.

29. Hoffman, M.T., Gunatilaka, M., Wijeratne, E. M. K., Gunatilaka, A. A. L., and A.E. Arnold. In prep. Endohyphal bacteria of an ascomycetous fungal endophyte increases IAA production.

30. Huelsenbeck, J. P. and F. Ronquist. 2001. MRBAYES: Bayesian inference of phylogenetic trees. Bioinformatics. 17:754–755.

31. James, T. Y., F. Kauff, C. Schoch, P. B. Matheny, V. Hofstetter, C. Cox, G. Celio, C. Gueidan, E. Fraker, J. Miadlikowska, H. T. Lumbsch, A. Rauhut, V. Reeb, A. E. Arnold, A. Amtoft, J. E. Stajich, K. Hosaka, G.-H. Sung, D. Johnson, B. O’Rourke, M. Crockett, M. Binder, J. M. Curtis, J. C. Slot, Z. Wang, A. W. Wilson, A. Schüßler, J. E. Longcore, K. O’Donnell, S. Mozley-Standridge, D. Porter, P. M. Letcher, M. J. Powell, J. W. Taylor, M. M. White, G. W. Griffith, D. R. Davies, R. A. Humber, J. B. Morton, J. Sugiyama, A. Y. Rossman, J. D. Rogers, D. H. Pfister, D. Hewitt, K. Hansen, S. Hambleton, R. A. Shoemaker, J. Kohlmeyer, B. Volkmann-Kohlmeyer, R. A. Spotts, M. Serdani, P. W. Crous, K. W. Hughes, K. Matsuura, E. Langer, G. Langer, W. A. Untereiner, R. Lücking, B. Büdel, D. M. Geiser, A. Aptroot, P. Diederich, I. Schmitt, M. Schultz, R. Yahr, D. Hibbett, F Lutzoni, D. McLaughlin, J. Spatafora, and R. Vilgalys. 2006. Reconstructing the early evolution of the fungi using a six gene phylogeny. Nature. 443:818–822. 76

32. Kobayashi, D. Y. and J. A. Crouch. 2009. Bacterial/fungal interactions: From pathogens to mutualistic endosymbionts. Annu. Rev. Phytopathol. 47:63–82.

33. Lackner G, N. Möbius, K. Scherlach, L. P. Partida-Martinez, R. Winkler, I. Schmitt, and C. Hertweck. 2009. Global distribution and evolution of a toxinogenic Burkholderia- Rhizopus symbiosis. Appl. Environ. Microbiol. 75:2982–2986.

34. Lane, D. J. 1991. 16S/23S rRNA sequencing. (eds. Stackebrandt and Goodfellow in: Nucleic Acid Techniques in Bacterial Systematics. John Wiley and Sons, Inc., New York, N.Y.

35. Levy, A., B. J. Chang, L. K. Abbott, J. Kuo, G. Harnett, and T. J. J. Inglis. 2003. Invasion of spores of the arbuscular mycorrhizal fungus Gigaspora decipiens by Burkholderia spp. Appl. Environ. Microbiol. 69:6250–6256.

36. Little, D. P. 2006. Evolution and circumscription of the true cypresses (Cupressaceae: Cupressus). Syst. Bot. 31:461–480.

37. Ludwig W, O. Strunk, R. Westram, L. Richter, H. Meier, Yadhukumar, A. Buchner, T. Lai, S. Steppi, G. Jobb, W. Förster, I. Brettske, S. Gerber, A. W. Ginhart, O. Gross, S. Grumann, S. Hermann, R. Jost, A. König, T. Liss, R. Lüssmann, M. May, B. Nonhoff, B. Reichel, R. Strehlow, A. Stamatakis, N. Stuckmann, A. Vilbig, M. Lenke, T. Ludwig, A. Bode, K. H. Schleifer. 2004. ARB: a software environment for sequence data. Nucl. Acids Res. 32:1363–1371.

38. Lumini E, V. Bianciotto, P. Jargeat, M. Novero, A. Salvioli, A. Faccio, G. Bécard, and P. Bonfante. 2007. Presymbiotic growth and sporal morphology are affected in the arbuscular mycorrhizal fungus Gigaspora margarita cured of its endobacteria. Cell. Microbio. 9:1716–1729.

39. MacDonald, R.M. and M. R. Chandler. 1981. Bacterium-like organelles in the vesicular-arbuscular mycorrhizal fungus Glomus caledonius. New Phyto. 89:241–246.

40. Maddison, D. R. and W. P. Maddison. 2005. ChromaSeq Module. Mesquite: a modular system for evolutionary analysis. V1.06. http://mesquiteproject.org.

41. Maddison, W. P. and D.R. Maddison. 2009. Mesquite: a modular system for evolutionary analysis. Version 2.6, http://mesquiteproject.org.

42. Manz, W., R. Amann, W. Ludwig, M. Wagner, and K. H. Schleifer. 1992. Phylogenetic oligodeoxynucleotide probes for the major subclass of proteobacteria: Problems and solutions. Syst. Appl. Microbiol. 15:593–600.

43. Márquez, L. M., R. S. Redman, R. J. Rodriguez, and M. J. Roossinck. 2007. A virus in a fungus in a plant: Three-way symbiosis required for thermal tolerance. Science 315: 513–517.

44. Minerdi D, M. Moretti, G. Gilardi, C. Barberio, M. L. Gullino, and A. Garibaldi. 2008. Bacterial ectosymbionts and virulence silencing in a Fusarium oxysporum strain. Environ. Microbio. 10:1725–1741.

45. Mirabal-Alonso, L. and E. Ortega-Delgado. 2007. Phosphate solubilizing bacteria isolated from the inside of Glomus mosseae spores from Cuba. First International Meeting on Microbial Phosphate Solubilization. Series: Developments in Plant and Soil Sciences. Vol. 102 Velazquez, E.; Rodriguez-Barrueco, C. (Eds.) Reprinted from Plant and Soil. 287:1–84.

46. Moran NA, C. Dale, H. Dunbar, W. A. Smith, H. Ochman. 2003. Intracellular symbionts of sharpshooters (Insecta: Hemiptera: Cicadellinae) form a distinct clade with a small genome. Environ. Microbio. 5:116–126.

47. Murphy, D. B. 2001. Fundamentals of Light Microscopy and Electronic Imaging. Wiley-Liss. New York, N.Y.

48. Nixon, K.C. 1999. The parsimony ratchet, a new method for rapid parsimony analysis. Cladistics 15:407–414.

49. O’Brien, H. E., J. Miadlikowska, and F. Lutzoni. 2005. Assessing host specialization in symbiotic cyanobacteria associated with four closely related species of the lichen fungus Peltigera. Eur. J. Phyco. 40:363–378.

50. Pan, J.J., A. M. Baumgarten, and G. May. 2008. Effects of host plant environment and Ustilago maydis infection on the fungal endophyte community of maize (Zea mays). New Phyto. 178:147–156.

51. Parrent, J.L., T. Y. James, R. Vasaitis, and A. F. S. Taylor. 2009. Friend or foe? Evolutionary history of glycoside hydrolase family 32 genes encoding for sucrolytic activity in fungi and its implications for plant-fungal symbioses. BMC Evol. Bio. 9:148.

52. Partida-Martinez, L. P., and C. Hertweck. (2005). Pathogenic fungus harbours endosymbiotic bacteria for toxin production. Nature 437:884–888.

53. Partida-Martinez L. P., S. Monajembashi, K. O. Greulich, and C. Hertweck. 2007. Endosymbiont-dependent host reproduction maintains bacterial-fungal mutualism. Current Bio. 17:773–777.

54. Posada, D., and K. A. Crandall. 1998. MODELTEST: Testing the model of DNA substitution. Bioinformatics. 14:817–818.

55. Ruiz-Lozano, J.M. and P. Bonfante. 1999. Identification of a putative P-transporter operon in the genome of a Burkholderia strain living inside the arbuscular mycorrhizal fungus Gigaspora margarita. J. Bact. 181:4106–4109.

56. Saiki R. K., S. Scharf, F. Faloona, K. B. Mullis, G. T. Horn, H. A. Erlich, and N. Arnheim. 1985. Enzymatic amplification of beta-globin genomic sequences and restriction site analysis for diagnosis of sickle cell anemia. Science. 230:1350–1354.

57. Saunders, M. and L. M. Kohn. 2009. Evidence for alteration of fungal endophyte community assembly by host defense compounds. New Phyto. 182:229–238.

58. Schrey, S.D., M. Schellhammer, M. Ecke, R. Hampp, and M. T. Tarkka, 2005. Mycorrhiza helper bacterium Streptomyces AcH 505 induces differential gene expression in the ectomycorrhizal fungus Amanita muscaria. New Phyto. 168:205–216.

59. Sharma, M., M. Schmid, M. Rothballer, G. Hause, A. Zuccaro, J. Imani, P. Kämpfer, E. Domann, P. Schäfer, A. Hartmann, and K-H. Kogel. 2008. Detection and identification of bacteria intimately associated with fungi of the order Sebacinales. Cell. Microbio. 10(11):2235–2246.

60. Shefferson R. P., M. Weiss, T. Kull, D. L. Taylor. 2005. High Specificity generally characterizes mycorrhizal associations in rare lady’s slipper orchids, genus Cypripedium. Mol. Ecol. 14:613–626.

61. Sikes, D. S. and P. O. Lewis. 2001. Beta software, version 1. PAUPRat: PAUP* implementation of the parsimony ratchet. Distributed by the authors. Dept. Ecol. Evol. Bio., Univ. of Conn., Storrs, USA.

62. Stackebrandt, E. and B. M. Göbel. 1994. Taxonomic note: a place for DNA-DNA reassociation and t65 rRNA sequence analysis in the present species definition in bacteriology. Int. J. Syst. Bacteriol. 44:846–849.

63. Stanosz, G. R., J. T. Blodgett, D. R. Smith, and E. L. Kruger. 2001. Water stress and Sphaeropsis sapinea as a latent pathogen of red pine seedlings. New Phyto. 149:531– 538.

64. Stewart, A. and J. W. Deacon. 1995. Vital fluorochromes as tracers for fungal growth studies. Biotech. Histochem. 70:57–65.

65. Swofford, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony (* and other methods). Ver. 4b10 Sinauer Assoc., Sunderland, MA. 79

66. U'ren J. M., J. W. Dalling, R. E. Gallery, D. R. Maddison, E. C. Davis, C. M. Gibson, and A. E. Arnold. 2009. Diversity and evolutionary origins of fungi associated with seeds of a neotropical pioneer tree: a case study for analysing fungal environmental samples. Myco. Res. 113:432–449.

67. Vilgalys, R. and M. Hester. 1990. Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J. Bac. 172:4238–4246.

68. Waller, F., B. Achatz, H. Baltruschat, J. Fodor, K. Becker, M. Fischer, T. Heier, R. Hückelhoven, C. Neumann, D. von Wettstein, P. Franken, and K-H. Kogel. 2005. The endophytic fungus Piriformospora indica reprograms barley to salt-stress tolerance, disease resistance, and higher yield. Proc. Natl. Acad. Sci. USA 102:13386–13391.

69. Wang, Q., G. M. Garrity, J. M. Tiedje, and J. R. Cole. 2007. Naive bayesian classifier for rapid assignment of rrna sequences into the new bacterial taxonomy. Appl. Environ. Microbiol. 73:5261–5267.

70. White, T. J., T. Bruns, S. Lee, and J. Taylor. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis, M. A., Gelfand, D. H., Sninsky, J. J. & White, T. J. (Eds.) PCR protocols: A guide to methods and applications. Academic Press, San Diego, CA.

71. Xavier, L.J.C., and J. J. She Germida. 2003. Bacteria associated with Glomus clarum spores influence mycorrhizal activity. Soil Bio. Biochem. 35:471–478.

Table 1. Locations and characteristics of sampling sites, and plant species surveyed for endophytes in each location.

Latitude, Elevation Mean annual Mean annual Site Habitats Plant species Longitude (m) temperaturea precipitation

(°C) a (mm)

Piedmont, Central NC (DUKE)

Duke Forest, NC (NC) Temperate mixed forest 36.00°N, 78.94°W 97 15.5 1092 J. virginiana L., P. orientalis (L.) Franco

Petran forest, NE AZ (CHU)

Chuska Mtns, AZ (CHU) Ponderosa pine forest 36.16°N, 108.91°W 2200 10.4 253 J. osteosperma (Torr.) Little, J. scopulorum Sarg.

Mogollon Rim, Central AZ (MOG)

Iron Springs, AZ (IS) Pinyon-juniper woodland 34.58°N, 112.60°W 1851 11.8 482 C. arizonicab Greene, J. deppeana Stead

Mingus Mtn, AZ (MM) Ponderosa pine-pinyon-juniper woodland 34.70°N, 112.13°W 2354 15.3 483 J. deppeana Stead

Strawberry, AZ (ST) Ponderosa pine-pinyon-juniper woodland 34.40°N, 111.49°W 1707 13.2 541 J. deppeana Stead

Sky Islands , SE AZ (SKY)

Mt Lemmon, AZ (MTL) Madrean evergreen woodland 32.44°N, 110.78°W 1600 9.2 751 C. arizonicab Greene, J. deppeana Stead

Chiricahua Mtns AZ (CHI) Madrean evergreen woodland 32.03°N, 109.38°W 1371 11.7 549 J. deppeana Stead

Campus Arboretum, SE AZ (UA)

Univ. of Arizona (UA) Cultivated trees, semi-urban setting 32.23°N, 110.95°W 787 20.2 305 C. arizonicab Greene, J. virginiana L., P. orientalis (L.) Franco

a Climate information obtained from http://www.wrcc.dri.edu/coopmap/ b Recently designated change from Cupressus arizonica Greene to Callitropsis arizonica (Greene) is acknowledged (36).

Table 2. Taxonomic classification for 29 fungal endophytes and associated endohyphal bacteria. Bacterial taxonomic classifications

are based on BLAST results and Ribosomal Database project (RDP) classifier program (Naive Bayesian rRNA Classifier Version 2.0,

July 2007) (69). Fifteen distinct 16S rDNA OTU, based on 97% sequence similarity, are designated in groups A-R. Asterisks indicate

one genotype found also as a bacterial endophyte in plant tissue, such that its status as only an endosymbiont of fungi is not clear

(Genotype A; see Table 3). Fungal identifications were based on comparisons with the FMP curated database in March 2009 and are

approximate at the genus and species levels, but strongly supported at the class level based on phylogenetic analyses (Figs. 3-5).

16S Top 16S Isolate Host plant Site Top BLAST match Fungal class genotype BLAST/GenBank acc. Bacterial lineage (based on RDP classifier)

2611 J. scopulorum AZ: CHU Dothidea sambuci AY930108.1 Dothideomycetes K Acinetobacter sp. Gammaproteobacteria, Pseudomonadales, FJ422393 Moraxellaceae 9026 P. orientalis AZ: UA Alternaria mali AY154683.1 Dothideomycetes L Acinetobacter junii Gammaproteobacteria, Pseudomonadales, NR_026208 Moraxellaceae 9054 P. orientalis AZ: UA Phoma glomerata AY183371.1 Dothideomycetes M Uncultured Sphingomonas Alphaproteobacteria, Sphingomonadales, EU341143 Sphingomonadaceae 9055 P. orientalis AZ: UA Alternaria mali AY154683.1 Dothideomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae 9060 P. orientalis AZ: UA Aureobasidium pullulans AY225166.1 Dothideomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 9094 P. orientalis AZ: UA Lecythophora sp. AY219880.1 Sordariomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 9096 P. orientalis AZ: UA Hormonema sp. AF182378.1 Dothideomycetes N Oxalobacteraceae sp. Betaproteobacteria, Burkholderiales, DQ490310 Oxalobacteraceae 9106 C. arizonica AZ: UA Preussia africana DQ865095.1 Dothideomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae 9107 C. arizonica AZ: UA Phoma glomerata AY183371.1 Dothideomycetes F Uncultured Haemophilus Gammaproteobacteria, Pasteurellales, sp. EU071476 Pasteurellaceae 9112 C. arizonica AZ: UA Lecythophora sp. AY219880.1 Sordariomycetes P Variovorax paradoxus Betaproteobacteria, Burkholderiales, DQ257419 Comamonadaceae 9120 C. arizonica AZ: UA Monodictys sp. AJ972795.1 Sordariomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 9122 P. orientalis NC: Duke Pestalotiopsis caudata EF055188.1 Sordariomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae

9128 P. orientalis NC: Duke Cladosporium oxysporum AJ300332.1 Dothideomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 9133 P. orientalis NC: Duke Botryosphaeria dothidea AY640254.1 Dothideomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae 9135 P. orientalis NC: Duke Phyllosticta spinarum AF312009.1 Dothideomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae 9140 P. orientalis NC: Duke Microdiplodia sp. EF432267.1 Dothideomycetes R Pantoea agglomerans Gammaproteobacteria, Enterobacteriales, FJ756346.1 Enterobacteriaceae 9143 P. orientalis NC: Duke Pestalotiopsis caudata EF055188.1 Sordariomycetes C Uncultured bacterium Gammaproteobacteria, Xanthomonadales, clone DQ984599 Xanthomonadaceae 9145 P. orientalis NC: Duke Microdiplodia sp. EF432267.1 Dothideomycetes D Pantoea oleae Gammaproteobacteria, Enterobacteriales, AF130967 Enterobacteriaceae 9147 P. orientalis NC: Duke Penicillium citreoni AY373908.1 Eurotiomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 11123 J. virginiana AZ: UA Pithya cupressina U66009.1 Pezizomycetes G Bacillus foraminis Firmicutes, "Bacilli", Bacillales, AJ717382 Bacillaceae 11164 J. deppeana AZ: MTL Phaeomoniella chlamydospora Eurotiomycetes A* Bacillus subtilis Firmicutes, "Bacilli", Bacillales, AY772236.1 FJ549011 Paenibacillaceae 11259 J. deppeana AZ: MTL Biscogniauxia. mediterranea Sordariomycetes B Paenibacillus sp. Firmicutes, "Bacilli", Bacillales, AJ390413.1 AM906086 Paenibacillaceae 11272 J. deppeana AZ: MOG Phaeomoniella chlamydospora Eurotiomycetes B Paenibacillus sp. Firmicutes, "Bacilli", Bacillales, AB278179.1 AM906086 Paenibacillaceae 11035 J. deppeana AZ: MTL Pestalotiopsis besseyi AY687313.1 Sordariomycetes A* Bacillus atrophaeus Firmicutes, "Bacilli", Bacillales, FJ194961 Paenibacillaceae 9084b P. orientalis AZ: UA Aureobasidium pullulans DQ680686.1 Dothideomycetes F Uncultured bacterium Gammaproteobacteria, Pasteurellales, clone EF509196 Pasteurellaceae 9106 clA C. arizonica AZ: UA Preussia africana DQ865095.1 Dothideomycetes E Uncultured bacterium Betaproteobacteria, Burkholderiales, clone EU236303 Burkholderiaceae 9106 clB C. arizonica AZ: UA Preussia africana DQ865095.1 Dothideomycetes O Acinetobacter sp. Gammaproteobacteria, Pseudomonadales, FJ267573 Moraxellaceae 9126 clB P. orientalis NC: Duke Pestalotiopsis caudata EF055188.1 Sordariomycetes Q Pantoea agglomerans Gammaproteobacteria, Enterobacteriales, EU598802.1 Enterobacteriaceae 9149b P. orientalis NC: Duke Phyllosticta spinarum AF312009.1 Dothideomycetes D Pantoea oleae Gammaproteobacteria, Enterobacteriales, AF130967 Enterobacteriaceae

Table 3. Bacterial endophytes isolated from foliage of focal species of Cupressaceae, collection sites, 16S rDNA OTU based on 97% sequence similarity when compared among themselves and with the endohyphal bacteria listed in Table 2, top RDP classifier match and accession information, and bacterial lineage.

Isolate Plant species Location 16S genotype RDP Classifier match Bacterial lineage

11124 J. deppeana AZ: MTL H Bacillus megaterium FJ527650 Firmicutes, "Bacilli", Bacillales, Bacillaceae

11136 C. arizonica AZ: MTL A Bacillus subtilis FJ549011 Firmicutes, "Bacilli", Bacillales, Bacillaceae

11180 J. osteosperma AZ: CHU A Bacillus sp. FJ514812 Firmicutes, "Bacilli", Bacillales, Bacillaceae

11219 C. arizonica AZ: MTL A Bacillus subtilis NR_024931 Firmicutes, "Bacilli", Bacillales, Bacillaceae

11290 J. deppeana AZ: MTL A Bacillus subtilis FJ549011 Firmicutes, "Bacilli", Bacillales, Bacillaceae

11351 C. arizonica AZ: MOG I Erwinia persicina EU681952 Gammaproteobacteria, Enterobacteriales, Enterobacteriaceae

11394 J. virginiana AZ: UA J Bacillus licheniformis EU071553 Firmicutes, "Bacilli", Bacillales, Bacillaceae

Figure legends

Figure 1. Fluorescent in situ hybridization (FISH) microscopy showing hyphae of two isolates of endophytic fungi harboring endohyphal bacteria. A (isolate 9084b,

Dothideomycetes) shows TAMRA fluorophore at site of internal structure in hyphal cells.

B (isolate 9143, Sordariomycetes) shows TAMRA fluorophore with DAPI counterstain

(blue), highlighting the fungal nuclear and mitochondrial DNA in addition to the bacterial

16S rDNA (yellow/green). Scale bars: A, 10µm for image A and 25µm for image B.

Figure 2. Phylogenetic relationships of 29 endohyphal bacteria, 7 bacterial endophytes

(paired circles), and 38 named taxa based on Bayesian analysis of 16S ribosomal DNA sequences. Branch support values indicate parsimony bootstrap bootstrap proportions

(>70%; before slash) and Bayesian posterior probabilities (>95%; after slash). Branches in bold indicate >70% neighbor-joining bootstrap proportions. Named bacterial sequences and GenBank accession numbers are listed in Appendix 3; genotype groups for fungi are indicated by letters. Sequence data representing previously described bacterial endosymbionts of Glomeromycota and Mucoromycotina (9, 10, 52) are marked with solid squares.

Figure 3. Phylogenetic relationships of fungal endophytes with representative

Dothideomycetes based on majority rule consensus tree from Bayesian analysis. Branch support values indicate parsimony bootstrap proportions (>70%; before slash) and

Bayesian posterior probabilities (>95%; after slash). Taxon labels in bold indicate endophytes isolated in this study and screened for bacterial symbionts. Isolate names are 85

followed by plant species and collection site. Endophytes in which endohyphal bacteria were observed indicate the GenBank accession number for the top 16S BLAST match, taxonomic placement, and 16S rDNA genotype group. Closed circles indicate infections that were observed initially but were lost during cultivation. The open circle (isolate

6731) represents an infected endophyte from another study (1).

Figure 4. Phylogenetic relationships of endophytes with representative Eurotiomycetes based on majority rule consensus tree from Bayesian analysis of 25 LSU rDNA sequences. Branch support values indicate parsimony bootstrap proportions (>70%; before slash) and Bayesian posterior probabilities (>95%; after slash). Taxon labels in bold indicate endophytes isolated in this study and screened for bacterial symbionts.

Isolate names are followed by plant species and collection site. Endophytes in which endohyphal bacteria were observed indicate the GenBank accession number for the top

16S BLAST match, taxonomic placement, and 16S rDNA genotype group. The closed circle (isolate 11164) indicates an infection was lost during cultivation. The open circle

(isolate 4466) represents an infected endophyte from another study (1).

Figure 5. Phylogenetic relationships of endophytes with representative Sordariomycetes based on majority rule consensus tree from Bayesian analysis of 95 LSU rDNA sequences. Branch support values indicate parsimony bootstrap proportions (>70%; before slash) and Bayesian posterior probabilities (>95%; after slash). Taxon labels in bold indicate endophytes isolated in this study and screened for bacterial symbionts.

Isolate names are followed by plant species and collection site. Endophytes in which 86

endohyphal bacteria were observed indicate the GenBank accession number for the top

16S BLAST match, taxonomic placement, and 16S rDNA genotype group. Closed circles indicate infection was lost during cultivation. The open circle (isolate 6722) represents infected endophyte from another study (1). 87

Figure 1.

100/98 Candidatus Glomeribacter gigasporarum * AJ251633.1 n Figure 2. 94/100 Candidatus Glomeribacter gigasporarum * X897272 n 88 100/- Candidatus Glomeribacter gigasporarum * AJ251634.1 n 100/- Candidatus Glomeribacter gigasporarum * AJ251635.1 n 98/100 Burkholderia sp 20577 * AJ938141.1 n 100/- Burkholderia rhizoxinica * AJ938142.1 n 100/98 Burkholderia sp 30887 * AJ938143.1 n 100/- Burkholderia sp 69968 * AJ938144.1 n 97/100 Burkholderia fungorum * AF215705 98/100 Burkholderia sordidicola * AF512827 Burkholderia cepacia ATCC25416T * AB334766 99/100 99/100 Burkholderia pseudomallei ATCC23343 * CP000573 96/100 Ralstonia solanacearum ATCC11696 * EF016361 100/- Ralstonia mannitolilytica * AJ270258 99/100 9147 Penicillium * P. orientalis NC * EU236303 * E 100/100 9094 Lecythophora * P. orientalis UA * EU236303 * E 100/- 9060 Aureobasidium * P. orientalis UA * EU236303 * E 100/- 9106clA Preussia * C. arizonica UA * EU236303 * E 100/- 9120 Monodictys * C. arizonica UA * EU236303 * E 98/- 9128 Cladosporium * P. orientalis NC * EU236303 *E 99/- 9096 Hormonema * P. orientalis UA * DQ490310 * N 94/100 9112 Lecythophora * C. arizonica UA * DQ257419 * P 98/100 Nitrosomonas sp. BF16c52 * AF386747 96/100 Xanthomonas campestris LMG568T * AM920689 97/100 Xanthomonas oryzae LMG5047 * NR_026319 100/- 9133 Botryosphaeria * P. orientalis NC * DQ984599 * C 94/100 9135 Phyllosticta * P. orientalis NC * DQ984599 * C 100/- 9106 Preussia * C. arizonica UA * DQ984599 * C 100/- 9143 Pestalotiopsis * P. orientalis NC * DQ984599 * C 100/- 9055 Alternaria * P. orientalis UA * DQ984599 * C 98/- 9122 Pestalotiopsis * P. orientalis NC * DQ984599 * C 99/- Pseudomonas syringae * AJ308316 93/- 9026 Alternaria * P. orientalis UA * FJ596644 * L 99/- 2611 Dothidea * J. scopulorum CHU * FJ422393 * K 95/- 9106clB Preussia * C. arizonica UA * FJ267573 * O 91/100 9084b Aureobasidium * P.orientalis UA * EF509196 * F 95/- 9107 Phoma* C. arizonica UA * EU071476 * F 98/100 9145 Microdiplodia * P. orientalis NC * AF130967 * D 98/- 9149b Phyllosticta * P. orientalis NC * AF130967 * D 100/- Escherichia coli ATCC11775T * NR_024570 100/- Erwinia amylovora ATCC15580 * U80195 99/- 11351 C. arizona IS * EU681952 * I == 100/- Pantoea agglomerans ATCC27155 * FJ357815 9126clb Pestalotiopsis * P. orientalis NC * EU598802.1 * Q 98/- 9140 Microdiplodia * P. orientalis * NC EF432267.1 * R 100/100 9054 Phoma * P. orientalis UA * EU341143 * M 100/100 Sphingomonas mali IFO10550 * NR_026374 91/100 Caulobacter leidyia ATCC15260 * NR_025324 98/98 Sphingomonas phyllosphaerae * AY563441 96/100 Howardella ureilytica * DQ925472 96/100 Lactovum miscens * AJ439543 98/100 11259 Biscogniauxia * J. deppeana MTL * AM906086 * B 98/- 11272 Phaeomoniella * J. deppeana IS * AM906086 * B 98/100 Paenibacillus graminis * AJ223987 97/- Cohnella phaseoli * EU014872 96/100 11123 Pithya * J. virginiana UA * AJ717382 * G 99/99 11124 J. deppeana MTL * FJ542812 * H == 100/- Bacillus anthracis * AM747220 98/100 Bacillus cereus ATCC31293 * FJ601653 99/99 100/100 Bacillus mycoides * AF155956 11394 J. virginiana UA * EU071553 * J == 100/100 Bacillus licheniformis * AM913932 98/100 11180 J. osteosperma CHU * FJ514812 * A == 100/100 11035 Pestalotiopsis * J. deppeana MTL * FJ194961 * A 100/99 11219 C. arizonica MTL * NR_024931 * A == 100/99 11136 C. arizonica MTL * FJ549011 * A == 100/- 11164 Phaeomoniella * J. deppeana MTL * FJ549011 * A 96/99 11290 J. deppeana MTL * FJ549011 * A == 90/100 Desulfomicrobium orale * NR_025378 91/100 Desulfomicrobium norvegicum * NR_025407 86/100 Campylobacter helveticus * NR_025948 Campylobacter jejuni subsp. jejuni * CP000768 90/- Bacteroides ovatus ATCC8483T * X83952 Bacteroides fragilis ATCC25285T * CR626927

0.10 Figure 3. 89

Dendrographa minor 99/100 Botryosphaeria ribis 9133 Platycladus orientalis NC * DQ984599 (Xanthomonadaceae) * B 100/100 469 Laetia thamnia - Panama 98/- 2834 Dryas integrifolia - Quebec 100/100 79/100 87/100 2779 Dryas integrifolia - Quebec 3300 Dryas integrifolia - Quebec 1182 Magnolia grandiflora - North Carolina 100/- 6730 Swartzia simplex - Panama 100/100 6731 Swartzia simplex - Panama * O 100/100 435b Laetia thamnia - Panama 100/- 9149b Platycladus orientalis UA * AF130967 (Enterobacteriaceae) * C 9130 Platycladus orientalis NC 100/100 100/- 100/- 9144 Platycladus orientalis NC 9135 Platycladus orientalis NC * DQ984599 (Xanthomonadaceae) * B 9134 Platycladus orientalis NC 9129 Platycladus orientalis NC 83/- Microxyphium citri 98/100 Raciborskiomyces longisetosum 99/100 9128 Platycladus orientalis NC * EU236303 (Burkholderiaceae) * D 100/100 2712 Huperzia selago - Quebec 2722 Huperzia selago - Quebec 462 Laetia thamnia - Panama 100/100 100/- 4140 Dryas integrifolia - Quebec 100/- 100/100 3358A Dryas integrifolia - Quebec 4221 Dryas integrifolia - Quebec 100/100 3326 Picea mariana - Quebec 100/100 5007 Picea mariana - Quebec 97/- Sydowia polyspora 78/- 97/- 1029 Magnolia grandiflora - North Carolina DC2611 Juniperus scopulorum CHU * FJ422393 (Moraxellaceae) * H * 99/- 4109 Huperzia selago - Quebec 100/- 90/- Delphinella strobiligena 99/- 4947A Picea mariana - Quebec 81/100 4089 Picea mariana - Quebec 98/- Dothidea insculpa -/- Dothidea sambuci 97/- Dothidea ribesia Stylodothis puccinioides 9096 Platycladus orientalis UA * DQ490310 (Oxalobacteraceae) * K 87/- Discosphaerina fagi -/100 -/- 3275 Dryas integrifolia - Quebec 90/- 3335 Dryas integrifolia - Quebec 3353 Dryas integrifolia - Quebec 100/- 9042 Platycladus orientalis UA 100/100 -/- 9060 Platycladus orientalis UA * EU236303 (Burkholderiaceae) * D 100/- 9079 Platycladus orientalis UA -/- 9084b Platycladus orientalis UA * EF509196 (Pasteurellaceae) * F 100/- 9002 Platycladus orientalis UA 100/- 9009b Platycladus orientalis UA 88/- 9028 Platycladus orientalis UA -/- 9031 Platycladus orientalis UA 9036 Platycladus orientalis UA 100/100 3329 Picea mariana - Quebec 3377 Picea mariana - Quebec Lojkania enalia 100/100 Trematosphaeria heterospora 94/98 5095 Picea mariana - Quebec 99/98 Westerdykella cylindrica 100/100 -/98 Preussia terricola 9106 Cupressus arizonica UA * DQ984599 (Xanthomonadaceae) * B 99/- 9116 Cupressus arizonica UA 100/100 99/- 9120 Cupressus arizonica UA * EU236303 (Burkholderiaceae) * D 100/100 100/- 9104 Cupressus arizonica UA 9051 Platycladus orientalis UA 5718 Dryas integrifolia - Quebec -/- Byssothecium circinans 100/100 98/- 3323 Dryas integrifolia - Quebec Bimuria novae zelandiae Letendraea helminthicola -/- 100/- 9203 Juniperus virginiana NC 77/- 100/100 9200 Juniperus virginiana NC 11251 Cupressus arizonica MTL 89/- 9149a Platycladus orientalis UA 100/100 100/- 9145 Platycladus orientalis NC * AF130967 (Enterobacteriaceae) *C 9140 Platycladus orientalis NC * FJ756346.1 (Enterobacteriaceae) * O 9137 Platycladus orientalis NC 94/- Phoma glomerata 100/- 9158 Platycladus orientalis NC 100/100 82/- 9054 Platycladus orientalis NC * EU341143 (Sphingomonadaceae) * J -/- 9107 Cupressus arizonica UA * EU071476 (Pasteurellaceae) * F 85/- 9065 Platycladus orientalis UA 5654 Picea mariana - Quebec 71/- 100/- 9005 Cupressus arizonica UA 76/- 9008 Cupressus arizonica UA 82/- 9007 Platycladus orientalis UA 73/- 9018 Platycladus orientalis UA 98/- -/- 9015 Platycladus orientalis UA 9027 Platycladus orientalis UA 9030 Platycladus orientalis UA 100/100 Curvularia brachyspora 99/100 Cochliobolus heterostrophus 100/100 Setoshaeria monoceras 97/100 Pyrenophora tritici repentis 9055 Platycladus orientalis UA * DQ984599 (Xanthomonadaceae) * B 100/99 100/99 9026 Platycladus orientalis UA * NR_026208 (Moraxellaceae) * I 100/100 -/- 9021 Platycladus orientalis UA 9009a Platycladus orientalis UA 100/100 9057 Cupressus arizonica UA 97/100 9059 Cupressus arizonica UA -/- 9058 Cupressus arizonica UA Setomelanomma holmii -/- Ampelomyces quisqualis 99/- Phaeosphaeria avenaria 100/100 72/- 2851 Huperzia selago - Quebec 100/- 2706 Huperzia selago - Quebec 5246 Huperzia selago - Quebec Figure 4. 90

Peltula umbilicata

Arthroderma curreyi

100/100 Auxarthron zuffianum

100/- Aphanoascus fulvescens 100/100

Coccidioides posadasii 100/-

Paracoccidioides brasiliensis

100/100 Ascosphaera apis 100/100

100/- Eremascus albus

Byssochlamys nivea

100/100 Penicillium freii 100/100

9147 P. orientalis NC * EU236303 (Burkholderiaceae) * C

Pyrenula pseudobufonia

Pyrenula cruenta 100/-

4466 Picea mariana * Quebec * O

100/100 100/- 11247 Juniperus deppeana MIN

100/100 11272 J. deppeana IS * AM906086 (Paenibacillaceae) * A -/-

11164 J. deppeana MTL* FJ549011 (Paenibacillaceae) * E 100/-

Dermatocarpon luridum 100/-

Verrucaria pachyderma

-/- Capronia mansonii

422 Laetia thamnia * Panama 100/- 100/-

Exophiala jeanselmei 100/-

Capronia pilosella

100/- Ceramothyrium carniolicum 100/-

Glyphium elatum Figure 5. 91

Leotia lubrica 100/100 Lulworthia fucicola Lulworthia grandispora Thyridium vestutum Gaeumannomyces graminis var graminis 100/100 6702 Trichilia tuberculata - Panama -/99 Diaporthe phaseolorum 100/100 Valsa ambiens -/- 98/98 Cryptodiaporthe corni -/100 Cryphonectria cubensis -/100 -/95 Cryphonectria havanensis Melanconis marginalis Discula destructive Gnomonia setacea 100/100 -/95 100/100 Gnomoniella fraxini Plagiostoma euphorbiae -/- 100/100 Cephalotheca sulfurea Menispora tortuosa -/100 Camarops microspora -/- -/99 Farrowia longicollea Farrowia seminude -/100 100/100 Chaetomium globosum -/95 100/100 Neurospora crassa Sordaria macrospora -/- 100/100 Sordaria fimicola -/100 9038 Platycladus orientalis UA -/99 9097 Cupressus arizonica UA * -/100 9069 Platycladus orientalis UA -/- 9093 Platycladus orientalis UA -/- 9094 Platycladus orientalis UA * EU236303 (Burkholderiaceae) * D -/- 9092 Platycladus orientalis UA 9085 Platycladus orientalis UA Oxydothis frondicola -/- Microdochium nivale Hyponectria buxi -/100 100/100 4653 Picea mariana - Quebec 100/100 Arthrinium phaeospermum -/100 100/96 Apiospora sinensis -/- 6722 Faramea occidentalis - Panama * O 9143 Platycladus orientalis NC * DQ984599 (Xanthomonadaceae) * B -/- 100/100 9126 Platycladus orientalis NC * DQ984599 (Xanthomonadaceae) * N -/- 9121 Platycladus orientalis NC 9089 Platycladus orientalis UA 100/99 100/100 100/100 11035 Juniperus deppeana SKY * FJ194961 (Paenibacillaceae) * E Cryptosphaeria eunomia Fasciatispora petrakii -/- -/100 Cainia graminis -/- Seynesia erumpens -/100 Hypoxylon fragiforme -/96 -/100 256 Laetia thamnia - Panama 2204 Pinus taeda - NC 100/100 6717 Gustavia superba - Panama 11122 Cupressus arizonica SKY -/- 9202 Juniperus virginiana NC -/- 100/100 9201 Juniperus virginiana NC -/100 Astrocystis cocoes Rosellinia necatrix -/99 Xylaria acuta 100/99 -/- Xylaria hypoxylon 6728 Faramea occidentalis - Panama 100/94 Halorosellinia oceanica Pleurothecium recurvatum 6756 Gustavia superba - Panama 6741 Theobroma cacao - Panama 100/100 -/- 6733 Swartzia simplex - Panama 412 Laetia thamnia - Panama Theobroma cacao -/- 6749 - Panama 6714 Gustavia superba - Panama 151 Laetia thamnia - Panama -/100 Microascus trigonosporus 100/100 Ascosalsum cincinnatulum Magnisphaera stevemossago -/100 Sagaaromyces abonnis 100/100 Corollospora maritima Varicosporina ramulosa Ascosacculus heteroguttulatus -/99 Natantispora lotica 99/- 100/100 Halosarpheia marina 425 Laetia thamnia - Panama Torrubiella luteorostrata -/- 100/100 416 Laetia thamnia - Panama Gustavia superba -/- 6708 - Panama -/- 4939 Picea marina - Quebec -/- Balansia henningsiana -/100 Hydropisphaera erubescens Cordyceps khaoyaiensis 100/100 4780 Dryas integrifolia - Iqaluit -/- Emericellopsis terricola -/- 9154 Platycladus orientalis NC 100/100 9168 Platycladus orientalis NC -/100 Hypocrea citrina -/- Verticillium epiphytum -/95 Hypomyces polyporinus 89/- Cordyceps irangiensis Cordyceps sinensis 92

Appendix 1A. GenBank accession numbers for LSU rDNA sequences of endophytic fungi for phylogenetic analyses. All endophyte sequences were obtained and published by Higgins et al., 2007.

Endophyte isolate Accession no.

151 EF420086 256 EF420088 412 EF420089 416 EF420090 422 EF420091 425 EF420092 435b EF420093 462 EF420094 469 EF420095 1029 EF420084 1182 EF420085 2204 EF420087 2706 DQ979418 2712 DQ979419 2722 DQ979420 2779 DQ979422 2834 DQ979423 2851 DQ979424 3275 DQ979425 3300 DQ979427 3323 DQ979428 3326 DQ979429 3329 DQ979430 3335 DQ979431 3353 DQ979432 3358A DQ979433 3377 DQ979434 4089 DQ979437 4109 DQ979438 4140 DQ979440 4221 DQ979441 4466 DQ979444 4653 DQ979449 4780 DQ979452 4939 DQ979453 4947A DQ979454 5007 DQ979455 5095 DQ979457 5246 DQ979458 93

5654 DQ979462 5718 DQ979463 6702 EF420096 6708 EF420097 6714 EF420098 6717 EF420099 6722 EF420100 6728 EF420101 6730 EF420102 6731 EF420103 6733 EF420104 6741 EF420105 6749 EF420106 6756 EF420107

Appendix 1B. Representative Ascomycota species used in phylogenetic analyses with GenBank identification numbers. Sequences for Coccidioides posadasii were obtained from TIGR (http://www.tigr.org).

Taxon Accession no. Ampelomyces quisqualis 31415592 Aphanoascus fulvescens 33113349 Apiospora sinensis 27447246 Arthrinium phaeospermum 27447247 Arthroderma curreyi 33113367 Ascosacculus heteroguttulatus 34453082 Ascosalsum cincinnatulum 34453080 Ascosphaera apis 11120650 Astrocystis cocoes 27447238 Auxarthron zuffianum 33113353 Balansia henningsiana 45155172 Bimuria novae-zelandiae 15808399 Botryosphaeria ribis 11120642 Byssochlamys nivea 33113391 Byssothecium circinans 15808400 Cainia graminis 20334356 Camarops microspora 27447236 Capronia mansonii 11120644 Capronia pilosella 12025061 Cephalotheca sulfurea 20334357 Ceramothyrium carniolicum 11120645 Chaetomium globosum 13469777 Coccidioides posadasii TIGR222929 Cochliobolus heterostrophus 45775574 Cordyceps irangiensis 13241779 Cordyceps khaoyaiensis 13241765 Cordyceps sinensis 29466997 Corollospora maritima 22203655 Cryphonectria cubensis 22671402 Cryphonectria havanensis 22671403 Cryptodiaporthe corni 22671407 Cryptosphaeria eunomia var. eunomia 27447241 Curvularia brachyspora 12025063 Delphinella strobiligena 15808401 Dendrographa leucophaea f. minor 47499217 Dermatocarpon luridum 52699696 Diaporthe phaseolorum 1399144 Discosphaerina fagi 15808402 Discula destructiva 22671422 Dothidea insculpta 52699697 Dothidea ribesia 15808403 Dothidea sambuci 45775610 Emericellopsis terricola 1698507 Eremascus albus 11120651 Exophiala jeanselmei 4206347 95

Fasciatispora petrakii 27447243 Farrowia longicollea 13469782 Farrowia seminuda 13469784 Gaeumannomyces graminis var. graminis 16565889 Glyphium elatum 17104831 Gnomonia setacea 16565895 Gnomoniella fraxini 16565884 Halorosellinia oceanica 27447237 Halosarpheia marina 34453085 Hydropisphaera erubescens 45155165 Hypocrea citrina 45775578 Hypomyces polyporinus 32261014 Hyponectria buxi 27447249 Hypoxylon fragiforme 27447244 Leotia lubrica 45775573 Letendraea helminthicola 15808405 Lojkania enalia 15808406 Lulworthia fucicola 22203665 Lulworthia grandispora 22203666 Magnisphaera stevemossago 34453095 Melanconis marginalis 22671436 Menispora tortuosa 45775611 Microascus trigonosporus 1399149 Microdochium nivale 2565289 Microxyphium citri 11120643 Natantispora lotica 34453084 Neurospora crassa 13469785 Oxydothis frondicola 27447250 Paracoccidioides brasiliensis 2271524 Peltula umbilicata 15216700 Penicillium freii 52699706 Phaeosphaeria avenaria 45775613 Phoma glomerata 31415593 Plagiostoma euphorbiae 22671445 Pleurothecium recurvatum 45775614 Preussia terricola 45775615 Pyrenophora tritici-repentis 45775601 Pyrenula cruenta 12025090 Pyrenula pseudobufonia 52699710 Raciborskiomyces longisetosum 15808410 Rosellinia necatrix 27447239 Sagaaromyces abonnis 34453078 Setomelanomma holmii 28144315 Setosphaeria monoceras 15808411 Seynesia erumpens 12025093 Sordaria fimicola 45155203 Sordaria macrospora 37992293 Stylodothis puccinioides 11120648 Sydowia polyspora 45775604 Thyridium vestitum 45775600 Torrubiella luteorostrata 13241778 Trematosphaeria heterospora 15808412 Varicosporina ramulosa 22203671 96

Verrucaria pachyderma 15216679 Verticillium epiphytum 8777749 Valsa ambiens 16565896 Westerdykella cylindrica 11120649 Xylaria acuta 45775605 Xylaria hypoxylon 45775577

Appendix 2. Details of phylogenetic analyses for three classes of fungi.

Parsimony Bayesian analyses analyses

Class Sequences Included Length of Trees excluded Trees in (this study/ characters shortest trees as burn-in majority rule named taxa) (total) consensus

Dothideomycetes 107 (49/58) 1208 (4595) 761 3585 5415

Eurotiomycetes 25 (4/21) 1056 (4443) 418 3000 3400

Sordariomycetes 95 (17/78) 1218 (4605) 1313 5550 8890 97

Appendix 3. Representative species of bacteria used in phylogenetic analyses with GenBank identification numbers.

GenBank Taxon Accession

Bacillus anthracis AM747220 Bacillus cereus sp. FJ601653 Bacillus sp. L164 AM913932 Bacillus mycoides AF155956 Bacteroides fragilis CR626927 Bacteroides ovatus ATCC 8483T X83952 Burkholderia cepacia ATCC 25416T AB334766 Burkholderia fungorum AF215705 Burkholderia pseudomallei ATCC 23343 CP000573 Burkholderia rhizoxinica AJ938142.1 Burkholderia sordicola AF215705 Burkholderia sp 20577 AJ938141.1 Burkholderia sp 30887 AJ938143.1 Burkholderia sp 69968 AJ938144.1 Campylobacter helveticus NR_025948 Campylobacter jejuni subsp jejuni CP000768 Candidatus Glomeribacter gigasporarum X897272 Candidatus Glomeribacter gigasporarum AJ251633.1 Candidatus Glomeribacter gigasporarum AJ251634.1 Candidatus Glomeribacter gigasporarum AJ251635.1 Caulobacter leidyia strain ATCC 15260 NR_025324 Cohnella phaseoli EU014872 Desulfomicrobium norvegicum NR_025407 Desulfomicrobium orale NR_025378 Erwinia amylovora ATCC15580 U80195 Escherichia coli ATCC11775T NR_024570 Howardella ureilytica strain DQ925472 Lactovum miscens AJ439543 Nitrosomonas sp. BF16c52 AF386747 Paenibacillus graminis AJ223987 Pantoea agglomerans ATCC27155 FJ357815 Pseudomonas syringae AJ308316 Ralstonia mannitolilytica AJ270258 Ralstonia solanacearum ATCC 11696 EF016361 Sphingomonas mali IFO10550 NR_026374 Sphingomonas phyllosphaerae AY563441 Xanthomonas campestris pv. campestris AM920689 Xanthomonas oryzae strain LMG 5047 NR_026319

Appendix 4. Percentages of taxonomic orders of Pezizomycotina obtained in each biogeographic region based on FMP curated ITS database results. Remaining taxa not shown here include mitosporic fungi, fungi of uncertain placement, unidentifiable fungi, and a small number of isolates from lineages other than the Pezizomycotina, such that columns may not sum to 100%.

Percentage of isolates by SKY MOG CHU UA DUKE taxonomic orders Botryosphaeriales 11.1% 9.9% 1.6% 0 44.0% Capnodiales 11.1% 1.4% 1.6% 11.3% 1.2% Chaetothyriales 0 1.4% 0 1.6% 0 Coniochaetales 0.7% 4.2% 3.2% 6.5% 0 Diaporthales 0.7% 0 3.2% 0 2.4% Dothideales 7.4% 8.5% 4.8% 12.9% 0 Eurotiales 0 0 9.7% 0 1.2% Halosphaeriales 0 0 3.2% 3.2% 0 Helotiales 5.9% 5.6% 12.9% 0 2.4% Hypocreales 0 0 3.2% 0 4.8% Myriangiales 0 0 0 0 1.2% Pezizales 25.2% 21.1% 27.4% 4.8% 1.2% Pleosporales 5.2% 11.3% 8.1% 35.5% 4.8% Sordariales 0.7% 1.4% 0 3.2% 0 Xylariales 18.5% 12.7% 8.1% 1.6% 17.9%

APPENDIX C

IAA PRODUCTION BY ENDOPHYTIC PESTALOTIOPSIS NEGLECTA IS

ENHANCED BY AN ENDOHYPHAL BACTERIUM

100

IAA production by endophytic Pestalotiopsis neglecta is enhanced by an endohyphal bacterium

Michele T. Hoffman1, Malkanthi K. Gunatilaka1, E. M. Kithsiri Wijeratne2, A. A. Leslie

Gunatilaka2 and A. Elizabeth Arnold1

1. College of Agriculture and Life Sciences

Division of Plant Pathology and Microbiology

School of Plant Sciences

1140 E. South Campus Drive, Forbes 303

University of Arizona

Tucson, AZ 85721 USA

2. Southwest Center for Natural Products Research and Commercialization

Office of Arid Lands Studies

250 E. Valencia Road

University of Arizona

Tucson, Arizona 85706 USA

For correspondence: AE Arnold, Tel: 520.621.7212; Fax: 520.621.9290; Email:

[email protected]

Running head: Endohyphal bacterium enhances IAA production by fungal endophyte

Contains 5 figures, and 2 appendices

101

Abstract

A variety of plant-associated microbes produce phytohormones with pervasive effects on plant growth. Specifically, indole-3-acetic acid (IAA) production has been recorded among diverse plant pathogens, mycorrhizal and rhizosphere endophytes, and in endophytic bacteria and yeasts from healthy, above-ground plant tissues. Here, we show for the first time that a filamentous foliar endophyte, Pestalotiopsis neglecta, produces

IAA, and that its production is significantly enhanced when the endophyte hosts an endohyphal bacterium (Luteibacter sp.). When cultivated axenically, the bacterium does not produce IAA in culture. Both P. neglecta and the P. neglecta/Luteibacter complex appear to rely on L-tryptophan for IAA synthesis. Culture filtrate from the endophyte- bacterium complex significantly increased root growth in vitro in seedlings of tomato relative both to controls and to filtrate from the endophyte alone. Our results reveal that this endophyte, recovered from a coniferous host (Platycladus orientalis), has the potential to strongly influence growth of a distantly related plant, demonstrate the previously unknown role of an endohyphal bacterium in enhancing phytohormone production by a plant-symbiotic fungus, and provide a new lens with which to view the considerable genotypic, species-level, and functional diversity of foliar endophytic fungi.

Keywords: bacterial endosymbionts, IAA, auxins, Pestalotiopsis, Luteibacter,

Xanthomonadales, fungal endophytes, phylogeny

102

Introduction 1

Benefits conferred to plants by mycorrhizal fungi, including enhanced uptake of nutrient 2 and water, are well-documented and generally represent an intuitively straightforward 3 alignment of plant- and fungal fitness (Parniske, 2008; Bonfante and Anca, 2009; 4

Kobayashi and Crouch, 2009; Hoeksema et al., 2010). In contrast, the potential 5 contributions of foliar endophytes to plant physiology are both less obvious and less 6 thoroughly studied. Such contributions are especially difficult to resolve in the case of 7 endophytes associated with woody plants, which frequently comprise an exceptional 8 richness of highly localized infections at the scale of only a few millimeters of leaf area 9

(Carroll and Petrini, 1983; Lodge et al., 1996; Arnold et al., 2000; Arnold, 2007). These 10

Class 3 endophytes – a term used to refer to the species-rich, phylogenetically diverse, 11 horizontally transmitted, tissue-specific, and localized endophytes known from all major 12 lineages of land plants (Rodriguez et al., 2009) – are a fundamental feature of plant 13 biology, occurring in plants from the tundra to tropical rainforests (Higgins et al., 2007; 14

Arnold and Lutzoni, 2007). The benefits or costs they extend to their hosts, the nature of 15 their interactions with host tissues on the intimate scales of cells and molecules, and the 16 mechanisms by which they escape, tolerate, or prevent induction of host defenses 17 represent three major questions in the study of these ubiquitous plant symbionts (Arnold 18 et al., 2003; Waller et al., 2005; Arnold and Engelbrecht, 2007). 19

Over recent decades, a large number of studies have documented the ability of 20 diverse microbes to synthesize phytohormones such as gibberellins, ethylene, cytokinins, 21 jasmonic acid, abscisic acid, and especially indole-3-acetic acid (IAA), often with 22 profound effects on plant growth, tissue differentiation, and reproduction (Costacurta and 23

103

Vanderleyden, 1995; Glick, 1995; Lindow et al., 1998; Robinson et al., 1998; Koga, 24

1995; Maor et al., 2004; Sirrenberg et al., 2007; Spaepen et al., 2007). Most of these 25 studies have focused on plant-pathogenic microbes, mycorrhizal fungi, rhizosphere 26 endophytes, and endophytic bacteria and yeasts from above-ground tissues (Gruin, 1959; 27

Sosa-Morales et al., 1997; Nassar et al., 2005; van der Lelie et al., 2009; Chagué et al., 28

2009; Xin et al., 2009). Together, they reveal that microbial production of IAA is 29 phylogenetically widespread, encompassing both bacteria (e.g., Erwinia herbicola; Yang 30 et al., 2006; Pantoea agglomerans; Barash and Manulis-Sasson, 2009) and diverse 31 lineages of fungi (e.g., Mucoromycotina, Basidiomycota, and some Ascomycota; Strack 32 et al., 2003; Bajo et al., 2008; Reineke et al., 2008). 33

In many cases, IAA production by plant-associated microbes strongly influences 34 plant physiology in ways that are relevant to plant-microbe interactions. For example, 35 production of IAA by some mycorrhizal fungi stimulates production of plant biomass and 36 promotes disease resistance (Bonfante and Anca, 2009). Sirrenberg et al. (2007) noted an 37 increase in Arabidopsis root growth following exposure to diffusible IAA in filtrate from 38 liquid cultures of the rhizosphere fungus Piriformospora indica (Sebacinales, 39

Basidiomycota). Microbial IAA production can inhibit plant hypersensitive responses, 40 limiting the production of plant-defense compounds such as chitinase and glucanase 41 enzymes (Tudzynski and Sharon, 2002). 42

Here, we show for the first time that Pestalotiopsis neglecta, a filamentous foliar 43 endophyte recovered from Platycladus orientalis (Cupressaceae), produces IAA in vitro, 44 and that its production is enhanced significantly when the endophyte hosts an endohyphal 45 bacterium (Luteibacter sp.). Culture filtrate from the endophyte-bacterium complex 46

104

significantly increased root growth in seedlings of a distantly related plant (tomato) 47 relative both to controls and to filtrate from the endophyte alone. Our results highlight the 48 previously unknown role of an endohyphal bacterium in enhancing phytohormone 49 production by a plant-symbiotic fungus and provide a new lens with which to view the 50 remarkable genotypic, species-level, and functional diversity of foliar endophytic fungi. 51

Methods 53

Identification of fungal endophyte 9143 54

As part of a larger study, an endophytic fungus (9143) was collected from surface- 55 sterilized foliage of a mature, healthy individual of Platycladus orientalis (Cupressaceae) 56 in Durham, NC (Hoffman and Arnold, 2008). The isolate was archived as a living 57 voucher in sterile water at the Robert L. Gilbertson Mycological Herbarium (ARIZ) at the 58

University of Arizona. Previous phylogenetic analyses (Hoffman and Arnold, 2010) 59 confirmed ordinal placement of this isolate (Xylariales) but were insufficient to identify it 60 to species. After observing its bacterial endosymbiont and IAA production (below), we 61 identified the isolate on the basis of conidial morphology (Steyaert, 1953) and then 62 confirmed our designation using phylogenetic analyses. 63

For morphological characterization, we examined conidia of cultures grown on 64

2% malt extract agar media for 7 days (Appendix 1). For phylogenetic analyses, we 65 examined ITS rDNA sequence data for 9143 (obtained previously; see Hoffman and 66

Arnold, 2008) in the context of 59 publicly available ITS rDNA sequences for 67

Pestalotiopsis spp. (Jeewon et al, 2003, 2004; Wei et al., 2007; Watanabe et al, 2010). 68

We manually aligned these 60 sequences using Mesquite version 1.06 (Maddison and 69

105

Maddison, 2005). Taxon sampling included the breadth of close relatives of P. neglecta, 70 representing our estimation of taxonomic placement based on morphology. 71

Phylogenetic relationships were inferred using parsimony and Bayesian 72

Metropolis-coupled Markov chain Monte Carlo (MCMCMC) methods. For maximum 73 parsimony (MP), a heuristic search with random stepwise addition and tree bisection- 74 reconnection (TBR) was implemented in PAUP* 4b10 (Swofford, 2002). Support was 75 assessed using parsimony bootstrap (1000 replicates). Bayesian MCMCMC analysis was 76 implemented in MrBayes v. 3.1.2 (Huelsenbeck and Ronquist; 2001) for 2 sets of 3 77 million generations each, initiated with random trees, four chains, and sampling every 78

1000th tree, using GTR+I+G based on sequence evaluation in Modeltest 3.7 (Posada and 79

Crandall, 1998). After elimination of the first 500 trees as burn-in, the remaining 5002 80 trees were used to infer a majority rule consensus. Sequence data for this isolate are 81 deposited in GenBank (ITS data used for species-level identification, as well as 82

LSUrDNA data from Hoffman and Arnold, 2010; EF419899.1, EF420070.1). 83

Identification of endohyphal bacterium 85

Using light microscopy and PCR with 16S rDNA primers, we determined that endophyte 86

9143 harbored an endohyphal bacterium, which we identified previously only as a 87 member of the Gammaproteobacteria (Hoffman and Arnold, 2010). To place this 88 bacterium at the species level, we sampled more densely within the phylum, with a 89 special focus on Xanthomonaceae, using 36 16S rDNA sequences obtained by BLAST 90 matches to our 16S rDNA sequence (obtained using 10F/1507R primers; HM117737). 91

Manual alignment was conducted in Mesquite v. 1.06 (Maddison and Maddison, 2005) 92

106

and the resulting data set analyzed using parsimony and Bayesian MCMCMC methods. 93

Relationships were inferred using Felsenstein non-parametric distance methods, saving 94

10 best trees in each replicate set of 100 in PAUP* 4b10 (Swofford, 2002). Using 95 maximum parsimony (MP), a heuristic search with random stepwise addition and tree 96 bisection-reconnection (TBR) was implemented in PAUP*. Support was then assessed 97 using parsimony bootstrap (1000 replicates). Bayesian analysis was implemented in 98

MrBayes v. 3.1.2 (Huelsenbeck and Ronquist, 2001) on the CIPRES teragrid portal 99

(Miller et al., 2009; http://www.phylo.org/sub_sections/portal) for 2 runs of 10 million 100 generations, initiated with random trees, four chains, and sampling every 1000th tree, 101 using GTR+I+G based on evaluation in Modeltest 3.7 (Posada and Crandall, 1998). 102

After elimination of the first 440 trees as burn-in, the remaining 19,122 trees were used to 103 infer a majority rule consensus. 104

105 Production of bacterium-free strain 106

To eliminate the bacterium from the endophyte, we grew isolate 9143 on 2% MEA 107 amended with 40µg/ml of antibiotic ciprofloxacin. Infection status (infected, cured) 108 was confirmed using light microscopy (400X), which ruled out external contaminants, 109 followed by DNA extraction and 16S rDNA PCR (Hoffman and Arnold, 2010). 110

Infected and cured cultures of 9143 were stored at -80°C in 80% glycerol. We found no 111 difference between infected and cured isolates of 9143 in terms of growth rate on 2% 112

MEA or water agar at 22°C, nor with regard to pH of the growth medium (Appendix 2). 113

114

Measurement of indole compound production in vitro 115

107

Small samples of mycelium from infected and cured isolates were plated on 2% MEA 116 and allowed to grow for 7 d. Small pieces were removed from the resulting cultures 117 using a 0.5mm cork borer and used to inoculate three sterile flasks containing 80ml of 118

Czapek Dox broth (CDB; Hymedia; pH 7.2) augmented with L-tryptophan (5mM). For 119 each treatment set, one additional flask containing only CDB was used as a negative 120 control. All flasks were checked for contamination (turbidity) after agitating at 120rpm 121 at 26°C for 24 h. Contaminated cultures were discarded and replaced. 122

After 72 hours of agitation, three 1ml aliquots of broth were removed from each 123 flask and treated with colorimetric Salkowski reagent (0.01 M Fe (NO3)3; 7.0 M 124

HCIO4) prior to standard spectophotometric measurements at 530nm (Salkowski, 1885; 125

Ehmann, 1977; Glickmann and Dessaux, 1995). Each 1ml extract was combined with 126

500µl of Salkowski reagent, incubated at room temperature for 30 minutes, and then 127 measured once on a Whatman spectrophotometer (530nm). CDB was used as a blank 128 between readings. The mean of the three readings was calculated for each culture, 129 concentrations inferred using an IAA standard curve (below), and analyzed with a t-test 130 in JMP® 8.0 (SAS Institute Inc., Cary, NC.). 131

The IAA standard curve was inferred using commercial IAA (MW 175.19, 132

Sigma; 1mM), prepared in a volumetric flask using 95% molecular grade methanol. 133

Fourteen 1.5ml Eppendorf™ tubes were prepared in proportional ratios with 25-1950µl 134 of IAA and 50-1975µl of reagent. After incubating in the dark for 30 minutes, each 135 sample was measured twice. A blank calibration control using 1ml water plus 500µl of 136 reagent was as a blank between readings. 137

138

108

Identification of the indole compound as IAA using chromatography 139

We used thin layer chromatography (TLC) and high performance liquid chromatography 140

(HPLC) to identify the indole compound that was observed on the basis of colorimetric 141 reaction with Salkowski reagent. For chromatographic analysis, two cultures each of 142 infected and cured mycelia were grown for two weeks in 500mL of sterile CDB 143 supplemented with 5mM L-tryptophan. One additional culture (9143-infected) was 144 grown in sterile CDB without L-tryptophan (1.0 L) to determine whether fungus 9143 145 uses a tryptophan-dependent pathway for IAA production. Each culture was filtered 146 using Whatman No.1 filter paper and the filtrate (1.0 L) extracted three times with 147

500mL of ethyl acetate (EtOAc). The combined EtOAc layer was washed three times 148 with 500mL of H2O, dried over anhydrous Na2SO4, and evaporated under reduced 149 pressure to yield the EtOAc extract. 150

Normal phase TLC analysis of the EtOAc extract was performed on aluminum- 151 backed plates coated with 0.20 mm layer of silica gel 60 F254 (E. Merck, Darmstadt). 152

Spots were visualized by inspection of plates under UV light (254 nm) and after spraying 153 with Van Urk-Salkowski reagent (Ehmann, 1977) followed by heating. Commercially 154 available IAA (Sigma) was used as an authentic sample (Eluant: 8% MeOH in 155 dichloromethane, Rf 0.3). Analytical reversed phase TLC investigations were performed 156 on aluminum-backed plates coated with 0.20 mm layer of silica gel 60 RP-18 F254S (E. 157

Merck, Darmstadt; Eluant: 20% MeCN in H2O, Rf 0.5). 158

Analytical HPLC analysis of the EtOAc extract was performed using a Kromasil 159

5µm C-18 column (4.6 x 250 mm) on a Shimadzu HPLC system equipped with DGU- 160

14A degasser, LC-10ADvp pump, SPD-M10Avp diode array detector and SCL-10Avp 161

109

system controller utilizing Shimadzu LC-MS solution software. Samples were 162 redissolved in MeOH (2.0 mg/mL) and injections (10µL) were made with Shimadzu SIL- 163

10ADvp auto injector. The mobile phase consisted of H2O/ MeCN/HCOOH 164

(69.75:30:0.25) with a flow rate of 1.2 mL min-1. 165

166

Seedling assays 167

Cured and infected isolates of 9143 were agitated for 15 days in 200ml CDB + 5µM L- 168 tryptophan. Cultures then were vacuum-filtered (0.44 mm nylon filter) and the liquid 169 retained in sterile 50ml tubes. The pH was measured from an aliquot of each (pH 5.7 for 170

9143-infected; pH 4.2 for 9143-cured; pH 7.1 for CDB alone; Denver Instruments pH 171 meter) and the solutions brought to pH 7.0-7.1, as needed, by amendment with 0.5M 172

NaOH. Approximately 25ml of this filtrate was further filter-sterilized using 0.2mm 173 syringe filters into a sterile container and used in seedling assays. 174

Tomato seeds (ACE 55; The Home Depot®) were surface-sterilized by agitating 175 in 50% bleach for 12-15 minutes, rinsed in sterile water, and placed on sterile filter paper 176 in sterile 60mm Petri dishes (ca. 20 seeds/dish). Three milliliters of sterile water was 177 added to each dish before sealing with Parafilm®. Seeds were allowed to germinate at 178

25°C for 5 days. Sets of 10 apparently healthy seedlings were chosen haphazardly and 179 transferred under sterile conditions to new 60mm plates containing sterile filter paper. 180

Each set of seedlings was watered with 3.5ml of sterile water. Four plates/treatment (40 181 seedlings/treatment) then received 50µl per seedling of one of five treatments: (a) CDB 182 from 9143-infected; (b) CDB from 9143-cured; (c) CDB + 5µM L-tryptophan; (d) CDB 183

+ IAA (0.1mg/ml; pH 7.0); or (e) sterile water. 184

110

Plates were sealed with Parafilm® and incubated under fluorescent lights at room 185 temperature for 5 d (12 h light/dark cycle). Seedlings then were harvested and stretched 186 to full length by mounting to paper with transparent cellophane tape. Shoot and root 187 lengths to longest points were measured for each seedling using calipers. Data were 188 analyzed in JMP® 8.0 using ANOVA after normalizing all measurements to the results of 189 the CDB + IAA treatment (treatment d). 190

191

Isolation of the endohyphal bacterium 192

Incubating the isolate 9143 on 2% MEA at 36°C for 7 d (Appendix 1) resulted in 193 emergence of bacterial growth from the apparently axenic endophyte culture. We 194 extracted DNA from the bacterium following Hoffman and Arnold (2010) and sequenced 195 an 1100bp portion of the 16SrDNA region of using primers 27F /1429R (Lane 1991). 196

The sequence of this bacterial isolate (BAC182) was identical to that of the bacterium 197 sequenced directly from endophyte 9143. BAC182 was grown overnight in sterile LB 198 broth and vouchered in sterile glycerol at -80°C. Using the above methods we found no 199 evidence of IAA production by axenic cultures of this bacterium (data not shown). 200

201

Results 202

Phylogenetic analyses were used in conjunction with morphology to positively identify 203 endophyte 9143 as Pestalotiopsis neglecta (Fig. 1, Appendix 1). Multisetulate conidia are 204 fusiform shaped concolorous cells, lacking knobbed appendages. Phylogenetic analysis 205 placed this fungal isolate as sister to one of three identified P. neglecta strains from 206

Watanabe et al. (2010). Our results are congruent with their analysis, placing two other 207 putative “P. neglecta” sequences in a separate subclade. Reclassification of the 208

111

taxonomic placement of Pestalotiopsis spp. has been recommended (see Jeewon et al., 209

2004; Hawksworth 2005; Wei et al., 2007), with greater emphasis placed on 210 morphological characteristics combined with rigorous phylogenetic methods. 211

Phylogenetic analyses of 16S rDNA confirmed placement of fungal endophyte 212

9143 endohyphal bacterium in the Xanthomonadaceae, with strong support as a member 213 of the genus Luteibacter (Luteibacter sp.; Fig. 2). 214

Chromogenic testing indicated that isolate 9143 produced an indole compound in 215 vitro, with enhanced production in the presence of the endohyphal bacterium (Fig. 3). At 216 the end of the 14 d cultivation period, mean concentration of the indole compound in 217 broth from 9143-infected (104.8µg/ml) was 78µg/ml greater than that produced by 9143- 218 cured (Fig. 3). Strong differences in indole compound concentration were observed 219 throughout the 14 d cultivation period (repeated measures ANOVA, F1, 4 = 358.7; p < 220

0.0001; Fig. 3). 221

TLC and HPLC identified the compound as indole-3-acetic acid (IAA) and 222 confirmed that endophyte 9143 produced significantly more IAA when the endohyphal 223 bacterium was present vs. absent (Fig. 4). No IAA was produced when 9143-infected and 224 cured isolates were grown in CDB without tryptophan, suggesting a tryptophan- 225 dependent pathway for IAA production (data not shown). 226

Growth of tomato seedlings, measured as total seedling length, shoot length, root 227 length, and root:shoot ratio, differed significantly as a function of treatment (Fig. 5). 228

Seedlings treated with filtrate from 9143-cured did not differ in any of these measures 229 relative to the CDB control (Fig. 5, post-hoc Tukey-Kramer test, alpha = 0.05). Seedling 230

112

length was significantly greater following treatment with 9143-infected, reflecting a 231 significant enhancement of root elongation (Fig. 5, Tukey-Kramer tests, alpha = 0.05). 232

233

Discussion 234

Endohyphal bacteria have been found previously in living hyphae of plant-associated 235

Glomeromycota, Mucoromycotina, and several ectomycorrhizal Dikarya (Tuber borchii; 236

Ascomycota; Laccaria bicolor and Piriformospora indica; Basidiomycota) (Bianciotto et 237 al., 1996; Barbieri et al., 2000; Bertaux et al., 2003; Levy et al., 2003; Partida-Martinez 238 and Hertweck, 2005). Recently, we reported their presence in foliar endophytic fungi 239 representing four classes of Pezizomycotina, and demonstrated that they are both 240 geographically widespread and phylogenetically diverse (Hoffman and Arnold, 2010). 241

The present study is the first to (1) definitively identify the partners in a close ecological 242 relationship between a foliar endophyte and an endohyphal bacterium, (2) show that a 243 species of Pestalotiopsis (P. neglecta) is capable of producing a phytohormone, (3) 244 demonstrate that a filamentous foliar endophyte can produce indole-3-acetic-acid (IAA), 245

(4) show that IAA produced by this endophyte is dependent on the L-tryptophan 246 pathway, and (5) demonstrate that an endohyphal bacterium (Luteibacter sp.) enhances 247 this IAA production, with exogenous culture filtrate of the endophyte with its endohyphal 248 symbiont yielding significantly enhanced growth in roots of a model plant. Together, our 249 results reveal that products of an endophyte from a coniferous host has the potential to 250 strongly influence growth of a distantly related plant (see also Rodriguez et al., 2009), 251 demonstrate the previously unknown role of an endohyphal bacterium in altering 252 phytohormone production by a plant-symbiotic fungus, and provide a new lens with 253

113

which to view the considerable genotypic, species-level, and functional diversity of foliar 254 endophytic fungi. 255

In the presence of endohyphal Luteibacter sp. and L-tryptophan, P. neglecta 256 cultivated in CDB produced ca. 100µg/ml of IAA. This amount is likely biologically 257 significant in the host, as demonstrated both by our seedling assays and by previous 258 studies with plant-pathogenic and rhizosphere fungi. IAA production observed here was 259 within the range produced by Ustilago maydis mutant strains (75-262µg/ml) as measured 260 using the same colorimetric reagents (Sosa-Morales et al., 1997). Using more precise 261 methods (GC-MS and HPLC-ESI_MS/MS), culture filtrate from Piriformospora indica 262 measured ca. 0.16µmol of IAA after 4.5 weeks of growth (Sirrenberg et al., 2007), 263 exceeding that demonstrated here. Isolates of Colletotrichum were found to generate 2- 264

32µg/ml of IAA using TLC and GC-MS (Robinson et al., 1998; Chagué et al., 2009). 265

Although our study does not yet address bacterial-fungal-plant interactions during 266 the endophyte symbiosis per se, it provides a first estimation of one way in which an 267 apparently facultative bacterial endosymbiosis can influence interactions between an 268 endophytic fungus and its host. Further experiments involving the establishment of this 269 endophytic fungus within host tissues, and exploration of relative benefits for hosts more 270 closely related to that from which the fungus was first isolated, represent important future 271 directions. The strong response we observed in root growth as a function of filtrate from 272 cultures containing both the endophyte and its endohyphal symbionts suggests that we 273 should evaluate the potential for this foliar endophyte to colonize root tissues, and/or 274 examine how fungi localized in foliage might influence growth in other tissues. We also 275 are interested to discover whether IAA production by the endophyte or 276

114

endophyte/endohyphal bacteria complex camouflages the fungus within the host, making 277 it less detectable by plants and thus enhancing its ability to grow without inducing a 278 hypersensitive response. 279

Recent work has shown that even closely related or conspecific fungi can harbor 280 different endohyphal bacteria (Hoffman and Arnold, 2010). These facultative 281 associations are by definition flexible and appear to be gained and lost readily (Hoffman 282 and Arnold, 2010). Our study provides the first evidence that the differences in bacterial 283 associates of endophytes might play a key role in determining the nature of endophyte- 284 host associations, and may account for some of the diversity and ecological plasticity 285 observed in endophyte-plant interactions (Arnold et al., 2009). More generally, 286 increasing but still limited exploration of ectohyphal bacteria, endohyphal bacteria, and 287 mycoviruses have begun to illustrate the powerful but often overlooked ways in which 288 microbes associated with fungal hyphae can influence the outcome of plant-fungus 289 interactions (Marquez et al., 2007, Bonfante and Anca, 2009; Kobayashi and Crouch, 290

2009). Advancing our knowledge of endophyte interactions with plant hosts, other 291 extrinsic microorganisms, and most recently, diverse endohyphal bacteria, will help 292 define the functional biology of these diverse and ubiquitous symbionts. 293

294 295

115

Acknowledgments 295

We thank the Division of Plant Pathology and Microbiology, the Department of Plant 296

Sciences, the College of Agriculture and Life Sciences at the University of Arizona. 297

Additional support from the National Science Foundation (NSF-0626520 and 0702825 to 298

AEA; NSF-IGERT to MTH) is gratefully acknowledged. We thank D.R. Maddison for 299 sharing pre-release versions of Mesquite and Chromaseq; R. Ryan, M. J. Epps, J. U’Ren, 300

M. del Olmo Ruiz and A. Woodard for lab assistance and helpful discussion. We are 301 especially grateful to H. VanEtten for access to chemistry facilities and J. L. Bronstein, L. 302

S. Pierson, III, and M. J. Orbach for comments on the manuscript. This paper represents a 303 portion of the doctoral dissertation research of MTH in Plant Pathology and 304

Microbiology at the University of Arizona. 305

306

References 307

Arnold, A.E., Maynard, Z., Gilbert, G., Coley, P.D. and Kursar, T. A. (2000). Are 308 tropical fungal endophytes hyperdiverse? Ecology Letters 3: 267-274. 309 310 Arnold, A.E., Mejía, L., Kyllo, D., Rojas, E., Maynard, Z. and Herre, E. A. (2003). 311 Fungal endophytes limit pathogen damage in a tropical tree. Proceedings of the National 312 Academy of Sciences USA 100: 15649-15654. 313 314 Arnold, A.E. and Engelbrecht, B.M.J. (2007). Fungal endophytes nearly double minimum 315 leaf conductance in seedlings of a tropical tree. Journal of Tropical Ecology 23: 369-372. 316 317 Arnold, A.E. (2007). Understanding the diversity of foliar fungal endophytes: progress, 318 challenges, and frontiers. Fungal Biology Reviews 21: 51-66 319 320 Arnold, A.E. and Lutzoni, F. (2007). Diversity and host range of foliar fungal 321 endophytes: Are tropical leaves biodiversity hotspots? Ecology 88: 541-549. 322 323 Arnold, A.E., Miadlikowska, J., Higgins, K.L., Sarvate, S.D., Gugger, P., Way, A., 324 Hofstetter, V., Kauff, F., and Lutzoni, F. (2009). Hyperdiverse fungal endophytes and 325 endolichenic fungi elucidate the evolution of major ecological modes in the Ascomycota. 326 Systematic Biology 58: 283-297. 327

116

328 Artursson, V., Finlay, R. D. and Jansson, J. K. (2005). Combined bromodeoxyuridine 329 immunocapture and terminal-restriction fragment length polymorphism analysis 330 highlights differences in the active soil bacterial metagenome due to Glomus mosseae 331 inoculation or plant species. Environ. Microbiol. 7:1952–1966. 332 333 Bajo, J., Santamaría, O. and Diez, J. J. (2008). Cultural characteristics and pathogenicity 334 of Pestalotiopsis funerea on Cupressus arizonica. Forest Pathology 38:263-274. 335 336 Barash, I. and Manulis-Sasson, S. (2009). Recent evolution of bacterial pathogens: The 337 gall-forming Pantoea agglomerans case. Annual Review of Phytopathology 47:133-152. 338 339 Barbieri. E., Potenza, L., Rossi, I., Sisti, D., Giomaro, G., Rossetti, S., Beimfohr, C., and 340 Stocchi, V. (2000). Phylogenetic characterization and in situ detection of a Cytophaga- 341 Flexibacter-Bacteroides phylogroup bacterium in Tuber borchii Vittad. ectomycorrhizal 342 mycelium. Appl. Environ. Microbiol. 66:5035–5042. 343 344 Barbieri, E., C. Guidi, J. Bertaux, P. Frey-Klett, J. Garbaye, P. Ceccaroli, R. Saltarelli, A. 345 Zambonelli and V. Srocchi (2007). Occurrence and diversity of bacterial communities in 346 Tuber magnatum during truffle maturation. Environ. Microbiol. 9(9):2234–2246. 347 348 Bertaux J., Schmid, M., Prevost-Boure, N. C., Churin, J. L., Hartmann, A., Garbaye, J. 349 and Frey-Klett, P. (2003). In situ identification of intracellular bacteria related to 350 Paenibacillus spp. in the mycelium of the ectomycorrhizal fungus Laccarria bicolor 351 S238N. Appl. Environ. Microbiol. 69:4243–4248. 352 353 Bianciotto, V., Bandi, C.,Minerdi, D., Sironi, M., Tichy, H. V. and Bonfante, P. (1996). 354 An obligately endosymbiont mycorrhizal fungus itself harbors obligately intracellular 355 bacteria. Appl. Environ. Microbiol. 62:3005–3010. 356 357 Bianciotto V, Lumini, E., Lanfranco, L., Minerdi, D., Bonfante, P. and Perotto, S. (2000). 358 Detection and identification of bacterial endosymbionts in arbuscular mycorrhizal fungi 359 belonging to the family Gigasporaceae. Appl. Environ. Microbiol. 66:4503–4509. 360 361 Bonfante, P. and Anca, I-A. (2009). Plants, mycorrhizal fungi, and bacteria: A network of 362 interactions. Annu. Rev. Microbiol. 63:363–383. 363 364 Cannon, P.F. and Simmons, C.M. (2002). Diversity and host preference of leaf 365 endophytic fungi in the Iwokrama Forest Reserve, Guyana. Mycologia 94: 210-220. 366 367 Carroll G. and Petrini, O. (1983). Patterns of substrate utilization by some fungal 368 endophytes from coniferous foliage. Mycologia 75(1): 53-63. 369 370 Chagué, V., Maor, R. and Sharon, A. (2009). CgOptI, a putative oligopeptide transporter 371 from Colletotrichum gloeosporioides that is involved in responses to auxin and 372 pathogenicity. BMC Micro. 9:173-185. 373

117

374 Costacurta, A. and Vanderleyden, J. (1995). Synthesis of phytohormones by plant- 375 associated bacteria. Critical Rev. Microbiol. 21:1-18. 376 377 Ehmann, A. (1977). The van Urk-Salkowski reagent - A sensitive and specific 378 chromogenic reagent for silica gel thin-layer chromatographic detection and 379 identification of indole derivatives. J. Chromatography. 132:267-276. 380 381 Gardes, M. and Bruns, T. D. (1993). ITS primers with enhanced specificity of 382 basidiomycetes: application to the identification of mycorrhizae and rusts. Mol. Ecol. 383 2:113–118. 384 385 Glick, B. R. (1995) The enhancement of plant growth by freeliving bacteria. Can J 386 Microbiol 41:109–117. 387 388 Glickmann, E. and Y. Dessaux. (1995). A critical examination of the specificity of the 389 Salkowski reagent for indolic compounds produced by phytopathogenic bacteria. Appl. 390 Environ. Microbiol. 61(2):793-796. 391 392 Gruen, H. E. (1959). Auxins & fungi. Ann. Rev. Plant Physiol. 10: 405-440. 393 394 Hawksworth, D. L. (2005). Host specificity in Pestalotiopsis. Mycological Research. 395 109:5. 396 397 Higgins, K. L., Arnold, A. E., Miadlikowska, J., Sarvate, S. D., and Lutzoni, F. (2007). 398 Phylogenetic relationships, host affinity, and geographic structure of boreal and arctic 399 endophytes from three major plant lineages. Molecular Phylogenetics and Evolution 42: 400 543-555. 401 402 Hoffman, M. T. and Arnold, A. E. (2008). Geographic locality and host identity shape 403 fungal endophyte communities in cupressaceous trees. Myco. Res. 112:331–344. 404 405 Hoffman, M. T. and Arnold, A. E. (2010). Diverse bacteria inhabit living hyphae of 406 phylogenetically diverse fungal endophytes. Appl. Environ. Microbiol. 407 408 Hoeksema, J. D., Chaudhary, V. B., Gehring, C. A., et al., (2010). A meta-analysis of 409 context-dependency in plant response to inoculation with mycorrhizal fungi. Ecology 410 Letters 13:394-407. 411 412 Huelsenbeck, J. P. and F. Ronquist (2001). MRBAYES: Bayesian inference of 413 phylogeny. Bioinformatics 17:754-755. 414 415 Jeewon, R., Liew, E.C.Y., Simpson, J. A., Hodgkiss, I. J. and Hyde, K.D. (2003). 416 Phylogenetic significance of morphological characters in the taxonomy of Pestalotiopsis 417 species. Molecular Phylogenetics and Evolution 27:372-383. 418 419

118

Jeewon, R., Liew, E.C.Y., and Hyde, K.D. (2004) Phylogenetic evaluation of species 420 nomenclature of Pestalotiopsis in relation to host association. Fungal Diversity 17:39-55. 421 422 Kimura, Y., Kouge, A., Nakamura, K., Koshino, H., Uzawa, J., Fujoika, S. and Kawano, 423 T. (1998). Pesthetoxin, a new phytotoxin produced by the gray blight fungus, 424 Pestalotiopsis theae. Biosci. Biotechnol. Biochem. 62:1624-1626. 425 426 Koga, J. (1995) Structure and function of indolepyruvate decarboxylase, a key enzyme in 427 indole-3-acetic acid biosynthesis. Biochim Biophys Acta 1249:1–13 428 429 Kobayashi, D. Y. and J. A. Crouch. 2009. Bacterial/fungal interactions: From pathogens 430 to mutualistic endosymbionts. Annu. Rev. Phytopathol. 47:63–82. 431 432 Kumar, D.S.S. and Hyde, K.D. (2004). Biodiversity and tissue-recurrence of endophytic 433 fungi in Tripterygium wilfordii. Fungal Diversity 17: 69-90. 434 435 Kumaresan, V. and Suryanarayanan, T.S. (2000). Fungal endophytes of the mangrove 436 Rhizophora apiculata. In: Abstracts of Asian Mycological Congress 2000 (eds. K.D. 437 Hyde, S.R. Lai, S.B. Pointing and W.S.W. Wong). Hong Kong SAR, China: 36. 438 439 Lane, D. J. 1991. 16S/23S rRNA sequencing. (eds. Stackebrandt and Goodfellow in: 440 Nucleic Acid Techniques in Bacterial Systematics. John Wiley and Sons, Inc., New 441 York, N.Y. 442 443 Levy, A., Chang, B. J., Abbott, L. K., Kuo, J., Harnett, G. and Inglis, T. J. J. (2003). 444 Invasion of spores of the arbuscular mycorrhizal fungus Gigaspora decipiens by 445 Burkholderia spp. Appl. Environ. Microbiol. 69:6250–6256. 446 447 Lindow, S.E., Desurmont, C., Elkins, R., McGourty, G., Clark, E. and Brandl, M.T. 448 (1998). Occurrence of indole-3-acetic acid-producing bacteria on pear trees and their 449 association with fruit russet. Phytopathology 88:1149-1157. 450 451 Liu, A.R., Wu, X.P., Xu, T., Guo, L.D. and Wei, J.G. (2006). Notes on endophytic 452 Pestalotiopsis from Hainan, China. Mycosystema 25: 389-397. 453 454 Lodge D. J., Fisher, P. J., and Sutton, B. C. (1996). Endophytic fungi of Manilkara 455 bidentata leaves in Puerto Rico. Mycologia 88(5): 733-738. 456 457 Maddison, D. R. and W. P. Maddison (2005). ChromaSeq Module. Mesquite: a modular 458 system for evolutionary analysis. V1.06. http://mesquiteproject.org. 459 460 Maddison, W. P. and D.R. Maddison (2009). Mesquite: a modular system for 461 evolutionary analysis. V2.6, http://mesquiteproject.org. 462 463

119

Marquez, L. M., Redman, R. S., Rodriguez, R. J., and Roossinck, M. J. (2007). A virus in 464 a fungus in a plant: Three-way symbiosis required for thermal tolerance. Science 465 315:513-515. 466 467 Maor, R., Haskin, S., Levi-Kedmi, H., and Sharon, A. (2004). In planta production of 468 indole-3-acetic acid by Colletotrichum gloeosporioides f. sp. aeschynomene. AEM 469 70(3):1852-1854. 470 471 Miller, M. A., Holder, M. T., Vos, R., Midford, P. E., Liebowitz,T., Chan, L., Hoover, P., 472 Warnow, T. The CIPRES Portals CIPRES. 2009-08-04. 473 474 Nassar, A.H., El-Tarabily, K. A., and Sivasithamparam, K. (2005). Promotion of plant 475 growth by an auxin-producing isolate of the yeast Williopsis saturnus endophytic in 476 maize (Zea mays L.) roots. Biology and Fertility of Soils 42: 97–108. 477 478 Okane, I., Nagagiri, A. and Ito, T. (1997). Preliminary study of endophytic fungi in 479 evergreen plants from Ishigaki and Iriomote Islands. The Institute for Fermentation, 480 Osaka Research Communications 18: 45-51. 481 482 Okane, I., Nagagiri, A. and Ito, T. (1998). Endophytic fungi in leaves of ericaceous 483 plants. Canadian Journal of Botany 76: 657-663. 484 485 Petrini, O. and Carroll, G. C. (1981). Endophytic fungi in the foliage of some 486 Cupressaceae in Oregon. Canadian Journal of Botany 59:629-636. 487 488 Photita, W., Taylor, P.W.J., Ford, R., Hyde, K.D. and Lumyong, S. (2005). 489 Morphological andmolecular characterization of Colletotrichum species from herbaceous 490 plants in Thailand. Fungal Diversity 18: 117-133. 491 492 Posada, D. and Crandall, K. A. (1998). MODELTEST: Testing the model of DNA 493 substitution. Bioinformatics. 14:817–818. 494 495 Partida-Martinez, L. P., and Hertweck, C. (2005). Pathogenic fungus harbours 496 endosymbiotic bacteria for toxin production. Nature 437:884–888. 497 498 Prusty, R., Grisafi, P. and Fink, G. R. (2004). The plant hormone indoleacetic acid 499 induces invasive growth in Saccharomyces cerevisiae. PNAS 101(12):4153-4157. 500 501 Reineke, G., Heinze, B., Schirawski, J., Buettner, H., Kahmann, R. and Basse, C. W. 502 (2008). Indole-3-acetic acid (IAA) biosynthesis in the smut fungus Ustilago maydis 503 and its relevance for increased IAA levels in infected tissue and host tumour formation. 504 Mol. Plant Pathology 9(3):339-355. 505 506 Robinson, M., Riov, J., and Sharon, A. (1998). Indole-3-acetic acid biosynthesis in 507 Colletotrichum gloeosporioides f. sp. aeschynomene. Appl. Environ. Microbiol. 508 64(12):5030-5032. 509

120

510 Salkowski, E. 1885. Ueber das Verhalten der Skatolcarbonsa¨ure im Organismus. 511 Z. Physiol. Chem. 9:23–33. 512 513 Scagel, C.F. and Linderman, R.G. (1998). Relationships between differential in vitro 514 indole-acetic acid or ethylene production capacity by ectomycorrhizal fungi and conifer 515 seedling responses in symbiosis. Symbiosis 24:13-34. 516 517 Sirrenberg, A. Göbel, C., Grond, S., Czempinski, N., Ratzinger, A., Karlovsky, P., 518 Santos, P., Feussner, I. and Pawlowski, K. (2007). Piriformospora indica affects plant 519 growth by auxin. Physiologia Plantarum 131:581-589. 520 521 Spaepen, S., Vanderleyden, J., and Remans, R. (2007). Indole-3-acetic acid in microbial 522 and microorganism-plant signaling. FEM Microbiology Reviews 31(4):425-448. 523 524 Sosa-Morales, M.E., Guevara-Lara, F., Martínez-Juárez, V.M., and Paredes-López, O. 525 (1997). Production of indole-3-acetic acid by mutant strains of Ustilago maydis (maize 526 smut/ huitlacoche). Appl. Microbiol. Biotechnol. 48:726-729. 527 528 Strack, D., Fester, T., Hause, B., Schliemann, W. and Walter, M. H. (2003). Arbuscular 529 mycorrhiza: Biological, chemical, and molecular aspects. Journ. of Chemical Ecology 530 29(9):1955. 531 532 Strobel, G.A., Hess, W.M., Li, J.Y., Ford, E., Sidhu, R.S. and Summerell, B. (1997). 533 Pestalotiopsis guepinii, a taxol-producing endophyte of the wollemi pine, Wollemia 534 nobilis. Australian Journal of Botany 45: 1073-1082. 535 536 Strobel, G.A., Li, J.Y., Ford, E., Worapong, J., Gary, I.B. and Hess, W.M. (2000). 537 Pestalotiopsis jesteri sp. nov. an endophyte from Fragraea bodenii Wernh., a common 538 plant in the southern highlands of Papua New Guinea. Mycotaxon 76: 257-266. 539 540 Strobel, G.A., Yang, X.S., Sears, J., Kramer, R., Sidhu, R.S. and Hess, W.M. (1996). 541 Taxolfrom Pestalotiopsis microspora, an endophytic fungus of Taxus wallichiana. 542 Microbiology 142: 435-440. 543 544 Suryanarayanan, T.S., Kumaresan, V. and Johnson, J.A. (1998). Foliar fungal endophytes 545 from two species of the mangrove Rhizophora. Canadian Journal of Microbiology 44: 546 1003-1006. 547 548 Swofford, D. L. 2002. PAUP*: Phylogenetic Analysis Using Parsimony (* and other 549 methods). Ver. 4b10 Sinauer Assoc., Sunderland, MA. 550 551 Takeda, Y. (2003). Phenotypes and Genotypes Related to Tea Gray Blight Disease 552 Resistance in the Genetic Resources of Tea in Japan. JARQ 37:31-35. 553 554 Toofanee, B.S. and Dulymamode, R. (2002). Fungal endophytes associated with 555

121

Cordemoya integrifolia. Fungal Diversity 11: 169-175. 556 557 Tudzynski, B. and Sharon, A. (2002). Biosynthesis, biological role and application of 558 fungal phytohormones. The Mycota X. Industrial Applications. Springer-Verlag Berlin 559 Heidelberg. 560 561 van der Lelie, D., Taghavi, S., Monchy, S., Schwender, J., Miller, L., Ferrieri, R., Rogers, 562 A., Wu, X., Zhu, W., Weyens, N., Vangronsveld, J. and Newman, L. (2009). Poplar and 563 its bacterial endophytes: Coexistence and harmony. Critical Reviews in Plant Sciences, 564 28:346-358. 565 566 Vilgalys, R. and M. Hester (1990). Rapid genetic identification and mapping of 567 enzymatically amplified ribosomal DNA from several Cryptococcus species. J. Bac. 568 172:4238–4246. 569 570 Wang, Y., Guo, L.D. and Hyde, K.D. (2005). Taxonomic placement of sterile 571 morphotypes of endophytic fungi from Pinus tabulaeformis (Pinaceae) in northeast China 572 based on rDNA sequences. Fungal Diversity 20: 235-260. 573 574 Wang, Y. and Guo, L.D. (2004). Endophytic fungi II. new records from pine in China. 575 Mycosystema 23: 24-27. 576 577 Wang, Z.W., Chen, W.P., Wei, G. and Chen, Y.X. (2002). Eight new recombination 578 species of Pestalotiopsis in China. Journal of Guangxi Agricultural and Biological 579 Science 22:151-156. 580 581 Watanabe, K., Motohashi, K., and Ono, Y. (2010). Description of Pestalotiopsos 582 pallidotheae: a new species from Japan. Mycoscience 583 584 Wei, J.G. and Xu, T. (2003a). Biodiversity of endophytic fungi Pestalotiopsis. 585 Biodiversity Science 11: 162-168. 586 587 Wei, J.G. and Xu, T. (2003b). Pestalotiopsis karstenii, a new record of endophytic fungi 588 from Camellia sasanqua in China. Mycosystema 22: 666-668. 589 590 Wei, J.G. and Xu, T. (2004). Pestalotiopsis kunmingensis sp. nov., an endophyte from 591 Podocarpus macrophyllus. Fungal Diversity 15: 247-254. 592 593 Wei, J.G., Xu, T., Guo, L.D. and Pan, X.H. (2005a). Endophytic Pestalotiopsis species 594 from southern China. Mycosystema 24: 481-493. 595 596 Wei, J.G., Xu, T., Guo, L.D., Liu, A.R., Pan, X.H., Zhang, J.C. and Yuan, G.Q. (2005b). 597 Delimitation of Pestalotiopsis species based on morphological and molecular 598 phylogenetic characters. Journal of Guangxi Agricultural and Biological Science 24: 599 304-313. 600 601

122

Wei, J.G., Xu, T., Guo, L.D., Liu, A.R. Zhang, Y. and Pan, X.H. (2007). Endophytic 602 Pestalotiopsis species associated with plants of Podocarpaceae, Theaceae and Taxaceae 603 in southern China. Fungal Diversity 24: 55-74. 604 605 White, T. J., Bruns, T., Lee, S. and Taylor, J. (1990). Amplification and direct sequencing 606 of fungal ribosomal RNA genes for phylogenetics. In: Innis, M. A., Gelfand, D. H., 607 Sninsky, J. J. & White, T. J. (Eds.) PCR protocols: A guide to methods and applications. 608 Academic Press, San Diego, CA. 609 610 Worapong, J., Intharaungdorn, S., Strobel, G.A., and Hess, W.M. (2003). A new record 611 of Pestalotiopsis theae, existing as an endophyte on Cinnamomum iners in Thailand. 612 Mycotaxon 88: 365–372. 613 614 Xin, G., Glawe, D. and Doty, S. L. (2009). Characterization of three endophytic, indole- 615 3-acetic acid-producing yeasts occurring in Populus trees. Myco. Research. 113:973-980. 616 617 Yang, S., Zhang, Q., Guo, J., Charkowski, A. O., Glick, B. R., Ibekwe, A. M., Cooksey, 618 D. A. and Yang, C-H. (2006). Global effect of indole-3-acetic acid biosynthesis on 619 multiple virulence factors of Erwinia chrysanthemi 3937. Applied and Environmental 620 Microbiology 73:1079-1088. 621 622 Zwickl, D. J. (2006). Genetic algorithm approaches for the phylogenetic analysis of large 623 biological sequence datasets under the maximum likelihood criterion. Ph.D. dissertation, 624 The University of Texas at Austin. 625 626 627

123

Figure legends 627

628

Figure 1. Majority rule consensus tree based on Bayesian analyses of 60 ITSrDNA 629 sequences representing Pestalotiopsis (with Seiridium as outgroup). Support value before 630 the slash represents maximum-parsimony bootstrap (>70%); number after the slash 631 represents Bayesian branch support > 90%. 632

633

Figure 2. Majority rule consensus tree (using P. syringae as outgroup) based on Bayesian 634 analyses of 37 proteobacterial 16S rDNA sequences with affinity for Luteibacter spp. in 635

BLAST analyses. Support value before the slash represents maximum-parsimony 636 bootstrap >70%, number after the slash represents Bayesian branch support > 90. 637

Branches in bold indicate neighbor-joining bootstrap support >70%. 638

639

Figure 3. Spectrophotometric absorbances at 530nm following treatment of supernatant 640 from CDB cultures of 9143-infected and 9143-cured with Salkowski reagent, indicating 641 production of an indole compound (identified as IAA; Fig. 4). Each point represents the 642 average of three readings (two-tailed t4 = 20.99; p < 0.0001), converted to concentration 643 amounts using the IAA standard curve (y =0.0108x-0.0049). Error bars indicate standard 644 deviation. 645

646

Figure 4. HPLC analysis (retention time and co-injection) results for (A) IAA standard, 647 and extracts from (B) 9143-cured and (C) 9143-infected. Elution time was 12.54 minutes 648 for the IAA standard, 12.89 for 9143-cured, and 12.95 by 9143-infected sample. , IAA 649

124

comprised 16% of the total sample from 9143-cured, and 29% from 9143-infected (Fig. 650

2C) based on the area ratio under the peak. 651

652

Figure 5. Growth responses of tomato seedlings to pH-consistent supernatant from 9143- 653 infected cultures in CDB; supernatant from 9143-cured in CDB; and control treatments 654

(CDB alone, water). (A) Total seedling length, (B) root length, (C) root to shoot ratio, (D) 655 shoot length. Data were normalized to measurements for plants that received CDB + 656

0.1mg/ml IAA treatment (data not shown). Different superscripts within each panel 657 indicate statistically different means. 658

659

660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676

125

Figure 1. 676

677 678

126

Figure 2. 678

679 680

127

Figure 3. 680 681

9143+ 9143-

IAA concentration (µg/ml)

Days

128

Figure 4.

(A) Indole-3-acetic acid standard

(B) 9143-cured

129

Figure 5.

A. B. F3,156 = 30.2; p< 0.0001 F3,156 = 32.8; p< 0.0001 a a

b b b b

c c

9143+ 9143- CDB H2O 9143+ 9143- CDB H2O

Seedling length (mm)

C. D. F3,156 = 19.3; p< 0.0001 F3,156 = 3.2; p< 0.0001

a a a b a b a b b

b 9143+ 9143- CDB H2O 9143+ 9143- CDB H2O

Treatments

130

Appendix I

Pestalotiopsis spore morphology

Pestalotiopsis neglecta asexual spore morphology description coincides with the morphology of conidia from endophyte 9143 (Fig A). These spores are fusiform, four-septate, a fuliginous brown in color, with end cells hyaline. The apical end is short with two or three spreading setulae, approximately 22um long. The basal end contains a pedicel about 4-7um long

(Steyaert, 1953).

Figure A.

Conidia from 9143-uninfected culture (400X) showing fusiform cells with 4 septae.

Figure B.

Results of growth assays over 14 days for 9143-infected (black diamond) and 9143-cured (open square) on 2% MEA at pH4.5 (panel A), pH8.0 (panel B), and pH=6.8 (standard 2% MEA;

131

panel C) at 22°C. Panel D shows growth on water agar at 22°C. 9143 did not grow at 36°C.

Error bars indicate standard error of the three replicates performed for all experiments.

A. B.

C. D.