Subalpine Conifers Host Core Endophytic Bacteria Conserved Across Sites and Tree Species

UNIVERSITY OF CALIFORNIA, MERCED

Subalpine conifers host core endophytic bacteria conserved across sites and tree species

A dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy

Environmental Systems

Alyssa A. Carrell

Committee in Charge:

A. Carolin Frank, Chair Michael Beman Lara Kueppers Jason Sexton

! !

Chapter 2 © Carrell and Frank 2014

All other chapters: ©

Alyssa A. Carrell, 2014

! !

The dissertation of Alyssa Ann Carrell is approved, and it is acceptable in quality and form for publication on microfilm and electronically:

______

Michael Beman

______

Lara Kueppers

______

Jason Sexton

______

A. Carolin Frank, Chair

University of California, Merced 2014

iii! !

Table of Contents List%of%Tables...... vi! List%of%Figures ...... vii! Acknowledgements ...... x! Curriculum%Vitae ...... xii! Abstract%of%the%Dissertation...... xiv! 1! Introduction...... 1! 1.1! Organization%of%Dissertation...... 3! 1.2! References...... 4! 2! Pinus&flexilis%and%Picea&engelmannii%share%a%simple%and%consistent%needle% endophyte%microbiota%with%a%potential%role%in%nitrogen%fixation...... 7! 2.1! Abstract...... 7! 2.2! Introduction ...... 8! 2.3! Materials%and%methods ...... 9! 2.3.1! Sample!collection!and!sterilization...... 9! 2.3.2! DNA!Extraction ...... 10! 2.3.3! DNA!Amplification...... 10! 2.3.4! Sequence!Analysis ...... 11! 2.4! Results ...... 12! 2.4.1! Phylotypes!recovered!from!samples...... 12! 2.4.2! Dominant!bacterial!taxa!associated!with!P.!flexilis!and!P.!engelmannii!needles ! 12! 2.4.3! Structure!of!conifer!needle!endophyte!communities ...... 14! 2.5! Discussion ...... 14! 2.6! References...... 18! 2.7! Tables ...... 27! 2.8! Figures...... 28! 3! The%association%between%Pinaceae%and%an%acetic%acid%bacterial%needle% endophyte%is%persistent%across%host%species%and%geographic%locations ...... 34! 3.1! Abstract...... 34! 3.2! Introduction ...... 34! 3.3! Methods...... 36! 3.3.1! Sample!collection!and!sterilization...... 36! 3.3.2! DNA!extraction...... 36! 3.3.3! DNA!Amplification...... 37! 3.3.4! Phylotype!generation!and!classification...... 37! 3.3.5! Community!Analysis...... 38! 3.4! Results ...... 38! 3.4.1! Phylotypes!recovered ...... 38! 3.4.2! Dominant!bacterial!community!members...... 39! 3.4.3! Community!Diversity!and!Structure...... 39! 3.5! Discussion ...... 40!

iv! !

3.5.1! All!samples!were!heavily!dominated!by!AAB1!and!most!samples!by!AAB2...... 40! 3.5.2! Remaining!community!structure!and!composition...... 41! 3.6! References...... 42! 3.7! Tables ...... 46! 3.8! Figures...... 47! 4! Enterobacteraceae%dominate%the%bud%and%needle%endophytic%communities% of%Pinus&contorta&in%a%subalpine%meadow ...... 57! 4.1! Abstract...... 57! 4.2! Introduction ...... 57! 4.3! Methods...... 59! 4.3.1! Sample!collection!and!sterilization...... 59! 4.3.2! DNA!extraction!and!amplification ...... 59! 4.3.3! Sequence!Analysis ...... 60! 4.4! Results ...... 61! 4.5! Discussion ...... 61! 4.5.1! Enterobacter!and!Citrobacter!species,!not!Gluconacetobacter!species,!dominate! the!needles!of!P.!contorta!at!Tuolomne!Meadows,!CA ...... 61! 4.5.2! Little!overlap!in!the!major!phylotypes!in!buds!and!needles...... 62! 4.6! References...... 63! 4.7! Figures...... 67! 5! Diversity%and%Structure%of%the%Bacterial%Endophyte%Communities%in%the% Foliage%of%Giant%Trees ...... 73! 5.1! Abstract...... 73! 5.2! Introduction ...... 74! 5.3! Methods...... 75! 5.3.1! Sample!collection!and!sterilization...... 75! 5.3.2! DNA!Extraction ...... 75! 5.3.3! DNA!Amplification...... 76! 5.3.4! Phylotype!generation!and!classification...... 76! 5.3.5! Community!Analysis...... 77! 5.4! Results ...... 77! 5.4.1! Phylotypes!Recovered ...... 77! 5.4.2! Relative!Abundance!of!Bacterial!Taxa ...... 77! 5.4.3! Community!Structure ...... 78! 5.5! Discussion ...... 79! 5.6! References...... 81! 5.7! Tables ...... 87! 5.8! Figures...... 88! 6! Conclusion ...... 93! 6.1! References...... 95!

v! !

List of Tables

Table 2-1 Samples successfully characterized by 16S rRNA in this study, along with the number of sequences after sequence quality control and removal of plant DNA .... 27 Table 3-1 Student t-test (p-values) of alpha-diversity measurements...... 46 Table 3-2 Anosim and Permanova analysis of unweighted UniFrac distance matrices... 46 Table 5-1 Samples successfully characterized by 16S rRNA in this study with number of sequences after quality control and removal of plant DNA...... 87

vi! !

List of Figures

Figure 2-1 Rarefaction curves for warm edge and treeline samples. There is no apparent asymptote in the rarefaction curves, suggesting that the sequencing depth does not encompass the full extent of phylotype richness in each of the communities. However, the rarefaction curves suggest that the lower number of phylotypes recovered from treeline samples was not due to insufficient sampling. The high and low of the error bars represent one standard deviation away from the mean ...... 28 Figure 2-2 Relative abundances of various major bacterial phyla and classes recovered from P. flexilis and P. engelmannii needles. Relative abundance of phyla (and classes of the Proteobacteria) was calculated as the percentage of sequences belonging to a particular lineage of all 16S rRNA gene sequences recovered from each sample...... 29 Figure 2-3 Phylogeny of major Alphaproteobacterial sequences in our samples. Maximum likelihood phylogeny of Alphaproteobacterial sequences that occurred at least 100 times along with the three most closely related sequences from GenBank 16S rRNA database (accession number indicated). Because our sequences are short (approximately 300 nt), many of our clades have low bootstrap support. Here, only bootstrap values above 50% are displayed. The tree is rooted with Burkholderia arboris...... 30 Figure 2-4 Heatmap showing the 10 most dominant phylotypes and their average relative abundances as percentages of all sample 16S rRNA gene sequences recovered in our conifer needles samples. (A) P. flexilis, (B) P. engelmannii. Color tones range from cool (blue) to warm (red) to indicate the lowest to highest relative abundance values. Phylotypes were considered dominant if they were both highly abundant and occurred frequently in samples of a given conifer species. Abbreviations: PT=Phylotype,Acetobac=Acetobacteraceae, Acidobac=Acidobacteria/Acidobacteriaceae, Burkhold=Burkholderiaceae, Bacteriod=Bacteriodetes, Flexibacteraceae, Methylobac=Methylobacteriaceae, Cytophag=Cytophagacae, Sphingomonad=Sphingomonadaceae...... 31 Figure 2-5 Shared and host species-specific phylotypes. Blue: phylotypes found in all P. engelmannii samples but not in any of the P. flexilis samples. Red: phylotypes found in all P. flexilis samples but not in any of the P. engelmannii samples. Purple: indicates phylotypes recovered from all samples (i.e. both species). An asterisk indicates that the phylotype is included in Figure 2...... 32 Figure 2-6 PCoA and UniFrac analysis of the bacterial communities associated with conifer needles. (A-C) PCoA of the unweighted UniFrac distance matrix. Points that are closer together on the ordination have communities that are more similar. Each point corresponds to a sample, and shapes correspond to (A) host species, (B) elevation, and (C) location. (D) Hierarchical clustering of composite communities of the conifer species. Leaves are labeled by color according to host species: red, P. flexilis; blue, P. engelmannii...... 33

vii! !

Figure 3-1 Rarefaction curve of endophytic bacterial OTUs at 97% similarity, rarified to 10,000 sequences per sample. Niwot Ridge samples had fewer OTUs than Horseshoe Meadows samples...... 47 Figure 3-2 Taxonomic relative abundances of endophytic bacterial phylum in P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Relative abundances of phylum were calculated per sample...... 48 Figure 3-3 Taxonomic relative abundances of endophytic bacterial classes in P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Relative abundances were calculated by abundance of class within each sample...... 49 Figure 3-4 Log-transformed heatmap of top ten OTUs from each sample. The relative abundance (in percentage) is indicated for each OTU in each sample. Color ranges from cool (blue) to warm (red) colors to represent the intensity of domination of each OTU with highest values in red. AAB1 is indicated with an asterisk...... 50 Figure 3-5 Chao richness measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA...... 51 Figure 3-6 Phylogenetic diversity measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Horseshoe Meadows samples generally had a higher phylogenetic diversity...... 52 Figure 3-7 Shannon diversity index measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Shannon diversity was greater in Horseshoe Meadow samples...... 53 Figure 3-8 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to tree species: red, P. contorta; blue, P. flexilis. Elliipses represent 95% confidence intervals for clustering based on tree species...... 54 Figure 3-9 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to sampling site: red, Niwot Ridge, CO; blue, Horseshoe Meadows, CA. Elliipses represent 95% confidence intervals for clustering based on site...... 55 Figure 3-10 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and shape correseponds to sample site: diamond, Horseshoe Meadows, CA; triangle, Niwot Ridge, CO. Color of points and ellipses correspond with species within site. Elliipses represent 95% confidence intervals for clustering based on location and site...... 56 Figure 4-1Relative abundances of various bacterial classes based on 16S rRNA gene sequences at 97% similarity from buds and needles of P. contorta. Relative

viii! !

abundances of classes were calculated as percentage of sequences belonging to a particular class recovered from each sample...... 67 Figure 4-2 Log-transformed heatmap of OTUs with a relative abundance >1%. The relative abundance (in percentage) is indicated for each OTU in each sample. Color ranges from cool (blue) to warm (red) colors to represent the intensity of domination of each OTU with highest values in red. Relative abundances <0.01% are indicated with black text; >0.01% are indicated with white text...... 68 Figure 4-3 The number of observed endophytic bacterial species at the 97% similarity of 16S rRNA gene sequences in needles is higher than the number of observed species in buds...... 69 Figure 4-4 Chao1 richness based on 16S rRNA gene sequences of endophytic bacteria in P. contorta was higher in needles than buds...... 70 Figure 4-5 Phylogenetic distance, or the sum of branches of phylogeny in each sample, based on 16S rRNA gene sequences of endophytic bacteria in P. contorta was higher in needles than buds...... 71 Figure 4-6 PCoA of the weighted UniFrac distance matrix of needle and bud endophytic bacterical communities of P. contorta. Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to tissue type: red, bud; blue, needle...... 72 Figure 5-1 Rarefaction curves indicating the number of OTUs at 97% similarity for lower, middle, and upper heights of trees at each location. FC: Freeman Creek (giant sequoia), BC: Big Creek (coast redwood), ST: Samuel P Taylor (coast redwood). The rarefaction curves do not asymptote indicating sampling was not sufficient.... 88 Figure 5-2 A: Bar chart showing relative abundance of major bacterial phyla (and classes for Proteobacteria) from coast redwood and giant sequoia trees calculated as percentage per sample. Each bar represents a tree sample. B-E: Pie charts showing the distribution of bacterial families from the four most abundant classes. B: Alphaproteobacteria, C: Bacilli, D: Betaproteobacteria, E: Gammaproteobacteria. 89 Figure 5-3 Boxplot showing (A) Chao1 and (B) Shannon Diversity indexes of our samples at the 97% similarity and rarified to 620...... 91 Figure 5-4 PCoA of an unweighted UniFrac distance matrix. Points that are closer on the PCoA are more similar phylogenetically. Each point represents the bacterial community of a sample colored by (A) tree species, (B) location and (C) canopy height. (D) Hierarchial clustering analysis by UPGMA of all tree samples. Leaves are labeled by shape according to tree species: square, giant sequoia; circle, coast redwood...... 92

ix! !

Acknowledgements

This dissertation represents a culmination of five years of research that would not have been possible without the guidance, support, and assistance of many people.

First I would like to express my deepest gratitude to my inspiring advisor, Carolin Frank. She is one of the most intelligent, supportive, caring, and down to earth person I have met. I will cherish our discussions of science, research, and life. She instilled my confidence in obtaining a doctorate, helping me realize my research potential. I would also like to thank my PhD committee: Mike Beman, Lara Kueppers, and Jay Sexton for their guidance and feedback throughout my dissertation project and writing phases.

I would also like to thank my lab colleagues for assistance in lab and collection of samples. In particular, I would like to thank Michele Conrad for sampling the trees for my first chapter. She taught me general molecular techniques, helped formulate my scientific questions, and became a dear friend over our work together. I would also like to thank undergraduates Johnny Dhaliwal and Ashley Graham that helped significantly with grinding pine needles. I especially would like to thank my dear friend Dana Carper for assistance with my last two projects. PCR, extractions, gels, and RuPaul helped create a bond that will never be broken.

I would like to thank all of my friends for their support during my time at graduate school. Chelsea Carey, thank you for being my soundboard for science and life problems. I am so glad our initial encounter wasn’t our last. Kristynn Sullivan, I appreciate that she listened to my science issues and pretended to understand and care. My office neighbor, Katie Amrine has always provided emotional support when I needed it and I am thankful to have her as a friend. I am also grateful to Travis Lawrence for being a great friend and for allowing me to steal his wife for lab work. In addition to the friends listed above, I have many friends to thank for helping me throughout my PhD journey: Christy and the entire Copeland Family, Yesenia, Faust, the Minter family, the McGuffins, and many more.

Finally, I would like to thank my family: Mom, Dad, Dave, Loren, Casey, and Hannah. Their support throughout my life has brought me to this point. My love for science has been fostered since I was child by their devotion to my education and interests. My husband, Stephen Minter, has always provided the encouragement to complete my doctorate. The past few years have been difficult with distance separation but I always knew I had his support and love. Also, I must also acknowledge his help with sampling and sterilization of samples. Thank you.

x! !

Please note, Chapter 2, in full, is a reprint of the material as it appears in Frontiers of Microbiology 2014. Carrell A,A., and A.C. Frank. The dissertation author was the primary investigator and author of this paper.

And thank you, dear reader, because if you are reading this line, you read at least one page of my dissertation.

xi! !

Curriculum Vitae

EDUCATION

Doctor of Philosophy in Environmental Systems, 2009-2014 University of California, Merced

Bachelors of Science, Biology, 2005-2009 University of California, Merced

PUBLICATIONS

Carrell, A. A., and A. C. Frank. 2014. Pinus flexilis and Piceae engelmannii share a simple and consistent needle endophyte microbiota with a potential role in nitrogen fixation. Frontiers in Microbiology.

Carrell, A. A., and A.C. Frank, 2014. Under Review. Diversity and Structure of the Bacterial Endophyte Communities in the Foliage of Giant Trees.

Carrell,. A. A., D.L. Carper, and A.C. Frank, 2014. In Prep. The association between Pinaceae and an acetic acid bacterial needle endophyte is persistent across host species and geographic locations.

PRESENTATIONS

Contributed Talk, Potential for aboveground endophytic N2 fixation in subalpine conifers revealed by 16S rRNA gene analysis, Yosemite Symbiosis Workshop, Wawona, CA, May 2014. Invited Talk, The Diversity of Beneficial Bacteria in Forest Trees, University of California, Merced, Genome Biology (BIO 142), February 2013 Poster Presentation, Diversity and Distribution of Endophytic Bacteria in Pinus flexilis Foliage, Ecological Society of America, Portland, OR. August 2012. Poster Presentation, Diversity and Distribution of Endophytic Bacteria in Pinus flexilis Foliage, American Society of Microbiology, San Francisco, CA. June 2012. Poster Presentation, Diversity of Endophytic Bacteria in Pinus flexilis Foliage, New Phytologist Symposium, Rhodes, Greece. May 2012. Poster Presentation, Diversity of Endophytic Bacteria in Pinus flexilis Foliage, UC Merced Research Week, Merced, CA April 2012. Invited Talk, Diversity of Conifer Endophytic Bacteria, California State University Stanislaus Biology Colloquium, Turlock, CA, March 2012. Contributed Talk, Endophytic Symbionts of Pinus flexilis Seedlings of Different

xii! !

Elevations, Subalpine and Alpine Species Range Shifts with Climate Change Annual Meeting, Berkeley, CA, November 2011. Poster Presentation, Diversity of Pinus flexilis Endophytic Bacteria, Joint Genome Institute Microbial Genomics and Metagenomics, Walnut Creek, CA, September 2011. Poster Presentation, Microbial Symbionts of Coniferous Trees, Symbiosis Workshop, Yosemite, CA, April 2011. Poster Presentation, Microbial Symbionts of Coniferous Trees, UC Merced Research Week, Merced, CA, April 2011. Contributed Talk, Developing Methods to Study Microbial Symbionts of Conifers, Subalpine and Alpine Species Range Shifts with Climate Change Annual Meeting, Berkeley, CA, October 2010. Invited Talk, What to Expect in Graduate School, Yosemite REU Summer Research Program, YNP, CA, June 2010.

AWARDS

UC Merced Bobcat Fellowship Award, $7,500, Spring 2014 UC Merced ES Summer Graduate Fellowship, $10,000, Summer 2013 UC Merced GRC Summer Fellowship, $7,500, Summer 2012 New Phytologist Symposium Travel Grant, $1,200, May 2012 UC Merced Graduate Division General Fellowship, $3855, Spring 2012 UC Merced Faculty Mentor Program Fellowship, $34,120, Fall 2011-Spring 2012 UC Merced GRC Spring Fellowship, $7,500, Spring 2011 NSF Graduate Research Fellowship Program, Honorable Mention, Fall 2011 UC Merced Graduate USAP Award, $10,000, Summer 2010

xiii! !

Abstract of the Dissertation

Subalpine conifers host core endophytic bacteria conserved across sites and tree species

Alyssa A. Carrell

Doctor of Philosophy, Environmental Systems Program

University of California, Merced, 2014

Dr. A. Carolin Frank, Chair

The overall aim of my dissertation was to determine if subalpine conifers host a consistent core of endophytic bacteria. Endophytic bacteria, symbiotic bacteria found within healthy plant tissue, have been found in all plants studied to date. Endophytic bacteria can influence plant healthy through growth promotion, resistance to pests and pathogens, and confer plasticity to abiotic and biotic stress. Despite the role they can play in plant health, they have largely been examined in agricultural crops with little exploration of endophytic bacteria in long-lived forest trees. Forest trees may rely on bacterial endophytes to confer robustness to variable and environmental conditions they face over their lifetime. Subalpine conifers are long-lived trees that inhabit nutrient limited soils and harsh environments and therefore a good candidate to examine endophytic bacteria that may confer host robustness. In animals, core bacterial symbionts with important roles in host health are often consistently associated with individuals of a host species. To uncover potential beneficial endophytic bacteria, I used next-generation sequencing of the 16S rRNA bacterial gene to determine if subalpine conifers were host to a consistent core of endophytes. I found Pinus flexilis and Picea engelmannii at Niwot Ridge, CO were consistently dominated by a potential N2 fixing phylotype (AAB1). Next, to determine if the consistent AAB1 association was a result of shared environment (Niwot Ridge) or shared ancestry (Pinus and Picea), I characterized the needle endophytic community of P. flexilis and Pinus contorta at two distance sites: Niwot Ridge, CO and Horseshoe Meadows, CA. I found AAB1 to dominate the endophytic

xiv! ! bacterial communities across both species and sites, suggesting the association was not dependent only on environment but may be a long-term association of AAB1 and Pinus and Picea. To further examine the distribution of AAB1 I examined the endophytic community of P. contorta at Tuolumne Meadows, CA. AAB1was present but not in high abundance, indicating the AAB1 association is facultative. I found that the remaining endophyte community (i.e. excluding AAB1) of P. flexilis, P. engelmanni and P. contorta was structured by geographic location. In contrast to the high elevation pines, a core and consistent endophytic bacteria taxa was not identified in Sequoia sempervirens and Sequoiadendron giganteum in the family Cupressaceae . In these species, the endophyte communities were highly variable within and across tree individuals and species. These results suggest that subalpine conifers in the Pinaceae associate with facultative but symbiotic N2 fixing endophytic bacteria, and that the symbiosis may have originated after the divergence of Pinaceae and the Cupressaceae.

iv! !

1 Introduction

Mutualistic microbial symbionts have profound impact on animal and plant development, physiology, lifestyle, evolution, and adaptation (Compant et al. 2010a). For example, symbiotic microbes enhance plant tolerance of stressful environments (i.e. saline soil or drought) (Arachevaleta et al. 1989, Mayak et al. 2004, Rodriguez et al. 2009, Mei and Flinn 2010) suggesting they should be considered when examining plant adaptation to environmental stressors. Plant-associated microbes consist of bacteria and fungi in the rhizosphere, phyllosphere, and endosphere. Among these plant-associated microbes, bacterial endophytes are the least studied but their niche inside the plant positions them to play important roles in plant biology (Compant et al. 2010a). As a result of their ability to colonize the interior of plants and to confer benefits to host plants, it is likely plants develop long-term mutualistic associations with endophytic bacteria. Studies have shown that endophytic bacteria have been detected in all plants studied (Lodewyckx et al. 2002) and provide a variety of beneficial effects to their host, including resistance to pests and pathogens (Berg et al. 2005, Seo et al. 2010), promotion of plant growth (Compant et al. 2010a, Paz et al. 2012, Anad and Chanway 2013), and resistance to abiotic and biotic stress (Sziderics et al. 2007, Ryan et al. 2008). Given the beneficial properties of endophytic bacteria, it is likely they are an important component of plant adaptation to environmental perturbations. Environmental changes can disrupt symbiosis, pushing existing partnerships from beneficial to harmful, or lead to better adapted partnerships (Compant et al. 2010b). Therefore, climate change or other environmental changes may, through its effect on microbes, reduce conifer fitness or enable conifers to adapt to climate change. To have a better grasp on plant adaption to environmental changes, we need to examine adaptations of the plant genome, as well as their relationship with symbiotic microbes.

Subalpine ecosystems may be highly vulnerable to climate change given their adaptations to cold conditions and short growing seasons (Fagre et al. 2003). Conifer trees are able to inhabit the harsh and nutrient poor subalpine sites where few other plants grow. A component of their ecological success may be attributed to their microbial symbionts including bacterial endophytes and future success may rely on the conifer-endophytic

1! 2! bacteria relationship. Before we can explore the potential impact of climate change on the conifer-endophytic bacteria relationship, we need an understanding of the present role and community structure of endophytic bacterial communities which is largely unknown (Pirttilä and Frank 2011).

Endophytic bacteria have been found in all tree tissues: xylem sap (Filteau et al. 2010), roots (Mocali et al. 2003, Moore et al. 2006, Izumi et al. 2008, Bonito et al. 2014), stems (Bal et al. 2012), leaves (Ulrich et al. 2008), buds (Pirttilä et al. 2000) and shoots (Izumi et al. 2008) and with potential roles in tree health. For example, Paenibacillus polymyxa P2b-2R, a N2 fixing bacterium, was isolated from stems and needles of Pinus contorta (lodgepole pine) growing in N poor soils (Bal et al. 2012). In seedlings inoculated with the N2 fixing strain, P. polymyxa P2b-2R colonized the seedlings intracellularly, associating with chloroplasts (Anad and Chanway 2013) that require fixed N. However, it remains unclear whether P. polymoxa P2b-2R and other N2 fixing endophytes are found consistently with conifer trees or if potential N2 fixers are rare. Both host and environmental factors influence endophyte community structure but the strength of contribution of each factor is not well understood. There appears to be specificity in interactions as well as environmental variability in associations. For example, studies of Arabidopsis and Asclepias viridis, Ambrosia psilostachya, Sorghastrum nutans, Panicum virgatum, and Ruellia humilis (Ding et al. 2013) found plant species first and then site to be responsible for endophytic bacterial community structure. Furthermore, the presence of a core endophytic bacterial community with plant specific association patterns was found in Populus, Quercus, and Pinus root endophytic communities (Bonito et al. 2014).

Core symbiotic bacterial taxa often provide specialized beneficial functions to the host. For example, insect guts are consistently (all hosts individuals) colonized by microbes that play a role in nutrient acquisition (Engel and Moran 2013). Exploration of endophytic bacterial diversity and patterns of association has largely been limited culture-dependent methods. While these methods are able to uncover specific roles of bacteria, it has been estimated only 1% of bacterial communities are able to be cultivated (Amann 2000). If endophytic bacteria are obligate or closely evolved symbionts, it is less likely co-evolved microbes will be cultivated independent of their host plant. Furthermore, it is not guaranteed isolated bacteria are ecologically relevant. Culture-independent methods are ideal to explore the endophytic bacteria-plant symbiosis. Metagenomic methods would be ideal to uncover the genes present in endophytic communities but are inhibited by the high rate of plastid and mitochondrial contamination. Plastid and mitochondria share ancestry with bacteria, making it difficult to differentiate the DNA of host plant and endophytic bacteria. The 16S rRNA gene of bacteria contains sites with significant variation from plant DNA allowing the differentiation of host DNA and bacterial DNA at the 16S rRNA gene. High-throughput sequencing of the 16S rRNA enables us to characterize the majority of the endophytic communities across many samples. With sequence-based techniques, we can identify the

! 3! core bacterial taxa present in the majority of host individuals or environments (Qin et al. 2010, Martinson et al. 2011, Sharp et al. 2012, Lundberg et al. 2012, Rastogi et al. 2012). It is likely core taxa are mutualistic that provide an important beneficial function to the host (Engel and Moran 2013)and thus would be the focus of future in vitro studies.

Subalpine conifers are long-lived trees that inhabit nutrient limited soils and harsh environments. A component of their ability to colonize nutrient limited soils and withstand harsh environments may be their association with mutualistic endophytic bacteria. To my knowledge, the role and community structure of subalpine conifer bacterial endophytic communities has not been examined to date. The overall aim of my dissertation was to determine if sub-alpine conifers are host to potential mutualistic endophytic bacteria by detecting consistent core endophytic bacterial taxa. To identify core bacterial taxa in the sub-alpine stress tolerant conifer Pinus flexilis, I used high-throughput sequencing of the 16S rRNA gene of individuals of P. flexilis and co-occurring Picea engelmannii in Niwot Ridge, CO, and identified core bacterial taxa that were consistently associated with tree individuals within this site. Next, I compared the community composition of P. flexilis and P. contorta in Niwot Ridge, CO and Horseshoe Meadows, CA to determine if the identified core taxa were dependent on host-tree or environment. I then examined the core bacterial taxa associated with P. contorta in Tuolumne Meadows, CA. Finally, I compared the sub-alpine endophytic communities to endophytic communities in distantly related conifers, Sequioa sempervirens and Sequoiadendron giganteum.

The potential beneficial roles of endophytic bacteria may impact ecosystems at various ecological levels. Despite the potential roles of endophytic bacteria, the role and structure of endophytic bacterial communities in natural ecosystems are unknown. At the organism level, endophytes may influence the health of the individual through growth promotion or resistance to pests and pathogens. If bacterial endophytic communities are structured by tree species, populations of specific trees may have an ecological advantage of other tree species in the same area, impacting competition dynamics. At the ecosystem level, nutrient acquisition conferred by endophytic bacteria may influence the carbon and nitrogen cycling. Thus, my dissertation examining the bacterial endophytic communities of subalpine conifers has implications from the organism level to the ecosystem level.

1.1 Organization of Dissertation

The research from this dissertation is divided into four self-contained chapters written in manuscript format. In Chapter 2, “Pinus flexilis and Piceae engelmannii share a simple and consistent needle endophyte microbiota with a potential role in nitrogen fixation” I compared the endophytic communities of P. flexilis and co-occurring P. engelmannii in the same site. This manuscript was published in Frontiers of Microbiology on July 4, 2014. In Chapter 3 “The association between Pinaceae and an acetic acid bacterial needle endophyte is persistent across host species and geographic locations” I determined if the results from Chapter 2 were consistent across sites. In Chapter 4 “Enterobacteraceae dominate the bud and needle endophytic communities of Pinus contorta in a subalpine

! 4! meadow,” I examined the core taxa of P. contorta buds and needles. Finally, in Chapter 5 “Diversity and Structure of the Bacterial Endophyte Communities in the Foliage of Giant Trees” I examined the structure of bacterial endophytes in Sequioa sempervirens and Sequoiadendron giganteum to determine if the giant trees were structured by the same factors as high elevation conifers. Chapter 5 has been submitted to Applied and Environmental Microbiology. All chapters are written in the pronoun “we”, which refers to the co-authors of the manuscripts and myself.

1.2 References

Amann,!R.!2000.!Who!is!out!there?!Microbial!aspects!of!biodiversity.!Systematic!and! Applied!Microbiology!23:1–8.! ! Anad,! R.,! and! C.! Chanway.! 2013.! N2Xfixation! and! growth! promotion! in! cedar! colonized!by!an!endophytic!strain!of!Paenibacillus!polymyxa.!Biol!Fertil!Soils! 49:235–239.!

Arachevaleta,!M.!C.,!W.!Bacon,!C.!S.!Hoveland,!and!D.!E.!Radcliffe.!1989.!Effect!of!tall! fescue! endophyte! on! plant! response! to! environmental! stress.! Agronomy! Journal!81:89–30.!

Bal,!A.,!R.!Anand,!O.!Berge,!and!C.!P.!Chanway.!2012.!Isolation!and!identification!of! diazotrophic! bacteria! from! internal! tissues! of! Pinus! contorta! and! Thuja! plicata.!Can!J!Forest!Res!42:807–813.!

Berg,!G.,!A.!Krechel,!M.!Ditz,!R.!A.!Sikora,!A.!Ulrich,!and!J.!Hallmann.!2005.!Endophytic! and! ectophytic! potatoXassociated! bacterial! communities! differ! in! structure! and!antagonistic!function!against!plant!pathogenic!fungi.!FEMS!Microbiology! Ecology!51:215–229.!

Bonito,!G.,!H.!Reynolds,!M.!S.!Robeson,!J.!Nelson,!B.!P.!Hodkinson,!G.!Tuskan,!C.!W.! Schadt,!and!R.!Vilgalys.!2014.!Plant!host!and!soil!origin!influence!fungal!and! bacterial! assemblages! in! the! roots! of! woody! plants.! Molecular! Ecology! 23:3356–3370.!

Compant,!S.,!C.!Clément,!and!A.!Sessitsch.!2010a.!Plant!growthXpromoting!bacteria!in! the! rhizoX! and! endosphere! of! plants:! Their! role,! colonization,! mechanisms! involved!and!prospects!for!utilization.!Soil!Biology!and!Biochemistry!42:669– 678.!

Compant,!S.,!M.!G.!van!der!Heijden,!and!A.!Sessitsch.!2010b.!Climate!change!effects! on! beneficial! plantXmicroorganism! interactions.! FEMS! Microbiol! Ecol! 73:197–214.!

! 5!

Ding,! T.,! M.! W.! Palmer,! and! U.! Melcher.! 2013.! Community! terminal! restriction! fragment! length! polymorphisms! reveal! insights! into! the! diversity! and! dynamics!of!leaf!endophytic!bacteria.!BMC!Microbiology!13:1.!

Engel,! P.,! and! N.! A.! Moran.! 2013.! The! gut! microbiota! of! insects! X! diversity! in! structure!and!function.!FEMS!Microbiology!Reviews!37:699–735.!

Fagre,! D.! B.,! D.! L.! Peterson,! and! A.! E.! Hessl.! 2003.! Taking! the! Pulse! of! Mountains:! Ecosystem! Responses! to! Climatic! Variability.! Pages! 263–282! in!H.!F.!Diaz,! editor.! Climate! Variability! and! Change! in! High! Elevation! Regions:! Past,! Present!&!Future.!Springer!Netherlands.!

Filteau,!M.,!L.!Lagacé,!G.!LaPointe,!and!D.!Roy.!2010.!Seasonal!and!regional!diversity! of!maple!sap!microbiota!revealed!using!community!PCR!fingerprinting!and! 16S!rRNA!gene!clone!libraries.!Systematic!and!applied!microbiology!33:165– 173.!

Izumi,!H.,!I.!C.!Anderson,!K.!Killham,!and!E.!R.!Moore.!2008.!Diversity!of!predominant! endophytic! bacteria! in! European! deciduous! and! coniferous! trees.! Canadian! journal!of!microbiology!54:173–179.!

Lodewyckx,!C.,!J.!Vangronsvel,!F.!Porteous,!E.!R.!B.!Moore,!S.!Taghaviad,!M.!Mezgeay,! and! D.! van! der! Lelie.! 2002.! Endophytic! Bacteria! and! Their! Potential! Applications.!Critical!Reviews!in!Plant!Sciences!21:583–606.!

Lundberg,! D.! S.,! S.! L.! Lebeis,! S.! H.! Paredes,! S.! Yourstone,! J.! Gehring,! S.! Malfatti,! J.! Tremblay,!A.!Engelbrektson,!V.!Kunin,!T.!G.!del!Rio,!R.!C.!Edgar,!T.!Eickhorst,! R.!E.!Ley,!P.!Hugenholtz,!S.!G.!Tringe,!and!J.!L.!Dangl.!2012.!Defining!the!core! Arabidopsis!thaliana!root!microbiome.!Nature!488:86–90.!

Martinson,! V.! G.,! B.! N.! Danforth,! R.! L.! Minckley,! O.! Rueppell,! S.! Tingek,! and! N.! A.! Moran.!2011.!A!simple!and!distinctive!microbiota!associated!with!honey!bees! and!bumble!bees.!Mol!Ecol!20:619–28.!

Mayak,!S.,!T.!Tirosh,!and!B.!R.!Glick.!2004.!Plant!growthXpromoting!bacteria!confer! resistance!in!tomato!plants!to!salt!stress.!Plant!Physiol!Biochem!42:565–72.!

Mei,!C.,!and!B.!S.!Flinn.!2010.!The!use!of!beneficial!microbial!endophytes!for!plant! biomass!and!stress!tolerance!improvement.!Recent!Pat!Biotechnol!4:81–95.!

Mocali,! S.,! E.! Bertelli,! F.! Di! Cello,! A.! Mengoni,! A.! Sfalanga,! F.! Viliani,! A.! Caciotti,! S.! Tegli,!G.!Surico,!and!R.!Fani.!2003.!Fluctuation!of!bacteria!isolated!from!elm! tissues!during!different!seasons!and!from!different!plant!organs.!Research!in! Microbiology!154:105–114.!

! 6!

Moore,!F.!P.,!T.!Barac,!B.!Borremans,!L.!Oeyen,!J.!Vangronsveld,!D.!Van!der!Lelie,!C.!D.! Campbell,! and! E.! R.! Moore.! 2006.! Endophytic! bacterial! diversity! in! poplar! trees!growing!on!a!BTEXXcontaminated!site:!the!characterisation!of!isolates! with! potential! to! enhance! phytoremediation.! Systematic! and! applied! microbiology!29:539–556.!

Paz,!I.!C.,!R.!C.!Santin,!A.!M.!Guimaraes,!O.!P.!Rosa,!A.!C.!Dias,!M.!C.!Quecine,!J.!L.! Azevedo,! and! A.! T.! Matsumura.! 2012.! Eucalyptus! growth! promotion! by! endophytic!Bacillus!spp.!Genet!Mol!Res!11:3711–20.!

Pirttilä,! A.,! and! A.! C.! Frank.! 2011.! Endophytes! of! forest! trees:! Biology! and! Application.!Springer.!

Pirttilä,!A.!M.,!H.!Laukkanen,!H.!Pospiech,!R.!Myllyla,!and!A.!Hohtola.!2000.!Detection! of!intracellular!bacteria!in!the!buds!of!Scotch!pine!(Pinus!sylvestris!L.)!by!in! situ!hybridization.!Appl!Environ!Microbiol!66:3073–7.!

Qin,!J.,!R.!Li,!J.!Raes,!M.!Arumugam,!K.!S.!Burgdorf,!C.!Manichanh,!T.!Nielsen,!N.!Pons,! F.!Levenez,!T.!Yamada,!D.!R.!Mende,!J.!Li,!J.!Xu,!S.!Li,!D.!Li,!J.!Cao,!B.!Wang,!H.! Liang,!H.!Zheng,!Y.!Xie,!J.!Tap,!P.!Lepage,!M.!Bertalan,!J.!M.!Batto,!T.!Hansen,!D.! Le! Paslier,! A.! Linneberg,! H.! B.! Nielsen,! E.! Pelletier,! P.! Renault,! T.! SicheritzX Ponten,!K.!Turner,!H.!Zhu,!C.!Yu,!S.!Li,!M.!Jian,!Y.!Zhou,!Y.!Li,!X.!Zhang,!S.!Li,!N.! Qin,! H.! Yang,! J.! Wang,! S.! Brunak,! J.! Dore,! F.! Guarner,! K.! Kristiansen,! O.! Pedersen,!J.!Parkhill,!J.!Weissenbach,!H.!I.!T.!C.!Meta,!P.!Bork,!S.!D.!Ehrlich,!and! J.! Wang.! 2010.! A! human! gut! microbial! gene! catalogue! established! by! metagenomic!sequencing.!Nature!464:59–65.!

Rastogi,!G.,!A.!Sbodio,!J.!J.!Tech,!T.!V.!Suslow,!G.!L.!Coaker,!and!J.!H.!Leveau.!2012.!Leaf! microbiota! in! an! agroecosystem:! spatiotemporal! variation! in! bacterial! community!composition!on!fieldXgrown!lettuce.!ISME!J!6:1812–22.!

Rodriguez,! R.! J.,! J.! F.! White! Jr.,! A.! E.! Arnold,! and! R.! S.! Redman.! 2009.! Fungal! endophytes:!diversity!and!functional!roles.!New!Phytol!182:314–30.!

Ryan,! R.! P.,! K.! Germaine,! A.! Franks,! D.! J.! Ryan,! and! D.! N.! Dowling.! 2008.! Bacterial! endophytes:! recent! developments! and! applications.! FEMS!Microbiol!Lett! 278:1–9.!

Seo,!W.!T.,!W.!J.!Lim,!E.!J.!Kim,!H.!D.!Yun,!and!K.!M.!Cho.!2010.!Endophytic!Bacterial! Diversity! in! the! Young! Radish! and! Their! Antimicrobial! Activity! against! Pathogens.! Journal! of! the! Korean! Society! for! Applied! Biological! Chemistry! 53:493–503.!

Sharp,! K.! H.,! D.! Distel,! and! V.! J.! Paul.! 2012.! Diversity! and! dynamics! of! bacterial! communities! in! early! life! stages! of! the! Caribbean! coral! Porites! astreoides.! ISME!J!6:790–801.!

! 7!

Sziderics,!A.!H.,!F.!Rasche,!F.!Trognitz,!A.!Sessitsch,!and!E.!Wilhelm.!2007.!Bacterial! endophytes! contribute! to! abiotic! stress! adaptation! in! pepper! plants! (Capsicum!annuum!L.).!Can!J!Microbiol!53:1195–202.!

Ulrich,! K.,! A.! Ulrich,! and! D.! Ewald.! 2008.! Diversity! of! endophytic! bacterial! communities! in! poplar! grown! under! field! conditions.! FEMS! microbiology! ecology!63:169–180.!

! !

2 Pinus flexilis and Picea engelmannii share a

simple and consistent needle endophyte

microbiota with a potential role in nitrogen

fixation.

2.1 Abstract

Conifers predominantly occur on soils or in climates that are suboptimal for plant growth. This is generally attributed to symbioses with mycorrhizal fungi and to conifer adaptations, but recent experiments suggest that aboveground endophytic bacteria in conifers fix nitrogen (N) and affect host shoot tissue growth. Because most bacteria cannot be grown in the laboratory very little is known about conifer-endophyte associations in the wild. Pinus flexilis (limber pine) and Picea engelmannii (Engelmann spruce) growing in a subalpine, nutrient-limited environment are potential candidates for hosting endophytes with roles in N2 fixation and abiotic stress tolerance. We used 16S rRNA pyrosequencing to ask whether these conifers host a core of bacterial species that are consistently associated with conifer individuals and therefore potential mutualists. We found that while overall the endophyte communities clustered according to host species, both conifers were consistently dominated by the same phylotype, which made up 19-53% and 14-39% of the sequences in P. flexilis and P. engelmannii respectively. This phylotype is related to Gluconacetobacter diazotrophicus and other N2 fixing acetic acid bacterial endophytes. The pattern observed for the P. flexilis and P. engelmannii needle microbiota—a small number of major species that are consistently associated with the host across individuals and species—is unprecedented for an endophyte community, and

7! 8! suggests a specialized beneficial endophyte function. One possibility is endophytic N fixation, which could help explain how conifers can grow in severely nitrogen-limited soil, and why some forest ecosystems accumulate more N than can be accounted for by known nitrogen input pathways.

2.2 Introduction

Bacterial endophytes inhabit the below- and aboveground tissues of all terrestrial plants examined and can affect plant physiology and growth under normal and stressed conditions. Endophytic bacteria can stimulate plant growth directly through production of phytohormones and volatiles (Arshad and Frankenberger 1991, Ping and Boland 2004, Hardoim et al. 2008), enhance nutrient acquisition (Boddey et al. 1999, Hurek et al. 2002), and suppress stress-induced ethylene synthesis (Glick 2004). Bacterial endophytes have also been found to protect against disease (Conn et al. 2008), and against abiotic stress such as salinity, and heavy metals (Idris et al. 2004, Mayak et al. 2004). At the ecosystem level, bacterial endophytes can persistently alternate soil biogeochemical cycles through the support of invasive plants (Rout et al. 2013).

Research to date on bacterial endophytes has focused mainly on agricultural ecosystems (Hallman et al. 1997), and to some extent, on invasive plants (Rout et al. 2013). Our knowledge of the role, diversity, and transmission of bacterial endophytes colonizing native plants is still limited. However, plant-beneficial endophytic properties are likely to have evolved in, and continually influence plants in natural ecosystems. A better understanding of the bacteria that inhabit wild plants has the potential to impact our understanding not only of basic plant physiology, but also of whole ecosystem processes such as carbon (C) and nitrogen (N) cycling.

While it is well established that many forest conifers depend on associations with mycorrhizal- (Smith and Read 2008) and foliar endophytic fungi (Carroll 1988, Arnold et al. 2003, Higgins et al. 2011), our understanding of the bacterial endophytes of conifers is limited (Pirttilä and Frank 2011). Bacteria have been isolated from the interior of roots, stems, needles, seeds and tissue culture of conifers (Pirttilä et al. 2000, Cankar et al. 2005, Izumi et al. 2008, Bal et al. 2012). Bacteria in the genus Methylobacterium likely play a role in the shoot tissue development of Pinus flexilis (Pirttilä et al. 2004, 2005). N2 fixing bacteria isolated from stems and needles of Pinus contorta have been found to increase the uptake atmospheric N2 in inoculated seedlings relative to control seedlings (Bal and Chanway 2012, Anand and Chanway 2013). A similar experiment on poplar clones inoculated with a consortium of N2 fixing bacteria suggests that N allocated to the leaves and stem in the poplar clones were derived largely from biological nitrogen fixation (Knoth et al. 2013). These results raise numerous questions: do non-nodulated trees in natural ecosystems acquire their N from endophytic N2 fixation? If so, are associations random or stable, and do single species or consortia of N2 fixing bacteria fix N2? Finally, does endophytic N2 fixation, along with other recently discovered nitrogen input pathways (DeLuca et al. 2002, Morford et al. 2011, Vitousek et al. 2013) explain the hidden input of N in temperate and boreal forests (Bormann et al. 1993, 2002,

! 9!

Binkley et al. 2000)? Unfortunately, a culture-based and experimental approach to answering these questions may not yet be feasible. Large discrepancies in the composition between cultured and uncultured endophytic communities (Ando et al. 2005) suggest that most bacterial endophytes may resist cultivation. In addition, there is no guarantee that an isolated species represents a dominant member of the endophytic community.

In general, sequence-based methods may be necessary to uncover some of the hidden symbioses in terrestrial ecosystems (Poole et al. 2012). High-throughput sequencing of the 16S rRNA can be used to identify the core subset of bacterial taxa present in the majority of host individuals, habitats, seasons, environments or developmental stages (Qin et al. 2010, Martinson et al. 2011, Sharp et al. 2012, Lundberg et al. 2012, Rastogi et al. 2012). Such core taxa often provide specialized beneficial functions to a host. In insect guts for example, a simple (i.e. consisting of few species) and consistent (i.e. present in all host individuals examined) microbiota typically plays a role in host nutrient acquisition (Engel and Moran 2013). Once established, the role of core uncultured members of the community can be further evaluated using in-situ methods (Woebken et al. 2012, Sharp et al. 2012).

Here, we used pyrosequencing of the 16S rRNA gene with the goal of identifying the core subset of bacterial species inside needles that are consistently associated with individuals of a conifer host species. Specifically, we asked whether the needle microbiota is conserved across individuals and stands of the high-elevation, stress-tolerant conifer species P. flexilis (limber pine) growing within one geographic area (Niwot Ridge in the Front Range of the Rocky Mountains, CO, USA), and whether the P. flexilis needle microbiota is shared with the co-occurring species P. engelmannii (Engelmann spruce).

This first comprehensive analysis of the bacterial endophyte microbiota in a gymnosperm revealed that the composition of P. flexilis and P. engelmannii needle endophyte communities is simple at the species level; that both conifer hosts are consistently dominated by potential N2 fixing taxa in the family Acetobacteraceae; and that the microbiota is largely shared between the two host species. We discuss the implications of these results in light of recent findings that bacterial endophytes appear to fix N2 inside conifer tissues in an experimental system.

2.3 Materials and methods

2.3.1 Sample collection and sterilization

Needles were collected from 14 trees at two elevations in the subalpine forest at Niwot Ridge, CO in September 2009—at tree line (TL) and near the local warm edge (WE) limit of the P. flexilis and P. engelmannii distributions (Table 1). To assess the difference in endophytic communities across locations, six P. flexilis and six P. engelmannii trees

! 10! were sampled at two separate warm edge stands (WE1 and WE2), approximately 0.5 km apart. To contrast inter- and intra tree variation in the endophytic community, one individual of each tree species at each stand was sampled in triplicate. Additionally, two P. flexilis and two P. engelmannii individuals were sampled at treeline, the upper elevation limit of upright tree growth. From each sampled tree, approximately 10 g of needles were removed with a sterile razor blade, placed in a sterile bag, and shipped to the University of California, Merced at 4°C for sterilization and extraction. Needles were sterilized through submersion in ethanol for one minute, 30% hydrogen peroxide for three minutes, followed by three rinses with sterile deionized water, and stored at -20°C. The final rinse after sterilization was saved to verify sterility (Izumi et al 2008).

2.3.2 DNA Extraction

Needles were ground to a fine powder in a sterile mortar with liquid nitrogen. In a 2 ml screw cap tube, 800µl of CTAB solution (1 ml CTAB buffer, 0.04 g of polyvinylpyrollidone, 5 µl of 2-mercaptoethanol) was added to 0.6 g of ground needle material. The tube was then incubated in a dry bath at 60°C for 2 hours with intermittent vortexing. After incubation, 0.3 g of 0.11 mm sterile glass beads was added to the tube and the sample was homogenized using a bead beater for 3 minutes. To remove proteins, an equal amount of chloroform was added to the tube, vortexed, and centrifuged for 10 minutes at 16 rcf. For precipitation of nucleic acids, the aqueous top phase was placed in a sterile 2 ml snap cap tube with 1/10 volume of cold 3 M sodium acetate and 1/2 volume cold isopropanol and placed in a –20°C freezer for 12 hours. The sample was then centrifuged for 3 minutes at 16 rcf, supernatant decanted, 700 µl of 70% ethanol added, and centrifuged for 10 minutes. The air-dried pellet was resuspended with 30 µl of DNA resuspension fluid (1.0 M Tris-HCL, 0.1 M EDTA) and stored at –20°C.

2.3.3 DNA Amplification

The extract was used to perform a nested PCR with Golay barcoded, chloroplast- excluding primers, 16S 799f (AACMGGATTAGATACCCKG) and 16S 1492r (TACGGHTACCTTGTTACGACTT) (Chelius and Triplett 2001, Redford et al. 2010), using the thermocycle profile described in Jiao et al. (2006). We used nested PCR to reduce the occurrence of plastid sequences, improve consistency (Hanshew et al. 2013) and minimize non-specific amplification from the barcoded primers that include 454 adapters (Berry et al. 2011). PCR amplification with primer 16S 799f resulted in a mitochondrial product of about 1000 bp and bacterial product of about 750 bp as described in Chelius and Triplett (2001). The bacterial product was then separated and extracted using E-Gel® SizeSelect™ Gels (Life Technologies, Carlsbad, CA, USA). The

! 11!

extracted bacterial product was amplified with the thermocycle profile described by Jiao et al. (2006) using the barcoded primer set, 799f and 1115r (AGGGTTGCGCTCGTTG), described by Redford et al. (2010) as an optimized primer set for phylogenetic analysis of pyrosequencing reads. The final product was then cleaned, quantified using Nanodrop, and pooled for pyrosequencing. The pooled product was sent to the Environmental Genomics Core Facility at the University of South Carolina for pyrosequencing on a 454 Life Sciences Genome Sequencer FLX machine.

2.3.4 Sequence Analysis

Sequences were analyzed and processed using the QIIME package (Caporaso et al. 2010b). Brieﬂy, sequences were quality filtered (minimum quality score of 25, minimum length of 200 bp, and no ambiguity in primer sequence) and assigned to their corresponding sample by the barcode sequences. Samples with less than 200 sequences were removed. These included three samples from a P. engelmannii WE1 individual, and one sample from one P. engelmannii WE2 individual. The remaining sequences were clustered into phylotypes using UCLUST (Edgar 2010), with a minimum coverage of 99% and a minimum identity of 97%. A representative sequence was chosen for each phylotype by selecting the longest sequence that had the highest number of hits to other sequences of that particular phylotype. Chimeric sequences were detected with ChimeraSlayer and removed before taxonomic analysis (Edgar et al. 2011). Representative sequences were aligned using PyNAST (Caporaso et al. 2010a) against the Greengenes core set (DeSantis et al. 2006). Taxonomic assignments were made using the Ribosomal Database Project (RDP) classiﬁer (Wang et al. 2007). Sequences classified as “Chloroplast” (0.2%) or “Mitochondria” (8%) were removed from the alignment.

To compare diversity levels between warm edge and treeline samples and control for differences in sequencing depth between samples from the two environments, we conducted rarefaction analyses with 800 randomly selected sequences per sample. The rarefaction curves are displayed in Figure 2-1. The relative abundance of bacterial classes in each sample, displayed in Figure 2-2, was calculated as the percentage of sequences belonging to a particular phylum of all 16S rRNA gene sequences recovered from each sample, with the Proteobacteria split into classes. The Alphaproteobacterial phylogenetic tree displayed in Figure 2-3 was created by first searching Alphaproteobacterial sequences that occurred at least 100 times in our data against the GenBank 16S rRNA database, using BLAST. The top hit for each sequence was then downloaded aligned from RDP. Our sequences, along with an outgroup sequence (Burkholderia arboris) were added to the alignment using ClustalW, before a maximum likelihood tree was inferred using RAxML (1,000 bootstrap replicates) (Stamatakis et al. 2005). To create the heatmap displayed in Figure 2-4, the heatmap function in QIIME was used. The function visualizes the operational taxonomic unit (OTU) table generated by QIIME (this table tabulates the number of times an OTU is found in each sample). In Figure 2-4, only the 10 most common OTUs (phylotypes) were included. To create Figure 2-5, a heatmap of all OTUs was generated to identify phylotypes unique to each species and shared across all samples, respectively. For each phylotype in Figures 2-4 and 2-5, the similarity to

! 12! known isolates was determined though a BLAST search against the NCBI 16S rRNA database. To create Figure 2-6, an approximately maximum-likelihood tree was constructed from the alignment using FastTree (Price et al. 2009). An unweighted UniFrac distance matrix was constructed from the phylogenetic tree. The unweighted Unifrac distances were visualized using principal coordinate analysis (PCoA) and an UPGMA tree was created from the UniFrac distance matrix. Confidence ellipses (95%) were drawn around groups on the PCoA plots using the ordiellipse function of the Vegan package in R (Oksanen, J. et al. 2008).

2.4 Results

2.4.1 Phylotypes recovered from samples

A total of 388 distinct bacterial phylotypes distributed across 166 bacterial phyla were recovered from 29,297 quality sequences from the 20 samples. These sequence data have been submitted to the GenBank databases under project accession No. SRP033097. On average, each sample yielded 1465 sequences after plant DNA was removed. The number of sequences recovered did not differ greatly between individuals, species, or locations (Table 2-1). The average number of phylotypes recovered also did not vary much within individuals or across species, but varied across locations. An average of 137 endophytic phylotypes were recovered from warm edge samples while an average of 53 phylotypes were recovered from treeline samples (Table 2-1). Rarefaction curves suggested that the lower number of phylotypes recovered from treeline samples was not due to insufficient sampling (Figure 2-1).

2.4.2 Dominant bacterial taxa associated with P. flexilis and P. engelmannii needles

We found a low degree of intra-individual, inter-individual, and interspecies variability in the taxonomic structure of the endophytic communities of our two conifer species. The majority of detected taxa belonged to two bacterial phyla: Acidobacteria and Proteobacteria (Figure 2-2). The proportion of Acidobacteria varied among samples and accounted for between 2 and 17 % in the warm edge samples, but only 0.5-3 % in the treeline samples. All of our samples were dominated by taxa in the Alphaproteobacteria (Figure 2-2). The Alphaproteobacterial phylotypes recovered here belong to five families: Acetobacteraceae, Beijerinckiaceae, Caulobacteraceae, Methylobacteriaceae, and Sphingomonadaceae.

Among the Alphaproteobacteria, and over all, the Acetobacteraceae (acetic acid bacteria) was the most common family recovered in our samples. Eighty-eight out of the total 188 Alphaproteobacterial phylotypes detected belonged to this family, which dominated all samples, accounting for at least 60% of phylotypes detected in each sample. This family is currently classified into 13 genera; Acetobacter, Gluconobacter, Gluconacetobacter, Acidomonas, Asaia, Kozakia (Kersters et al. 2006), Swaminathania (Loganathan and Nair

! 13!

2004), Saccharibacter (Jojima et al. 2004), Neoasaia (Yukphan et al. 2005), Granulibacter (Greenberg et al. 2006), Commensalibacter (Roh et al. 2008), Tanticharoenia (Yukphan et al. 2008), and Ameyamaea (Yukphan et al. 2009). Figure 3 shows the phylogenetic relationship among the most common Alphaproteobacterial taxa in our samples, demonstrating a high diversity of Acetobacteraceae. However, the high predominance of Alphaproteobacteria and Acetobacteraceae at the class and family levels was driven by the consistent occurrence and high relative abundance of a few single phylotypes (Figure 2-4A,B). One single phylotype (1045), which is 97% similar to Gluconacetobacter diazotrophicus, was the most common in all our samples. The exact phylogenetic placement of phylotype 1045 cannot be completely resolved with the 16S V-5/V-6 region used in this study (Figure 2-3), and may represent a novel species in the genus Gluconacetobacter or a related genus within the Acetobacteraceae.

The relative abundance and identity of the rest of the dominant endophytic taxa differed somewhat between the two conifer species, although there was also substantial overlap. The most prominent difference was seen between warm edge and treeline individuals, rather than between the two conifer species, although our small sample size of treeline trees (two trees of each species) prohibits us from drawing any conclusions based on these. Overall, the community of dominant taxa in P. flexilis was less diverse; in addition to 1045 only one more phylotype (1516) was consistently present above 10% in the warm edge samples, and another (1068) consistently present above 4% (Figure 2-4A). Both phylotypes 1516 and 1068 belong to the Acetobacteraceae. Phylotype 1516 is 95% similar to Gluconacetobacter liquefaciens, and phylotype 1068 is 98% similar to Tanticharoenia sakaeratensis. The rest of the 10 most abundant phylotypes in P. flexilis warm edge samples were present at much lower relative abundance (Figure 2-4A). An analysis of all phylotypes in all samples ensured that no individual sample was dominated by a phylotype other than the 10 most abundant in each tree species (data not shown).

The two conifer species shared five of their 10 most dominant phytotypes (Figure 2-4A,B). P. engelmannii warm edge samples had a slightly more diverse community of dominant phylotypes, with more phylotypes present above 4%, and with more variation in dominant phylotypes within and among individuals. Like P. flexilis, P. engelmannii had a high relative abundance of phylotype 1516 (5-16%). P. engelmannii trees at warm edge sites were also dominated by phylotypes not found in abundance in treeline P. engelmannii or P. flexilis samples of any stand; phylotype 1507 is a Methylobacterium species; 1295 is 95% similar to Terriglobus roseus in the family Acidobacteria; and 1324 is 95% similar to Kozakia balinesis, another species in the family Acetobacteraceae (Figure 2-4B).

Phylotypes shared across all individuals in one conifer species but not found in the other conifer species were rare (Figure 2-5). Only a limited subset of the needle endophytic microbiota appeared to be host species-specific; instead the differences in endophytic community composition between the two hosts were due to different relative abundances of a set of shared dominant phylotypes. The relative abundance of phylotype 1045 was highest in trees growing at treeline, making up 53 and 49 % of the sequences in the P. flexilis treeline individuals, and 38 and 39% of the sequences in the P. engelmannii

! 14! treeline individuals (Figure 2-4A,B). In the two P. flexilis treeline individuals, phylotype 1644, which is 97% similar to Asaia lannaensis (also in the Acetobacteraceae family), was the second most common after phylotype 1045. One P. engelmannii individual at treeline also had an increased relative abundance of phylotype 1644. The other P. engelmannii treeline individual had an increased relative abundance of a different phylotype—816, which is 99% similar to Pandoraea pnomenusa in the Burkolderiales family.

2.4.3 Structure of conifer needle endophyte communities

PCoA of the unweighted UniFrac diversity distance matrix demonstrated that within the warm edge habitat, host species identity controls endophyte community composition (Figure 2-6). P. flexilis samples from the two warm edge locations clustered to the exclusion of P. engelmannii samples from the same locations. Conifer needle endophyte community structure may also be a function of elevation, as the endophyte communities of our treeline samples clustered together to the exclusion of communities from warm edge samples. However, endophytic communities from more treeline samples will need to be characterized to confirm this observation.

2.5 Discussion

Our results demonstrate that P. flexilis and P. engelmannii growing at Niwot Ridge are consistently colonized by a limited set of bacterial species belonging to the Alphaproteobacteria and the Acidobacteria, and that the two species share some of the most dominant phylotypes. Most remarkably, all needle samples of both conifer species were dominated by the same phylotype, a species most closely related to the N2 fixing species G. diazotrophicus. However, is important to point out that 97% similarity over the V-5/V-6 region is not sufficient to identify a particular species. The primer pair used in this study have been used with other plants without resulting in a similar dominance across samples of one phylotype (Bodenhausen et al. 2013). Therefore, the patterns observed here are not likely to be caused by primer bias. It is also possible that specific taxa could be selected by our treatment (e.g. transportation in plastic bags at 4°C). However, this is unlikely since endophyte communities of other conifer tissues and species treated the same way are not consistently dominated by a few taxa (Carrell and Frank, unpublished). In addition, we find the same pattern in needles of P. contorta where the time from sampling to sterilization was less then 2 hours (Carrell and Frank, unpublished).

Bacterial species that are consistently recovered from a certain host species are predicted to be critical to the function of the microbial community (Shade and Handelsman 2012). For host-associated communities, this may translate into distinct functional roles for those species within the host. However, abiotic factors could also shape host-associated bacterial communities, contributing to consistency in microbiota across samples if the

! 15! environment is consistent. The dominance of Acidobacteria and Alphaproteobacteria in all our samples probably reflects the ability of these bacteria to survive the conditions within conifer needles. In addition, functionally relevant conifer-bacteria partnerships could underlie their dominance in the endophytic community.

The relative abundance of specific phylotypes from the Acidobacteria was low and their presence was not consistent across samples. Given that Acidobacteria have been found to dominate the alpine dry meadow site soils at Niwot Ridge (Lipson and Schmidt 2004), their overall prevalence might reflect their ability to tolerate the needle environment and their abundance in the soil rather than a functional relationship with the conifer host. Alphaproteobacteria are also common in the soil at Niwot Ridge (Lipson and Schmidt 2004), however, in contrast to the Acidobacteria, the dominance of Alphaproteobacteria in all our samples was driven by a few specific phylotypes from the Acetobacteraceae (e.g. 1045, 1516, and 1644 in P. flexilis, and 1045, 1516, 1324 and 1644 in P. engelmannii), suggesting that in addition to tolerating the needle environment, those bacterial taxa serve distinct functional roles within the needles of our subalpine conifers. An association based merely on the ability to survive the environment inside pine needles would more likely result in a variable (within and among individuals) community of taxa in the Acetobacteraceae.

Several high-throughput surveys of endophytic 16SrRNA have been published recently, but no study to date has reported a similarly simple and consistent endophyte microbiota. A study of the Populus deltiodes root endophyte community reported that a Pseudomonas phylotype accounted for 34% of bacterial sequences in all samples combined, however the extent to which this phylotype was consistently associated with individual trees was not reported (Gottel et al. 2011). Similarly, a study of the leaf endo- and epiphytes of Arabidopsis thaliana found a high relative abundance of a single Pseudomonas phylotype at 10.9% on average in the endophyte community (Bodenhausen et al. 2013), but inter-individual variability in the relative abundance of phylotypes was not studied. Finally, two recent large-scale studies of A. thaliana root endophytes did not report consistent dominance of one or a few phylotypes (Bulgarelli et al. 2012, Lundberg et al. 2012), and comparison across A. thaliana root microbiomes did not reveal a common core at the phylotype level (Schlaeppi et al. 2014).

It is important to point out that the phylotypes with the highest number of sequences are not necessarily the most abundant in the community. One potential source of bias is the primer used in this study (799f), which was designed to exclude chloroplast sequences, but likely excludes other sequences as well (e.g. the Cyanobacteria). However, because this primer has been used with very different results in terms of community diversity, variability and taxonomic distribution (Bodenhausen et al. 2013, Schlaeppi et al. 2014), it unlikely underlies the community consistency observed here.

The high relative abundance of a few phylotypes across individuals, host species and elevations may be due to ecologically significant associations between conifers and specific bacteria in the family Acetobacteraceae. Species in this family are often found as endophytes, with documented functions in N fixation (Boddey et al. 2003),

! 16! phytohormone production (Lee et al. 2004) and pathogen antagonism (Blanco et al. 2005). Acetobacteraceae are also common symbionts of insects (Crotti et al. 2010). A number of strains from the family are known to solubilize phosphate (P) (Loganathan and Nair 2004). Although P solubilizing endophytes can increase plant uptake of P (Verma et al. 2001, Wakelin et al. 2004), this is not a likely function of above-ground endophytes. Therefore, N2 fixation, pathogen protection, or phytohormone production is more likely to underlie the high relative abundance of Acetobacteraceae inside P. flexilis and P. engelmannii needles. G. diazotrophicus, Gluconacetobacter johannae, Gluconacetobacter azotocaptans, Swaminathania salitolerans, and Acetobacter peroxydans all fix N2 in association with plants (Gillis et al. 1989, Fuentes-Ramirez et al. 2001, Loganathan and Nair 2004, Muthukumarasamy et al. 2005), and Gluconacetobacter kombuchae and Acetobacter nitrogenifigens, are free-living N2 fixers (Dutta and Gachhui 2006, 2007). Among N2 fixing endophytes in the Acetobacteraceae, G. diazotrophicus is probably the best studied. This species colonizes the intercellular spaces of sugarcane stems (Dong et al. 1994), and may, together with other diazotrophic endophytes of sugarcane, provide the host plant with substantial amounts of N (Boddey 1995). G. diazotrophicus has also been found to dominate N2 fixation 'hotspots' in soil (Reed et al. 2010).

Fertilization experiments in the subalpine forest in Colorado have suggested that this habitat is still N limited, despite increasing N deposition (Rueth and Baron 2002), raising the possibility that this partnership involves endophytic N2 fixation. More research is needed to evaluate the role of phylotype 1045 in our conifer species, however, evaluation of biological N2 fixation associated with non-legumes can be challenging (Boddey 1995). Phylotype 1045 may not be a culturable strain as our initial attempts to culture any Acetobacteraceae from needles have been unsuccessful (Wilson and Frank, unpublished). N2-fixing potential and expression of nitrogenase in e.g. soil and ocean water is commonly estimated via amplification of nifH with degenerate primers (Farnelid et al. 2013, Collavino et al. 2014). Unfortunately, although a few studies report amplification of nifH (nitrogenase) from DNA extracted from plants (e.g. sugarcane, sweet potato and rice roots) (Ueda et al. 1995, Reiter et al. 2003, Ando et al. 2005), this has proven challenging in our system..

The idea that conifers form associations with N2-fixing bacteria is not new. Early attempts to estimate rates of N2 fixation in temperate and boreal forests detected significant amounts of atmospherically derived N in the conifer canopy (Richards 1964, Jones et al. 1974, Favilli and Messini 1990). Moreover, natural abundance studies that exploit naturally occurring differences in 15N composition between plant-available N sources in the soil and that of atmospheric N2, have shown that the Pinaceae, which presumably do not fix appreciable amounts of N2, run counter to the expectation of high 15 N abundance and low foliar content for non-N2 fixing plants (Delwiche et al. 1979, Virginia and Delwiche 1982, Näsholm et al. 2009).

Endophytic N2 fixation could explain why some gymnosperms are able to grow in extremely N limited environments (Anand et al. 2013). It is generally presumed that conifers and other non-nodulated plants get their N from the soil. Conifer uptake of

! 17! amino acids and proteins has been demonstrated but its quantitative importance is still under debate (Näsholm et al., 2009). Rhizospheric N2 fixation has been suggested as a conifer N source , but the activity has been found too low to support N requirement, at least in pines (Chanway and Holl 1991). Ectomychorrhizae are another suggested N source (Hobbie et al. 2000, Chapman and Paul 2012), but work on nonmycorrhizal conifer seedlings provides strong support for endophytic N2 fixation as an alternative explanation: Conifer seedlings inoculated with N2 fixing Paenibacillus strain isolated from P. contorta growing in N limited soil (Bal et al. 2012) have been found to acquire significant amounts of seedling foliar N from the atmosphere compared to control seedlings (Bal and Chanway 2012, Anand and Chanway 2013).

In light of this, it is striking that several potentially N2 fixing taxa dominate the community of needle-associated endophytes of conifers growing in the N limited subalpine environment at Niwot Ridge. Phylotype 1045, which is 97% similar to G. diazotrophicus, and the most common in every single sample, is the main candidate for an N2 fixing symbiosis in P. flexilis and P. engelmannii at this location. Interestingly, we found that phylotype 1045 was present in higher relative abundance in P. flexilis, which is found on drier and likely lower N soils than P. engelmannii, which is found on more mesic and potentially higher N soils. We also found that in both conifer species, the relative abundance of this phylotype was higher in trees growing at treeline than in trees growing at the warm edge, which could reflect slower N turnover at treeline or higher N demands by treeline trees. However, total abundance of phylotype 1045 may not differ among warm edge and treeline trees; it may simply be a function of overall lower diversity in treeline environments. More endophyte community data from the treeline environment will be required to establish links between the conifer endophyte community structure and elevation.

A consistent core microbiota similar to the one observed here would be promoted by selective uptake/transfer of bacteria. Not much is known with regards to the transmission of bacterial endophytes in conifer trees, but possible routes include soil, litter, seed and stomata. Given the long lifetime of P. flexilis and P. engelmannii, as well as needle longevity (e.g. 4-10 years for P. flexilis) (Schoettle and Rochelle 2000), the time spent within individual hosts is expected to be longer than for most host-microbe associations, possibly leading to convergence in the endophyte community assembly among different individuals. The consistency in the endophytic community between the two host species is intriguing. This pattern could reflect a strong local and environmental influence on endophyte uptake, leading to a shared endophyte community between co-occurring species in a shared habitat. Alternatively, the selective uptake/transfer of specific strains of Acetobacteraceae could be an ancestral trait that is shared between Pinus and Piceae species, which together with Cathaya form a clade in the Pinaceae phylogeny (Lin et al. 2010). If so, the dominance of phylotype 1045 in both host species may be the result of a long-standing association with possible co-diversification of host and endophyte. Our current sequences—approximately 300 nucleotides across over the V-5/V-6 region—do not provide enough resolution to evaluate possible divergence between strains

! 18! represented by phylotype 1045 from the two conifer species. To understand why P. flexilis and P. engelmannii at Niwot Ridge have such similar endophyte communities, further research is needed that includes a broader range of conifer species, habitats, and geographic locations, along with the amplification of longer segments of the 16S rRNA gene.

This first comprehensive analysis of the bacterial endophyte communities in a gymnosperm provides circumstantial evidence for endophytic N2 fixation in subalpine conifers growing in an N limited environment. Demonstration of a role of conifer endophytes in host N acquisition, along with the extent and potential significance in the conifer forest N budget will require direct tests for endophytic N2 fixation and N transfer to the conifer host.

2.6 References

Anand, R., and C. P. Chanway. 2013. N2-fixation and growth promotion in cedar colonized by an endophytic strain of Paenibacillus polymyxa. Biol Fert Soils 49:235–239.

Anand, R., S. Grayston, and C. Chanway. 2013. N2-Fixation and Seedling Growth Promotion of Lodgepole Pine by Endophytic Paenibacillus polymyxa. Microbial Ecology 66:369–374.

Ando, S., M. Goto, S. Meunchang, P. Thongra-ar, T. Fujiwara, H. Hayashi, and T. Yoenama. 2005. Detection of nifH squences in sugarcane (Saccharurnomcinarurn L.) and Pineapple (Ananas cornosus [L.] Merr.). Soil Science and Plant Nutrition 51:303–308.

Arnold, A. E., L. C. Mejia, D. Kyllo, E. I. Rojas, Z. Maynard, N. Robbins, and E. A. Herre. 2003. Fungal endophytes limit pathogen damage in a tropical tree. Proc Natl Acad Sci U S A 100:15649–54.

Arshad, M., and W. T. Frankenberger. 1991. Microbial production of plant hormones. Plant Soil 133:1–8.

Bal, A., R. Anand, O. Berge, and C. P. Chanway. 2012. Isolation and identification of diazotrophic bacteria from internal tissues of Pinus contorta and Thuja plicata. Can J Forest Res 42:807–813.

Bal, A., and C. P. Chanway. 2012. Evidence of nitrogen fixation in lodgepole pine inoculated with diazotrophic Paenibacillus polymyxa. Botany 90:891–896.

Berry, D., K. Ben Mahfoudh, M. Wagner, and A. Loy. 2011. Barcoded primers used in multiplex amplicon pyrosequencing bias amplification. Appl Environ Microb 77:7846–9.

! 19!

Binkley, D., S. Yowhan, and D. W. Valentine. 2000. Do Forests Receive Occult Inputs of Nitrogen? Ecosystems 3:321–331.

Blanco, Y., M. Blanch, D. Pinon, M. E. Legaz, and C. Vicente. 2005. Antagonism of Gluconacetobacter diazotrophicus (a sugarcane endosymbiont) against Xanthomonas albilineans (pathogen) studied in alginate-immobilized sugarcane stalk tissues. J Biosci Bioeng 99:366–71.

Boddey, R. M. 1995. Biological nitrogen fixation in sugar cane: a key to energetically viable biofuel production. Crit. Rev. Plant Sci. 14:263–279.

Boddey, R. M., L. G. da Silva, V. M. Reis, B. J. R. Alves, and S. Urquiaga. 1999. Assessment of bacterial nitrogen fixation in grass species. in E. W. Triplett, editor. Nitrogen Fixation in Bacteria: Molecular and Cellular Biology. Horizon Scientific Press, UK.

Boddey, R. M., S. Urquiaga, and V. M. Reis. 2003. Endophytic nitrogen fixation in sugarcane: present knowledge and future applications. Plant and Soil 252:139+149.

Bodenhausen, N., M. W. Horton, and J. Bergelson. 2013. Bacterial communities associated with the leaves and the roots of Arabidopsis thaliana. PLoS ONE 8:e56329.

Bormann, B. T., H. Bormann, W. B. Bowden, R. S. Piece, S. P. Hamburg, D. Wang, M. C. Snyder, C. . Li, and R. C. Ingersoll. 1993. Rapid N2 Fixation in Pines, Alder, and Locust: Evidence From the Sandbox Ecosystems Study. Ecosystems 74:583– 598.

Bormann, B. T., C. K. Keller, D. Wang, and H. Bormann. 2002. Lessons from the Sandbox: Is Unexplained Nitrogen Real? Ecosystems 5:727–733.

Bulgarelli, D., M. Rott, K. Schlaeppi, E. Ver Loren van Themaat, N. Ahmadinejad, F. Assenza, P. Rauf, B. Huettel, R. Reinhardt, E. Schmelzer, J. Peplies, F. O. Gloeckner, R. Amann, T. Eickhorst, and P. Schulze-Lefert. 2012. Revealing structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488:91–5.

Cankar, K., H. Kraigher, M. Ravnikar, and M. Rupnik. 2005. Bacterial endophytes from seeds of Norway spruce (Picea abies L. Karst). FEMS Microbiol Lett 244:341–5.

Caporaso, J. G., K. Bittinger, F. D. Bushman, T. Z. DeSantis, G. L. Andersen, and R. Knight. 2010a. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics 26:266–7.

Caporaso, J. G., J. Kuczynski, J. Stombaugh, K. Bittinger, F. D. Bushman, E. K. Costello, N. Fierer, A. G. Pena, J. K. Goodrich, J. I. Gordon, G. A. Huttley, S. T. Kelley, D.

! 20!

Knights, J. E. Koenig, R. E. Ley, C. A. Lozupone, D. McDonald, B. D. Muegge, M. Pirrung, J. Reeder, J. R. Sevinsky, P. J. Turnbaugh, W. A. Walters, J. Widmann, T. Yatsunenko, J. Zaneveld, and R. Knight. 2010b. QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7:335–6.

Carroll, G. 1988. Fungal endophytes in stems and leaves: from latent pathogen to mutualistic symbiont. Ecology 69:2–9.

Chanway, C., and F. B. Holl. 1991. Biomass increase and associative nitrogen fixation of mycorrhizal Pinus contorta seedlings inoculated with a plant growth promoting Bacillus strain. Can J Bot 69:507–511.

Chapman, W. K., and L. Paul. 2012. Evidence that Northern Pioneering Pines with Tuberculate Mycorrhizae are Unaffected by Varying Soil Nitrogen Levels. Microb Ecol 64:964–972.

Chelius, M. K., and E. W. Triplett. 2001. The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microb Ecol 41:252–263.

Collavino, M. M., H. J. Tripp, I. E. Frank, M. L. Vidoz, P. A. Calderoli, M. Donato, J. P. Zehr, and O. M. Aguilar. 2014. nifH pyrosequencing reveals the potential for location-specific soil chemistry to influence N 2 -fixing community dynamics: N 2 -fixing communities in agricultural soils. Environmental Microbiology:n/a–n/a.

Conn, V. M., A. R. Walker, and C. M. Franco. 2008. Endophytic actinobacteria induce defense pathways in Arabidopsis thaliana. Mol Plant Microbe Interact 21:208–18.

Crotti, E., A. Rizzi, B. Chouaia, I. Ricci, G. Favia, A. Alma, L. Sacchi, K. Bourtzis, M. Mandrioli, A. Cherif, C. Bandi, and D. Daffonchio. 2010. Acetic acid bacteria, newly emerging symbionts of insects. Appl Environ Microbiol 76:6963–70.

DeLuca, T. H., O. Zackrisson, M. C. Nilsson, and A. Sellstedt. 2002. Quantifying nitrogen-fixation in feather moss carpets of boreal forests. Nature 419:917–20.

Delwiche, C. ., P. J. Zinke, C. M. Johnson, and R. A. Virginia. 1979. Nitrogen isotope distribution as a presumptive indicator of nitrogen fixation. Bot Gaz 140:65–69.

DeSantis, T. Z., P. Hugenholtz, N. Larsen, M. Rojas, E. L. Brodie, K. Keller, T. Huber, D. Dalevi, P. Hu, and G. L. Andersen. 2006. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl Environ Microbiol 72:5069–72.

Dong, Z., M. J. Canny, M. E. McCully, M. R. Roboredo, C. F. Cabadilla, E. Ortega, and R. Rodes. 1994. A Nitrogen-Fixing Endophyte of Sugarcane Stems (A New Role for the Apoplast). Plant Physiol 105:1139–1147.

! 21!

Dutta, D., and R. Gachhui. 2006. Novel nitrogen-fixing Acetobacter nitrogenifigens sp. nov., isolated from Kombucha tea. Int J Syst Evol Microbiol 56:1899–903.

Dutta, D., and R. Gachhui. 2007. Nitrogen-fixing and cellulose-producing Gluconacetobacter kombuchae sp. nov., isolated from Kombucha tea. Int J Syst Evol Microbiol 57:353–7.

Edgar, R. C. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–1.

Edgar, R. C., B. J. Haas, J. C. Clemente, C. Quince, and R. Knight. 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27:2194– 200.

Engel, P., and N. A. Moran. 2013. The gut microbiota of insects - diversity in structure and function. FEMS Microbiology Reviews 37:699–735.

Farnelid, H., M. Bentzon-Tilia, A. F. Andersson, S. Bertilsson, G. Jost, M. Labrenz, K. Jürgens, and L. Riemann. 2013. Active nitrogen-fixing heterotrophic bacteria at and below the chemocline of the central Baltic Sea. The ISME Journal 7:1413– 1423.

Favilli, F., and A. Messini. 1990. Nitrogen fixation at phyllospheric level in coniferous plants in Italy. Plant Soil 128:91–95.

Fuentes-Ramirez, L. E., R. Bustillos-Cristales, A. Tapia-Hernandez, T. Jimenez-Salgado, E. T. Wang, E. Martinez-Romero, and J. Caballero-Mellado. 2001. Novel nitrogen-fixing acetic acid bacteria, Gluconacetobacter johannae sp. nov. and Gluconacetobacter azotocaptans sp. nov., associated with coffee plants. Int J Syst Evol Microbiol 51:1305–14.

Gillis, M., K. Kersters, B. Hoste, D. Janssens, R. M. Kroppenstedt, M. P. Stephan, K. R. S. Teixeira, J. Dobereiner, and J. De Ley. 1989. Acetobacter diazotrophicus sp. nov., a Nitrogen-Fixing Acetic Acid Bacterium Associated with Sugarcane. Int J Syst Bacteriol 39:361–364.

Glick, B. R. 2004. Bacterial ACC deaminase and the alleviation of plant stress. Adv Appl Microbiol 56:291–312.

Gottel, N. R., H. F. Castro, M. Kerley, Z. Yang, D. A. Pelletier, M. Podar, T. Karpinets, E. Uberbacher, G. A. Tuskan, R. Vilgalys, M. J. Doktycz, and C. W. Schadt. 2011. Distinct microbial communities within the endosphere and rhizosphere of Populus deltoides roots across contrasting soil types. Appl Environ Microbiol 77:5934–44.

Greenberg, D. E., L. Ding, A. M. Zelazny, F. Stock, A. Wong, V. L. Anderson, G. Miller, D. E. Kleiner, A. R. Tenorio, L. Brinster, D. W. Dorward, P. R. Murray, and S.

! 22!

M. Holland. 2006. A novel bacterium associated with lymphadenitis in a patient with chronic granulomatous disease. PLoS Pathog 2:e28.

Hallman, J., A. Quadt-Hallmann, W. F. Mahaffee, and J. W. Kloepper. 1997. Bacterial endophytes in agricultural crops. Can J Microbiol 43:895–914.

Hanshew, A. S., C. J. Mason, K. F. Raffa, and C. R. Currie. 2013. Minimization of chloroplast contamination in 16S rRNA gene pyrosequencing of insect herbivore bacterial communities. J Microbiol Methods 95:149–55.

Hardoim, P. R., L. S. van Overbeek, and J. D. Elsas. 2008. Properties of bacterial endophytes and their proposed role in plant growth. Trends Microbiol 16:463–71.

Higgins, K. L., P. D. Coley, T. A. Kursar, and A. E. Arnold. 2011. Culturing and direct PCR suggest prevalent host generalism among diverse fungal endophytes of tropical forest grasses. Mycologia 103:247–60.

Hobbie, E. A., S. A. Macko, and M. Williams. 2000. Correlations between foliar δ15N and nitrogen concentrations may indicate plant-mycorrhizal interactions. Oecologia:273–283.

Hurek, T., L. L. Handley, B. Reinhold-Hurek, and Y. Piche. 2002. Azoarcus grass endophytes contribute fixed nitrogen to the plant in an unculturable state. Mol Plant Microbe Interact 15:233–42.

Idris, R., R. Trifonova, M. Puschenreiter, W. W. Wenzel, and A. Sessitsch. 2004. Bacterial communities associated with flowering plants of the Ni hyperaccumulator Thlaspi goesingense. Appl Environ Microbiol 70:2667–77.

Izumi, H., I. C. Anderson, K. Killham, and E. R. Moore. 2008. Diversity of predominant endophytic bacteria in European deciduous and coniferous trees. Can J Microbiol 54:173–9.

Jojima, Y., Y. Mihara, S. Suzuki, K. Yokozeki, S. Yamanaka, and R. Fudou. 2004. Saccharibacter floricola gen. nov., sp. nov., a novel osmophilic acetic acid bacterium isolated from pollen. Int J Syst Evol Microbiol 54:2263–7.

Jones, K., E. King, and M. Eastlick. 1974. Nitrogen Fixation by Free-living Bacteria in the Soil and in the Canopy of Douglas Fir. Ann Bot (Lond) 38:765–762.

Kersters, K., P. Lisdiyanti, K. Komagata, and J. Swings. 2006. The family Acetobacteraceae: the genera Acetobacter, Acidomonas, Asaia, Gluconacetobacter, Gluconobacter, and Kozakia. Pages 163–200 in M. Dworkin, S. Falkow, E. Rosenberg, K. H. Schleifer, and E. Stackebrandt, editors. The prokaryotes. Third edition. Springer, New York, NY.

! 23!

Knoth, J. L., S. H. Kim, G. J. Ettl, and S. L. Doty. 2013. Biological nitrogen fixation and biomass accumulation within poplar clones as a result of inoculations with diazotrophic endophyte consortia. New Phytol.

Lee, S., M. Flores-Encarnación, M. Contreras-Zentella, L. Garcia-Flores, J. E. Escamilla, and C. Kennedy. 2004. Indole-3-acetic acid biosynthesis is deficient in Gluconacetobacter diazotrophicus strains with mutations in cytochrome c biogenesis genes. Journal of bacteriology 186:5384–5391.

Lin, C.-P., J.-P. Huang, C.-S. Wu, C.-Y. Hsu, and S.-M. Chaw. 2010. Comparative Chloroplast Genomics Reveals the Evolution of Pinaceae Genera and Subfamilies. Genome Biology and Evolution 2:504–517.

Lipson, D. A., and S. K. Schmidt. 2004. Seasonal changes in an alpine soil bacterial community in the colorado rocky mountains. Appl Environ Microbiol 70:2867– 79.

Loganathan, P., and S. Nair. 2004. Swaminathania salitolerans gen. nov., sp. nov., a salt- tolerant, nitrogen-fixing and phosphate-solubilizing bacterium from wild rice (Porteresia coarctata Tateoka). Int J Syst Evol Microbiol 54:1185–90.

Lundberg, D. S., S. L. Lebeis, S. H. Paredes, S. Yourstone, J. Gehring, S. Malfatti, J. Tremblay, A. Engelbrektson, V. Kunin, T. G. del Rio, R. C. Edgar, T. Eickhorst, R. E. Ley, P. Hugenholtz, S. G. Tringe, and J. L. Dangl. 2012. Defining the core Arabidopsis thaliana root microbiome. Nature 488:86–90.

Martinson, V. G., B. N. Danforth, R. L. Minckley, O. Rueppell, S. Tingek, and N. A. Moran. 2011. A simple and distinctive microbiota associated with honey bees and bumble bees. Mol Ecol 20:619–28.

Mayak, S., T. Tirosh, and B. R. Glick. 2004. Plant growth-promoting bacteria confer resistance in tomato plants to salt stress. Plant Physiol Biochem 42:565–72.

Morford, S. L., B. Z. Houlton, and R. A. Dahlgren. 2011. Increased forest ecosystem carbon and nitrogen storage from nitrogen rich bedrock. Nature 477:78–81.

Muthukumarasamy, R., I. Cleenwerck, G. Revathi, M. Vadivelu, D. Janssens, B. Hoste, K. U. Gum, K. D. Park, C. Y. Son, T. Sa, and J. Caballero-Mellado. 2005. Natural association of Gluconacetobacter diazotrophicus and diazotrophic Acetobacter peroxydans with wetland rice. Syst Appl Microbiol 28:277–86.

Näsholm, T., K. Kielland, and U. Ganeteg. 2009. Uptake of organic nitrogen by plants. New Phytologist 182:31–48.

Oksanen, J., Kindt, P., Legendre, B., O’Hara, B., Simpson, G.L., and Stevenson, M.H.H. 2008. vegan: community ecology package, R package version 1.11-4.

! 24!

Ping, L., and W. Boland. 2004. Signals from the underground: bacterial volatiles promote growth in Arabidopsis. Trends Plant Sci 9:263–6.

Pirttilä, A., and A. C. Frank. 2011. Endophytes of forest trees: Biology and Application. Springer.

Pirttilä, A. M., P. Joensuu, H. Pospiech, J. Jalonen, and A. Hohtola. 2004. Bud endophytes of Scots pine produce adenine derivatives and other compounds that affect morphology and mitigate browning of callus cultures. Physiol Plant 121:305–312.

Pirttilä, A. M., H. Laukkanen, H. Pospiech, R. Myllyla, and A. Hohtola. 2000. Detection of intracellular bacteria in the buds of Scotch pine (Pinus sylvestris L.) by in situ hybridization. Appl Environ Microbiol 66:3073–7.

Pirttilä, A. M., H. Pospiech, H. Laukkanen, R. Myllyla, and A. Hohtola. 2005. Seasonal variations in location and population structure of endophytes in buds of Scots pine. Tree Physiol 25:289–97.

Poole, A. M., D. B. Stouffer, and J. M. Tylianakis. 2012. “Ecosystomics”: ecology by sequencer. Trends Ecol Evol 27:309–10.

Price, M. N., P. S. Dehal, and A. P. Arkin. 2009. FastTree: computing large minimum evolution trees with profiles instead of a distance matrix. Mol Biol Evol 26:1641– 50.

Qin, J., R. Li, J. Raes, M. Arumugam, K. S. Burgdorf, C. Manichanh, T. Nielsen, N. Pons, F. Levenez, T. Yamada, D. R. Mende, J. Li, J. Xu, S. Li, D. Li, J. Cao, B. Wang, H. Liang, H. Zheng, Y. Xie, J. Tap, P. Lepage, M. Bertalan, J. M. Batto, T. Hansen, D. Le Paslier, A. Linneberg, H. B. Nielsen, E. Pelletier, P. Renault, T. Sicheritz-Ponten, K. Turner, H. Zhu, C. Yu, S. Li, M. Jian, Y. Zhou, Y. Li, X. Zhang, S. Li, N. Qin, H. Yang, J. Wang, S. Brunak, J. Dore, F. Guarner, K. Kristiansen, O. Pedersen, J. Parkhill, J. Weissenbach, P. Bork, S. D. Ehrlich, and J. Wang. 2010. A human gut microbial gene catalogue established by metagenomic sequencing. Nature 464:59–65.

Rastogi, G., A. Sbodio, J. J. Tech, T. V. Suslow, G. L. Coaker, and J. H. Leveau. 2012. Leaf microbiota in an agroecosystem: spatiotemporal variation in bacterial community composition on field-grown lettuce. ISME J 6:1812–22.

Redford, A. J., R. M. Bowers, R. Knight, Y. Linhart, and N. Fierer. 2010. The ecology of the phyllosphere: geographic and phylogenetic variability in the distribution of bacteria on tree leaves. Environ Microbiol 12:2885–93.

! 25!

Reed, S. C., A. R. Townsend, C. C. Cleveland, and D. Nemergut. 2010. Microbial community shifts influence patterns in tropical forest nitrogen fixation. Oecologia 264:521–531.

Reiter, B., H. Bürgmann, K. Burg, and A. Sessitsch. 2003. Endophytic nifH gene diversity in African sweet potato. Canadian journal of microbiology 49:549–555.

Richards, B. N. 1964. Fixation of atmospheric nitrogen in coniferous forests. Aust For 28:68–74.

Roh, S. W., Y. D. Nam, H. W. Chang, K. H. Kim, M. S. Kim, J. H. Ryu, S. H. Kim, W. J. Lee, and J. W. Bae. 2008. Phylogenetic characterization of two novel commensal bacteria involved with innate immune homeostasis in Drosophila melanogaster. Appl Environ Microbiol 74:6171–7.

Rout, M. E., T. H. Chrzanowski, T. K. Westlie, T. H. Deluca, R. M. Callaway, and W. E. Holben. 2013. Bacterial endophytes enhance competition by invasive plants. Am J Bot 100:1726–37.

Rueth, H. M., and J. S. Baron. 2002. Differences in Englemann Spruce Forest Biogeochemistry East and West of the Continental Divide in Colorado, USA. Ecosystems 5:45–57.

Schlaeppi, K., N. Dombrowski, R. G. Oter, E. Ver Loren van Themaat, and P. Schulze- Lefert. 2014. Quantitative divergence of the bacterial root microbiota in Arabidopsis thaliana relatives. Proceedings of the National Academy of Sciences 111:585–592.

Schoettle, A. W., and S. G. Rochelle. 2000. Morphological variation of Pinus flexilis (Pinaceae), a bird-dispersed pine, across a range of elevations. Am J Bot 87:1797–806.

Shade, A., and J. Handelsman. 2012. Beyond the Venn diagram: the hunt for a core microbiome. Environ Microbiol 14:4–12.

Sharp, K. H., D. Distel, and V. J. Paul. 2012. Diversity and dynamics of bacterial communities in early life stages of the Caribbean coral Porites astreoides. ISME J 6:790–801.

Smith, S., and D. Read. 2008. Mycorrhizal Symbiosis. Academic Press, London.

Stamatakis, A., T. Ludwig, and H. Meier. 2005. RAxML-III: a fast program for maximum likelihood-based inference of large phylogenetic trees. Bioinformatics 21:456–63.

! 26!

Ueda, T., Y. Suga, N. Yahiro, and T. Matsuguchi. 1995. Remarkable N2-fixing bacterial diversity detected in rice roots by molecular evolutionary analysis of nifH gene sequences. Journal of bacteriology 177:1414–1417.

Verma, S. C., J. K. Ladha, and A. K. Tripathi. 2001. Evaluation of plant growth promoting and colonization ability of endophytic diazotrophs from deep water rice. Journal of biotechnology 91:127–141.

Virginia, R. A., and C. . Delwiche. 1982. Natural 15N abundance of presumed N2-fixing plants from selected ecosystems. Oecologia 54:317–325.

Vitousek, P. M., D. N. Menge, S. C. Reed, and C. C. Cleveland. 2013. Biological nitrogen fixation: rates, patterns and ecological controls in terrestrial ecosystems. Philos Trans R Soc Lond B Biol Sci 368:20130119.

Wakelin, S. A., R. A. Warren, P. R. Harvey, and M. H. Ryder. 2004. Phosphate solubilization by Penicillium spp. closely associated with wheat roots. Biology and Fertility of Soils 40:36–43.

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 73:5261–7.

Woebken, D., L. Prufert-Bebout, L. Burow, B. Bebout, T. Hoehler, J. Pett-Ridge, A. M. Sporman, P. K. Weber, and S. W. Singer. 2012. Identification of a previously unknown cyanobacterial group as active diazotrophs in coastal microbial mats using NanoSIMS. Isme J 6:1427–1439.

Yukphan, P., T. Malimas, Y. Muramatsu, M. Takahashi, M. Kaneyasu, W. Potacharoen, S. Tanasupawat, Y. Nakagawa, K. Hamana, Y. Tahara, K. Suzuki, M. Tanticharoen, and Y. Yamada. 2009. Ameyamaea chiangmaiensis gen. nov., sp. nov., an acetic acid bacterium in the alpha-Proteobacteria. Biosci Biotechnol Biochem 73:2156–62.

Yukphan, P., T. Malimas, Y. Muramatsu, M. Takahashi, M. Kaneyasu, S. Tanasupawat, Y. Nakagawa, K. Suzuki, W. Potacharoen, and Y. Yamada. 2008. Tanticharoenia sakaeratensis gen. nov., sp. nov., a new osmotolerant acetic acid bacterium in the alpha-Proteobacteria. Biosci Biotechnol Biochem 72:672–6.

Yukphan, P., T. Malimas, W. Potacharoen, S. Tanasupawat, M. Tanticharoen, and Y. Yamada. 2005. Neoasaia chiangmaiensis gen. nov., sp. nov., a novel osmotolerant acetic acid bacterium in the alpha-Proteobacteria. J Gen Appl Microbiol 51:301– 11.

! 27!

2.7 Tables

Table 2-1 Samples successfully characterized by 16S rRNA in this study, along with the number of sequences after sequence quality control and removal of plant DNA

Sampling No. of No. of 97% Tree species Tree ID Sample location Sequences phylotypes 01 WE.01 1736 141 WE.02 1838 131 P. flexilis WE.03 1826 138 WE1 02 WE.04 1212 162 03 WE.05 1176 128 P. engelmannii 04 WE.11 1668 125 05 WE.12 2098 180 WE.06 1750 114 06 WE.07 1804 106 P. flexilis WE.08 1809 117 07 WE.09 1602 146 WE2 08 WE.10 875 101 WE.13 1709 157 P. engelmannii 09 WE.14 1055 134 WE.15 1254 164 10 WE.16 1287 150 11 TL.01 1199 31 P. flexilis 12 TL.02 1332 41 TL1 P. engelmannii 13 TL.11 865 58 14 TL.12 1202 80

! 28!

2.8 Figures

Warm Edge

Treeline

Observed Species

Sequences Per Sample

! 29!

Acidobacteria 1 Actinobacteria Armatimonadetes Bacteroidetes 0.8 Chlamydiae Firmicutes Proteobacteria (Alpha) Proteobacteria (Beta) 0.6 Proteobacteria (Delta) Proteobacteria (Gamma) Other 0.4

0.2

TL.12TL.11WE.16WE.15WE.14WE.13WE.12WE.11TL.02TL.01WE.10WE.09WE.08WE.07WE.06WE.05WE.04WE.03WE.02WE.01 P. engelmannii P. flexilis

Figure 2-2 Relative abundances of various major bacterial phyla and classes recovered from P. flexilis and P. engelmannii needles. Relative abundance of phyla (and classes of the Proteobacteria) was calculated as the percentage of sequences belonging to a particular lineage of all 16S rRNA gene sequences recovered from each sample.

! 30!

NR 041442.1 Methylobacterium persicinum PT 1507 91 PT 1325 NR 025596.1 Methylocella tundrae 87 NR 024997.1 Sphingomonas aquatilis 66 96 PT 1348 77 PT 604 NR 026338.1 Sphingomonas trueperi 67 NR 074264.1 Caulobacter sp. 70 PT 672 93 NR 043726.1 Brevundimonas terrae PT 469 PT 267 99 PT 1016 PT 805 54 PT 1516 PT 1290 PT 1644 NR 041564.1 Asaia lannaensis NR 024773.1 Kozakia baliensis NR 042706.1 Acidisoma sibirica 59 PT 295 NR 074276.1 Granulibacter bethesdensis NR 026132.1 Gluconacetobacter liquefaciens 94 NR 074292.1 Gluconacetobacter diazotrophicus PT 123 53 PT 945 Methylobacteriaceae PT 1045 PT 1324 Beijerinckiaceae 94 PT 1068 Sphingomonadaceae NR 041601.1 Tanticharoenia sakaeratensis PT 1277 Caulobacteraceae 64 71 PT 1622 Acetobacteraceae PT 849 NR 042634.1 Burkholderia arboris Scale: 0.1

Figure 2-3 Phylogeny of major Alphaproteobacterial sequences in our samples. Maximum likelihood phylogeny of Alphaproteobacterial sequences that occurred at least 100 times along with the three most closely related sequences from GenBank 16S rRNA database (accession number indicated). Because our sequences are short (approximately 300 nt), many of our clades have low bootstrap support. Here, only bootstrap values above 50% are displayed. The tree is rooted with Burkholderia arboris.

! 31!

A %

TL.01 TL.02 PT GenBank# ID Match Phylum/class/family WE.01 WE.02 WE.03 WE.04 WE.05 WE.06 WE.07 WE.08 WE.09 WE.10 3.3 2.7 2.4 2.1 3.5 3.0 2.7 2.1 2.7 4.0 0.0 0.0 123 NR_074284.1 98 Gluconacetobacter diazotrophicus Proto/Alpha/Acetobac 1.4 1.4 2.0 2.2 0.4 3.7 4.7 1.6 1.7 0.1 0.6 0.5 816 NR_042813.1 99 Pandoraea pnomenusa Proto/Beta/Burkhold 3.2 2.0 1.7 0.7 1.5 1.0 1.1 1.2 2.1 1.5 0.0 0.0 849 NR_024959.1 95 Gluconacetobacter johannae Proto/Alpha/Acetobac 1.9 0.7 0.4 0.7 0.6 1.2 1.1 0.4 0.7 0.3 8.4 7.7 1016 NR_024773.1 96 Kozakia baliensis Proto/Alpha/Acetobac 36.0 34.4 22.5 19.1 30.1 33.5 30.5 23.4 29.5 24.1 53.2 48.9 1045 NR_074292.1 97 Gluconacetobacter diazotrophicus Proto/Alpha/Acetobac 7.4 6.3 5.5 3.5 5.6 4.0 3.2 5.6 4.2 4.0 0.2 0.2 1068 NR_041601.1 98 Tanticharoenia sakaeratensis Proto/Alpha/Acetobac 0.9 1.2 2.7 4.0 3.1 2.5 1.2 3.8 1.1 1.6 0.0 0.1 1257 NR_074294.1 95 Terriglobus saanensis Proto/Acidobac/Acidobac 0.3 0.6 2.6 2.2 1.6 0.8 1.6 2.1 2.5 3.5 0.0 0.0 1507 NR_041442.1 93 Methylobacterium persicinum Proto/Alpha/Methylobac 10.0 16.5 11.0 11.2 9.3 11.5 11.9 7.9 17.5 22.1 2.8 2.7 1516 NR_026132.1 95 Gluconacetobacter liquefaciens Proto/Alpha/Acetobac 0.3 0.5 0.4 0.5 1.4 0.7 0.5 1.1 0.7 0.5 26.9 24.3 1644 NR_041564.1 97 Asaia lannaensis Proto/Alpha/Acetobac

B 1 1 1

1 % TL. TL.12 Match Phylum/class/order WE. PT GenBank# ID WE.12 WE.13 WE.14 WE.15 WE.16

3.8 4.9 2.2 2.1 1.5 2.0 0.0 0.0 29 NR_074294.1 93 Terriglobus saanensis Proto/Acidobac/Acidobac 4.3 5.0 3.0 4.4 1.4 0.8 0.0 0.1 603 NR_029359.1 96 Hymenobacter roseosalivarius Proto/Bacteroid/Cytophag 2.3 3.3 2.5 2.5 2.4 2.4 0.0 0.3 604 NR_026338.1 98 Sphingomonas trueperi Proto/Alpha/Sphingomonad 1.7 2.3 1.3 0.9 0.7 0.9 4.6 13.1 816 NR_042813.1 99 Pandoraea pnomenusa Proto/Beta/Burkhold 15.6 18.7 14.7 14.1 18.2 17.5 37.5 39.0 1045 NR_074292.1 97 Gluconacetobacter diazotrophicus Proto/Alpha/Acetobac 2.5 8.5 4.9 5.1 4.6 4.4 0.1 0.7 1295 NR_043918.1 98 Terriglobus roseus Proto/Acidobac/Acidobac 2.2 3.7 3.5 3.5 4.7 5.3 0.1 0.0 1324 NR_024773.1 95 Kozakia baliensis Proto/Acidobac/Acetobac 10.8 9.7 9.5 7.7 6.7 5.1 0.0 0.2 1507 NR_041442.1 93 Methylobacterium persicinum Proto/Alpha/Methylobac 11.0 16.2 5.4 6.4 5.6 5.2 2.9 1.5 1516 NR_026132.1 95 Gluconacetobacter liquefaciens Proto/Alpha/Acetobac 0.2 0.7 0.4 0.2 0.6 0.7 18.6 1.7 1644 NR_041564.1 97 Asaia lannaensis Proto/Alpha/Acetobac

Figure 2-4 Heatmap showing the 10 most dominant phylotypes and their average relative abundances as percentages of all sample 16S rRNA gene sequences recovered in our conifer needles samples. (A) P. flexilis, (B) P. engelmannii. Color tones range from cool (blue) to warm (red) to indicate the lowest to highest relative abundance values. Phylotypes were considered dominant if they were both highly abundant and occurred frequently in samples of a given conifer species. Abbreviations: PT=Phylotype,Acetobac=Acetobacteraceae, Acidobac=Acidobacteria/Acidobacteriaceae, Burkhold=Burkholderiaceae, Bacteriod=Bacteriodetes, Flexibacteraceae, Methylobac=Methylobacteriaceae, Cytophag=Cytophagacae, Sphingomonad=Sphingomonadaceae.

! 32!

Phylotype Match % identity 123 Gluconacetobacter diazotrophicus 98* 267 Caulobacter sp. 97 295 Acidisphaera rubrifaciens 95 333 Granulicella tundricola 98 663 Terriglobus roseus 99 669 Roseatele terrae 98 805 Tanticharoenia sakaeratensis 93 816 Pandoraea pnomenusa 99* 1016 Kozakia baliensis 96* 1045 Gluconacetobacter diazotrophicus 97* 1067 Terriglobus saanensis 97 1068 Tanticharoenia sakaeratensis 98* 1290 Tanticharoenia sakaeratensis 95 1295 Terriglobus roseus 98* 1433 Geobacter metallireducens 89 1516 Gluconacetobacter liquefaciens 95* 1644 Asaia lannaensis 97* 1813 Herbaspirillum rhizosphaerae 99

Figure 2-5 Shared and host species-specific phylotypes. Blue: phylotypes found in all P. engelmannii samples but not in any of the P. flexilis samples. Red: phylotypes found in all P. flexilis samples but not in any of the P. engelmannii samples. Purple: indicates phylotypes recovered from all samples (i.e. both species). An asterisk indicates that the phylotype is included in Figure 2.

! 33!

Species B Elevation A D TL.12 TL.11 TL.02 TL.01 WE.05 WE.08 WE.02 TL WE.06 WE.09 WE WE.01 Location WE.03 C WE.04 WE.07 WE.10 x xxxx WE.15 WE.16 WE.11 x WE.13 TL xx P. flexilis PC2: 16% WE.14 WE1 x P. engelmannii WE.12 PC1: 19% x WE2 0.04

Figure 2-6 PCoA and UniFrac analysis of the bacterial communities associated with conifer needles. (A-C) PCoA of the unweighted UniFrac distance matrix. Points that are closer together on the ordination have communities that are more similar. Each point corresponds to a sample, and shapes correspond to (A) host species, (B) elevation, and (C) location. (D) Hierarchical clustering of composite communities of the conifer species. Leaves are labeled by color according to host species: red, P. flexilis; blue, P. engelmannii.

! !

3 The association between Pinaceae and an acetic

acid bacterial needle endophyte is persistent

across host species and geographic locations

3.1 Abstract

A recent culture-independent 16S rRNA survey revealed that needles of Pinus flexilis (limber pine) and Picea engelmannii (Engelmann spruce) sampled at Niwot Ridge, CO were both consistently dominated by a single phylotype 97% similar to a dinitrogen (N2) fixing endophytic Gluconacetobacter. To test if the shared microbiota reflects shared environment (at Niwot Ridge) or shared ancestry (of Pinus and Picea), we characterized the needle endophyte community in P. flexilis and the co-occuring Pinus contorta (lodgepole pine) at Niwot Ridge, CO, and at Horseshoe Meadows, CA, an outlier site in the P. flexilis distribution. All samples —from both sites and species—were dominated by the Gluconacetobacter phylotype. The rest of the endophytic community varied across sites and species. The observed pattern is most consistent with a long-standing association between trees in the pine family and acetic acid bacterial endophytes, predating the divergence of the genera Picea and Pinus, and with a remaining, rare endophyte community that is variable and structured by location and host species.

3.2 Introduction

Bacterial symbiotic organisms found within healthy plant tissue have been shown to confer tolerance of harsh environments by extending host access to limited nutrients. Despite these documented roles, the diversity and distribution of bacterial endophytes in natural plant populations remain relatively unexplored. A recent non-culture-based 16S rRNA survey revealed that needles of Pinus flexilis (limber pine) and Picea engelmannii (Engelmann spruce) growing above 3000 m at Niwot Ridge, CO were consistently dominated by a single phylotype that made up 14-53 % of the sequences in each sample (Carrell and Frank 2014). The 300-nucleotide-long phylotype (here called AAB1 for

34! 35!

acetic acid bacteria 1) was 97% similar to an N2 fixing endophytic Gluconacetobacter spp. isolated from sugarcane and other crops, perhaps indicating that a N2 fixation underlies the observed association. A slightly higher relative abundance of AAB1 in trees growing in the exposed and N-limited treeline environment than at lower elevations indicates that environmental conditions might influence the association (Carrell and Frank 2014). Interestingly, AAB1 dominated the needle endophyte community of both P. flexilis and P. engelmannii growing in the subalpine environment at Niwot Ridge. This could be due to the shared environment; local adaptation between high elevation pines at Niwot Ridge and a broad host range endophyte could lead to a shared microbiota. If true, we would expect other pine species at Niwot Ridge to host the same dominant bacterial taxon in their needles. Moreover, if the host-endophyte relationship is shaped by local, environment-driven adaptation, we would expect P. flexilis in other parts of its range to host different dominant phylotypes, and we would expect those to be shared with co-occurring species. Alternatively, the shared dominant phylotype could reflect shared ancestry; it is possible that there is a coevolved symbiotic relationship with AAB1 that predates the divergence of Pinus and Picea species ~140 million years ago (MYA). We would then expect to see Pinus and Picea species to host AAB1 throughout their ranges, perhaps with ecosystem properties influencing the relative abundance or presence of the phylotype if the association is facultative.

Many animals have evolved persistent and highly specialized associations with bacteria that provide some kind of benefit to the host (Peek et al. 1998, Taylor et al. 2005, Stewart et al. 2008, Webster and Taylor 2011). Among the most commonly reported are nutritional symbioses between insects and bacteria that complement the host’s low-nutrient diet (Dale and Moran 2006, Moran 2006, Martinson et al. 2011). Similar persistent associations between plants and endophytic bacteria have not been reported. Instead, previously studied endophyte communities appear to fulfill environment-specific host needs without specific bacterial taxa. Studies of Arabidopsis roots show conservation of the microbiota across individuals and sites, but at higher taxonomic rank (Bulgarelli et al. 2012, Lundberg et al. 2012). A recent study showed some co-diversification between plants and endophyte communities, but there was no evidence of coevolution between Arabidopsis species and conserved, dominating phylotypes (Schlaeppi et al. 2014).

The closest example of persistent, coevolving symbioses in plants are perhaps found in leaf nodulated angiosperm species (e.g. Pavetta and Psychotria). These plants are consistently associated with single bacterial N2-fixing partners in a specific manner; however, different plant species often do not host related symbionts, suggesting intermittent interaction between plants and N2-fixing bacteria (Lemaire et al. 2011). Gunnera species are also persistently found associated with N2-fixing nodulating Nostoc, but with a low specificity in the association and high cyanobacterial diversity across geographic locations and even within single plants and nodules (Nilsson et al. 2000). Similarly, legumes are consistently associated with root nodulating Rhizobia but with low specificity of host legumes and Rhizobial strains suggesting frequent horizontal transfer

! 36! of Rhizobia (Martínez-Romero 2009). Given the paucity of strictly co-evolved associations in plants, the consistent association of pine trees with endophytes in the Acetobacteraceae is intriguing and requires further investigation.

In this current study, we test if the observed pattern, where P. flexilis and P. engelmannii at Niwot Ridge are both dominated by the AAB1 phylotype, is most consistent with shared environment or shared ancestry. Using Illumina sequencing of the 16S rRNA gene, we characterized the needle endophyte community in Pinus contorta (lodgepole pine), an additional species that co-occurs with P. flexilis at Niwot Ridge, CO. We also characterized P. flexilis and P. contorta growing at similar elevation in the Eastern Sierra Nevada, CA, an outlier of the general P. flexilis distribution.

3.3 Methods

3.3.1 Sample collection and sterilization

Needles were collected from 20 trees at two sites: Horseshoe Meadow, CA (HS), and Niwot Ridge, CO (NR). At each site, five P. flexilis and five co-occurring tree species (P. contorta) were sampled. From each tree, approximately 10 g of needles were removed aseptically from all sides of the tree, composited into a sterile bag, and transported on ice to University of California, Merced for immediate sterilization. Sterilization was achieved by submersion in 30% hydrogen peroxide for three minutes, rinsed three times by shaking with sterile deionized water for one minute, and stored at –20°C. Sterility was confirmed by negative PCR amplification of the final rinse.

3.3.2 DNA extraction

Needles were ground to a fine powder in a sterile mortar in the presence of liquid nitrogen. DNA was extracted using a CTAB extraction method as described by Carrell and Frank, 2014. Briefly, 800ul of CTAB solution (1 ml CTAB buffer, 0.04 g of polyvinylpyrollidone, 5 µl of 2-mercaptoethanol) was added to 0.6g of tissue, incubated for 2 hours at 60°C, and homogenized with glass beads for 3 minutes. Proteins were removed with the addition of an equal volume of chloroform, centrifuged for 10 minutes at 16 rcf, and the top aqueous phase was place in a sterile tube. Nucleic acids were precipitated with the addition of 1/10 volume of cold 3 M sodium acetate and 1/2 volume cold isopropanol, placed in a –20°C freezer for 12 hours, followed by centrifugation for 30 minutes at 16 rcf. The supernatent was removed, 700 µl of 70% ethanol added, and centrifuged for 10 minutes. The air-dried pellet was resuspended with 30 µl of DNA resuspension fluid (1.0 M Tris-HCL, 0.1 M EDTA) and stored at –20°C.

! 37!

3.3.3 DNA Amplification

The 16S rRNA genes of endophytic bacterial were amplified using nested PCR to reduce the occurrence of plastid sequences (Chelius and Triplett 2001a) and improve consistency. Plant DNA amplification was suppressed with the primer pair 16S 799f (AACMGGATTAGATACCCKG) and 16S 1492r (TACGGHTACCTTGTTACGACT) in the first PCR reaction (PCR1). Amplification with 16S 799f and 16S 1492r result in mitochondrial amplicon of about 1000 bp and bacterial amplicon of about 750 bp (Chelius and Triplett, 2001a). In the final round of PCR (PCR2), an appropriate amplicon length for Illumina sequences was achieved from PCR1 amplicons with the Illumina adapted, Golay-barcoded primer pair 799f and 1115r (AGGGTTGCGCTCGTTG), an optimized primer set for phylogenetic analysis of short reads ( Redford et al., 2010). We reduced primer bias by decreasing the number of cycles (Jiao et al. 2006) with the following thermocycle profile used for PCR1 and PCR2: one cycle of 3 min at 95°C; 20 cycles of 40 s at 95°C, 40 s at 50°C, 1.5 min at 72°C; and a final 10 min of elongation at 72°C. The 50 µl PCR1 reaction contained 5 ul of DNA extract, 20 µL 5 PRIME Hot Master Mix (5 PRIME, Inc.), 0.5 µg/µl Bovine Serum Albumin (Thermo Scientific), 21.5 µL PCR grade water (Fischer BioReagents), and 0.2 µM of forward and reverse primers. The 25 µL PCR2 reaction contained 3 µL of PCR1 product, 10 µL 5 PRIME Hot Master Mix, 0.5 µg/µl Bovine Serum Albumin (Thermo Scientific), 8.75µL PCR grade water (Fischer BioReagents), and 0.2 µM of forward and reverse primers. PCR2 amplicons were then cleaned, pooled, and gel extracted (QIAquick Gel Extraction Kit) to ensure selection of the correct band size and to remove most mitochondrial products. Pooled samples were then submitted for Illumina sequencing at the University of California, Davis Genome Center.

3.3.4 Phylotype generation and classification

Sequences were analyzed and processed using the QIIME (1.8.0) package (Caporaso et al. 2010). Forward and reverse paired-end reads were joined with fastq-join, with the barcode filtered from the dataset if the forward and reverse read did not overlap (Aronesty 2011). Joined pair-end reads were quality filtered with QIIME defaults settings (maximum number of consecutive low quality base calls of 3 bases; minimum number of consecutive high quality base calls as a fraction of the input read length of 0.75 total read length; maximum unacceptable Phred quality score of 3; no N characters) which have been found to sufficient for community analysis (Bokulich et al. 2012). The remaining sequences were clustered into phylotypes using denovo operational taxonomic unit (OTU) picking, reads were clustered against each other at 97% similarity using uclust 1.2.22q (Edgar 2010). A representative sequence was chosen for each phylotype by selecting the longest sequence that had the highest number of hits to other sequences of that particular phylotype. Representative sequences were aligned using PyNAST (Caporaso et al. 2010) against the Greengenes core set (DeSantis et al. 2006). Taxonomic assignments were made using uclust with greengenes representative set of sequences as reference. Sequences classified as “Chloroplast” (0.5%) or “Mitochondria” (10%) were removed from the alignment. An approximately maximum-likelihood tree was

! 38! constructed from the aligned of representative sequences, using FastTree (Price et al. 2009). A log-transformed heatmap of the top 10 OTUs from each sample was created with QIIME. P. flexilis and P. engelmannii foliage from Niwot Ridge sampled in 2009 was found to be heavily dominated by an OTU that we here refer to as AAB1 and most samples contained AAB2. We aligned AAB1 and AAB2 to our top OTUs from this study to determine the similarity between sequences.

3.3.5 Community Analysis

For all community analyses, OTUs were rarified to an even sampling depth of 10,575 sequences. We measured α diversity using various indices of diversity for each sample using QIIME. We calculated Chao1 that measures species richness, the Shannon diversity index that integrates richness and evenness, and Faith's phylogenetic diversity that calculates the sum of branch lengths of OTUs from each sample. Diversity measurements were compared between locations using analysis of variance (ANOVA, α=0.05). Student's t-tests were used to compare α diversity between species (all P. flexilis vs all P. contorta), and to compare within locations (P. flexilis vs P. contorta within a location).

We also measured β diversity by calculating phylogenetic dissimilarity of samples from the approximate maximum-likelihood tree with an unweighted UniFrac distance matrix using QIIME. The unweighted Unifraq distances were visualized using principal coordinate analysis (PCoA). ANOSIM and Permutational Multivariate Analysis of Variance (PERMANOVA), each with 999 permutations, were used to calculate the significance of clustering of samples by location or tree species. Students t-test, ANOVA, ANOSIM, and PERMANOVA measurements were calculated in R.

3.4 Results

3.4.1 Phylotypes recovered

Foliage was sampled from ten P. flexilis and ten P. contorta individuals across two locations to assess the endophytic bacterial communities using Illumina sequencing technology. The 16S rRNA gene sequencing produced a total of 5,620,675 reads before quality control. After filtering for quality and length, 5,371,571 total reads remained. Removal of mitochondrial and plastid DNA resulted in 5,054,430 sequences, representing a total of 2092 operational taxonomic units (OTUs) at the 97% similarity level. The number of sequences varied across all samples with an average of 252,721 sequences and between 10,575 and 755,640 reads per sample. A rarefaction curve was generated in QIIME at the 97% similarity OTU level and rarified to 10,575 sequences per sample (Figure 3-1). We recovered more OTUs from Niwot Ridge samples than from Horseshoe samples (p = 0.001). The amount of OTUs recovered did not vary within locations or between species (Table 3-1).

! 39!

3.4.2 Dominant bacterial community members

All of our samples were dominated by the same phyla: Proteobacteria (71-88%), Bacteroidetes (4-6%), Actinobateria (2-9%), Acidobacteria (2-4%) and Firmicutes (1- 13%). Horseshoe Meadow samples had a higher abundance of Firmicutes (9-13%) than Niwot Ridge samples (1%) (Figure 3-2). Within Proteobacteria, bacterial communities were dominated by Alphaproteobacteria followed by Betaproteobacteria and Gammaproteobacteria (Figure 3-3). At the family level, all of our samples were dominated by Acetobacteraceae (>45% of total sequences). This family accounted for 67% of sequences in P. flexilis and 65% of sequences in P. contorta from Niwot Ridge (Figure 3-4). Horseshoe Meadow samples of both species were dominated to a lesser degree by Acetobacteracae, with a relative abundance of 48% in P. flexilis and 43% in P. contorta. Other families dominating all of our samples were Acidobacteriaceae, Sphingobacteriaceae, Caulobacteraceae, Methylobacterium, and Sphingomondaceae. Streptococcus was only an abundant family in samples from Horseshoe Meadow.

A heatmap was generated in QIIME from the ten OTUs that occurred the most frequent in each sample (Figure 3-4). The heatmap contained 24 OTUs and accounts for between 40 and 76% of total sequences per sample. The two most abundant OTUs belonged to the genus Gluconacetobacter of the family Acetobacteraceae and totaled 49% of all sequences. In Niwot Ridge samples, the top two OTUs accounted for 52% of P. contorta and 48% of P. flexilis bacterial sequences. The top two OTUs accounted for 38% of sequences in P. contorta and 41% of sequences in P. flexilis in Horseshoe Meadow samples. We used ClustalW to align the dominant Gluconacetobacter endophytic OTU detected in trees at Niwot Ridge in 2009 (Carrell and Frank 2014) with the two most abundant Acetobacteraceae OTUs from this study and found that AAB1 was 100% similar with our most dominant OTU. Furthermore, our second dominant OTU, which was 95% similar to Gluconacetobacter liquefaciens, was 100% similar to a common but not consistent OTU from Niwot Ridge in 2009 (here called AAB2 for acetic acid bacteria 2) (Carrell and Frank 2014).

Other genera that occurred in high frequency across most samples were Acidisoma, Methylobacterium and Mucilaginibacter. Staphylococcus was abundant in some Horseshoe Meadow samples but not in Niwot Ridge samples. Terriglobus had a high abundance in a single P. flexilis sample of Niwot Ridge but was not abundant in any other samples.

3.4.3 Community Diversity and Structure

Chao1 (a measure of species richness), PD (a measure of phylogenetic diversity within a sample) and Shannon Index (a measure of evenness) were used to assess diversity and richness of the endophytic bacterial communities. Chao1 (Figure 3-5), PD (Figure 3-6), and Shannon index (Figure 3-7) values did not vary within or across tree species, or within location (Table 3-1). Chao1, PD, and the Shannon index were higher in Niwot Ridge than Horseshoe Meadow suggesting diversity and richness varied by location.

! 40!

Unifrac distances were compared from each sample to determine the phylogenetic distances between all samples. Unifrac distance did not significantly differ by tree species (Figure 3-8) but differed by geographic location (Figure 3-9) (Table 3-2). Within a location, unifrac distances were significantly different between P. contorta and P. flexilis. This suggests that the endophytic communities were structured primarily by location and secondarily by tree species (Figure 3-10). Location (Anosim/Permanova; R = 0.5409, P-value = 0.002/Pseudo-F = 2.7205, P-value = 0.005) and then tree species (Anosim/Permanova; R = 0.1777, P-value = 0.019/Pseudo-F = 1.4493, P-value = 0.040) were responsible for community structure with the exclusion of the dominating Gluconacterobacter phylotypes as well.

3.5 Discussion

Many animals form associations with bacteria that persist over millions of years and result in coevolution and co-speciation between host and symbiont (Moran 2006). In insects, specialized relationships with bacteria often allow animals to exploit nutrient-limited environments by providing essential nutrients such as amino acids (Moran 2007). To our knowledge, similarly long-standing associations have not been reported for plants and associated bacteria. Our previous finding that P. flexilis and P. engelmannii growing in the subalpine environment at Niwot Ridge are both dominated by the same, potentially N2 fixing phylotype, prompted us to investigate if coevolution may have occurred between pines and bacterial endophytes within their needles, or alternatively, if the shared microbiota reflects the shared environment. To do this, we characterized needle endophyte communities from P. contorta, another species that co-occurs with P. flexilis at Niwot Ridge, as well as P. flexilis and P. contorta from an area in the Sierra Nevada where the two species also co-occur. P. contorta includes four distinctly different but interfertile varieties. The variety growing at Niwot Ridge is P. contorta Dougl. var. latifolia Engelm (Rocky Mountain lodgepole pine), and the variety growing in the Sierra Nevada is P. contorta var. murrayana (Sierra lodgepole pine).

3.5.1 All samples were heavily dominated by AAB1 and most samples by AAB2

We found that every sample—from both sites and species—were dominated by AAB1, the same phylotype that dominated P. flexilis and P. engelmannii needles sampled in 2009. To date, endophytic bacterial communities have not been found to be consistently dominated by a single phylotype across all individual sample (Lundberg et al. 2012, Bodenhausen et al. 2013, Bonito et al. 2014). The persistence of this association appears to be stable across time, host species, host species variety, and large geographic distance. We can therefore reject the hypothesis that the shared dominance of AAB1 in needles of P. flexilis and P. engelmannii that was observed in a previous study is caused strictly by shared environment. Although there are similarities between the two sites (e.g. high altitude), there are also significant differences; for example, the Eastern Sierra Nevada receive substantially less precipitation than the eastern Rocky Mountains. However, even

! 41! if the two sites had highly similar environmental characteristics, it seems unlikely that trees in the two geographically separate sites would associate with such similar bacterial taxa based on environment only. More likely, the tendency of trees in the pine family to associate with specific bacteria in the family Acetobacteraceae is a trait that predates the divergence of the tree species. Strict co-speciation between trees and bacteria (i.e. with no horizontal transmission of symbionts) should result in divergence of the 16S rRNA genes in AAB bacteria that reflect the divergence of the hosts. To characterize the endophyte community using Illumina sequencing, we amplified the V-5/V-6 region of 16S using primers that exclude plant chloroplast sequences and yield a bacterial and a plant mitochondrial band that can be separated. The amplified product is ~ 300 nt long. Assuming a 16S rRNA divergence rate of 1% per 50 million years (Ochman and Wilson 1987), the sequences would be expected to be around 97% different, and would likely cluster together as one phylotype (clustering was done at the 97% level). Even when clustering is done at higher levels (e.g. 98%), the sequences are too short to provide resolution that would be necessary to test if co-speciation has occurred (Poretsky et al. 2014). To infer coevolution, would require full-length sequences of AAB1 from several species of pine.

AAB2 was the second dominant phylotype in most of our samples across both sites and is similar to G. liquefaciens, a known N2 fixer of elephant grass (Videira et al. 2012). However, unlike AAB1, AAB2 was not found in all tree samples suggesting that while it may play a role in subalpine conifer health, it is not as critical as AAB1. It is possible the co-occurrence of AAB1 and AAB2 enhance the possible N2 fixation but given the 16S rRNA gene sequence alone, it is difficult to determine the role of these core OTUs. Given the high relative abundance and fairly consistent presence, the role and interactions of the two microbes needs to be further explored.

3.5.2 Remaining community structure and composition

Proteobacteria (71-88%), Bacteroidetes (4-6%), and Actinobateria (2-9%) were the most abundant phyla associated with all of our tree samples, all phyla that were found abundant in our first study and found in the phyllosphere of forest trees (Redford et al. 2010). The endophyte community excluding AAB1 and AAB2 (i.e. the ‘rare’ endophyte community) was found to vary across host species and geographic site, and is likely structured by both factors. For example, in our present study, we found some Horseshoe Meadows samples to contain genera not found in samples at Niwot Ridge: Staphylococcus, Streptococcus, and Acinetobacter. ANOVA and PERMANOVA analysis of unweighted UniFrac distances determined that location was the most important factor structuring the endophytic community, followed by tree species. We also found bacterial endophyte community richness, evenness, and phylogenetic diversity to vary first by location and then by tree species. These results differ from studies of Arabidopsis (Bodenhausen et al. 2013) and Asclepias viridis, Ambrosia psilostachya, Sorghastrum nutans, Panicum virgatum, and Ruellia humilis (Ding et al. 2013) foliage that found species first and then location to be responsible for endophytic bacterial

! 42! community structure. It is possible that with additional locations, species would dominate as a factor structuring the rare endophyte community of pine trees.

The rare endophyte community was also found to vary across time. P. flexilis contained the genera Tanticharoenia, Pandoraeae, Kozakia, and Asaia in Niwot Ridge in 2009 that were not detected in Niwot Ridge P. flexilis in 2012. Instead P. flexilis sampled at Niwot Ridge in 2012 contained genera Acidisoma, Belnapia, and Mucilaginibacter, which were not present in 2009 samples. Variation of community composition between time points is unlikely due to difference in amplicon library sequencing method (454 pyrosequencing for 2009 samples, and Illumina for 2012 samples). It has been demonstrated that 454 pyroseqeuncing and Illumina sequencing of mock communities result in differences in relative abundances of community members but not in the presence and absence of community members (Nelson et al. 2014). Sequence analysis accounts for the greatest amount of variation between communities prepared with difference and both time point communities were processed with the same general pipeline (Nelson et al. 2014).

To conclude, we found that both P. flexilis and P. contorta at two geographically distant sites were dominated by AAB1 as previously reported for P. flexilis and P. engelmannii within one location. This observation rejects the hypothesis that the common environment was the only factor responsible for the shared dominant phylotype reported in the previous study. It is more likely that the presence of AAB1 in both Pinus and Picea is a result of shared ancestry. This striking pattern, which could be the result of host-symbiont co-diversification merits further study, and will require amplification and sequencing from longer parts of 16S rRNA from multiple Pinaceae species.

3.6 References

Aronesty, E. 2011. ea-utils: Command-line tools for processing biological sequencing data.

Bodenhausen, N., M. W. Horton, and J. Bergelson. 2013. Bacterial Communities Associated with the Leaves and the Roots of Arabidopsis thaliana. PLoS ONE 8:e56329.

Bokulich, N. A., S. Subramanian, J. J. Faith, D. Gevers, J. I. Gordon, R. Knight, D. A. Mills, and J. G. Caporaso. 2012. Quality-filtering vastly improves diversity estimates from Illumina amplicon sequencing. Nature Methods 10:57–59.

Bonito, G., H. Reynolds, M. S. Robeson, J. Nelson, B. P. Hodkinson, G. Tuskan, C. W. Schadt, and R. Vilgalys. 2014. Plant host and soil origin influence fungal and bacterial assemblages in the roots of woody plants. Molecular Ecology 23:3356– 3370.

! 43!

structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488:91–5.

Caporaso, J. G., J. Kuczynski, J. Stombaugh, K. Bittinger, F. D. Bushman, E. K. Costello, N. Fierer, A. G. Pena, J. K. Goodrich, J. I. Gordon, and others. 2010. QIIME allows analysis of high-throughput community sequencing data. Nature methods 7:335–336.

Chelius, M. K., and E. W. Triplett. 2001a. The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microb Ecol 41:252–263.

Chelius, M. K., and E. W. Triplett. 2001b. The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microb Ecol 41:252–263.

Dale, C., and N. A. Moran. 2006. Molecular interactions between bacterial symbionts and their hosts. Cell 126:453–65.

Ding, T., M. W. Palmer, and U. Melcher. 2013. Community terminal restriction fragment length polymorphisms reveal insights into the diversity and dynamics of leaf endophytic bacteria. BMC Microbiology 13:1.

Edgar, R. C. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics (Oxford, England) 26:2460–2461.

Jiao, J. Y., H. X. Wang, Y. Zeng, and Y. M. Shen. 2006. Enrichment for microbes living in association with plant tissues. J Appl Microbiol 100:830–7.

Lemaire, B., P. Vandamme, V. Merckx, E. Smets, and S. Dessein. 2011. Bacterial leaf symbiosis in angiosperms: host specificity without co-speciation. PloS One 6:e24430.

Martínez-Romero, E. 2009. Coevolution in Rhizobium-Legume Symbiosis? DNA and Cell Biology 28:361–370.

! 44!

Martinson, V. G., B. N. Danforth, R. L. Minckley, O. Rueppell, S. Tingek, and N. A. Moran. 2011. A simple and distinctive microbiota associated with honey bees and bumble bees. Mol Ecol 20:619–28.

Moran, N. A. 2006. Symbiosis. Curr Biol 16:R866–71.

Moran, N. A. 2007. Symbiosis as an adaptive process and source of phenotypic complexity. Proc Natl Acad Sci U S A 104:8627–33.

Nelson, M. C., H. G. Morrison, J. Benjamino, S. L. Grim, and J. Graf. 2014. Analysis, Optimization and Verification of Illumina-Generated 16S rRNA Gene Amplicon Surveys. PLoS ONE 9:e94249.

Nilsson, M., B. Bergman, and U. Rasmussen. 2000. Cyanobacterial diversity in geographically related and distant host plants of the genus Gunnera. Archives of Microbiology 173:97–102.

Ochman, H., and A. C. Wilson. 1987. Evolution in bacteria: evidence for a universal substitution rate in cellular genomes. J Mol Evol 26:74–86.

Peek, A. S., R. A. Feldman, R. A. Lutz, and R. C. Vrijenhoek. 1998. Cospeciation of chemoautotrophic bacteria and deep sea clams. Proceedings of the National Academy of Sciences of the United States of America 95:9962–9966.

Poretsky, R., L. M. Rodriguez-R, C. Luo, D. Tsementzi, and K. T. Konstantinidis. 2014. Strengths and Limitations of 16S rRNA Gene Amplicon Sequencing in Revealing Temporal Microbial Community Dynamics. PLoS ONE 9:e93827.

Price, M. N., P. S. Dehal, and A. P. Arkin. 2009. FastTree: computing large minimum evolution trees with profiles instead of a distance matrix. Mol Biol Evol 26:1641– 50.

Stewart, F. J., C. R. Young, and C. M. Cavanaugh. 2008. Lateral symbiont acquisition in a maternally transmitted chemosynthetic clam endosymbiosis. Molecular Biology and Evolution 25:673–687.

Taylor, M. J., C. Bandi, and A. Hoerauf. 2005. Wolbachia bacterial endosymbionts of filarial nematodes. Advances in Parasitology 60:245–284.

! 45!

Videira, S. S., D. M. de Oliveira, R. F. de Morais, W. L. Borges, V. L. D. Baldani, and J. I. Baldani. 2012. Genetic diversity and plant growth promoting traits of diazotrophic bacteria isolated from two Pennisetum purpureum Schum. genotypes grown in the field. Plant and Soil 356:51–66.

Webster, N. S., and M. W. Taylor. 2011. Marine sponges and their microbial symbionts: love and other relationships. Environ Microbiol.

! 46!

3.7 Tables

Table 3-1 Student t-test (p-values) of alpha-diversity measurements.

Chao1 PD Shannon Location 0.017 0.001 0.043 Species 0.791 0.844 0.171 Within Niwot Ridge, CO 1 1 1 Within Horseshoe Meadows, CA 1 0.78 1 *PD: Phylogenetic distance

Table 3-2 Anosim and Permanova analysis of unweighted UniFrac distance matrices

Anosim Permanova R P-value Pseudo-F P-value Species within location 0.253 0.003 1.48 0.002 Location 0.5 0.001 2.3258 0.001 Species 0.018 0.324 1.0155 0.400

! 47!

3.8 Figures

Niwot Ridge Horseshoe Meadows Number of OTUs

Sequences

Figure 3-1 Rarefaction curve of endophytic bacterial OTUs at 97% similarity, rarified to 10,000 sequences per sample. Niwot Ridge samples had fewer OTUs than Horseshoe Meadows samples.

! 48!

Acidobacteria Actinobacteria Armatimona- Bacteroidetes Firmicutes Fusobacteria Proteobacteria TM7 Relative Frequencies Relative

PC PF PC PF

Niwot Ridge Horseshoe Meadows

Figure 3-2 Taxonomic relative abundances of endophytic bacterial phylum in P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Relative abundances of phylum were calculated per sample.

! 49!

Taxon

Acidobacteriia

Actinobacteria

Sphingobacteriia

Bacilli

Alphaproteobacteria

Betaproteobacteria

Deltaproteobacteria

Gammaproteobacteria Relative Frequencies Relative Other

PC PF PC PF

Niwot Ridge Horseshoe Meadows

Figure 3-3 Taxonomic relative abundances of endophytic bacterial classes in P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Relative abundances were calculated by abundance of class within each sample.

! 50#

Niwot Ridge Horseshoe Meadows PC PF PC PF Genera Family Class 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 1 Granulicella 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 1 Acidobacteriaceae Acidobacteria 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 2 Terriglobus 2 2 1 1 1 2 2 0 2 2 0 0 1 1 2 1 1 1 1 0 Acidisoma 5 4 4 4 2 4 3 2 4 5 0 1 2 4 4 4 3 2 4 1 Acidisoma 2 3 3 2 1 2 2 6 2 2 0 0 1 2 3 2 1 1 2 2 Belnapia Acetobacteraceae 17 15 16 16 0 12 17 13 16 16 10 9 8 16 14 15 0 8 14 6 * 31 37 39 38 48 35 35 36 31 32 21 19 18 37 38 38 42 29 33 19 Gluconacetobacter 1 1 1 1 0 3 3 1 3 2 0 0 1 1 1 1 1 4 1 Alphaproteobacteria Brevundimonas 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Caulobacteraceae 1 2 2 2 2 1 1 1 1 1 0 0 1 2 2 1 1 1 1 1 Caulobacter 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 Methylobacterium Methylobacteriaceae 3 3 4 4 1 3 3 1 4 4 0 1 2 4 4 2 0 3 3 1 1 1 1 1 2 2 1 0 1 1 0 0 1 1 1 1 2 1 1 0 Sphingomonas Sphingomondaceae 1 1 1 1 2 1 1 0 1 1 0 0 0 1 1 1 1 1 1 0 1 1 1 1 2 2 2 0 2 2 0 0 0 1 1 1 2 2 2 0 Mucilaginibacter Sphingobacteriaceae Bacteriodetes 3 3 2 2 6 3 3 0 2 2 0 1 1 2 3 3 5 3 3 1 1 1 1 1 4 1 1 1 1 0 0 0 1 1 1 1 4 1 1 0 Burkholderia Bukholderiacea 1 1 1 1 3 1 1 0 1 1 0 0 1 1 1 2 1 1 0 Neisseria Neisseriaceae Betaproteobacteria 0 0 0 0 0 0 0 0 0 1 0 1 Herbaspirillum Oxalobacteraceae 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 1 0 3 Staphylococcus Staphylococcaceae 0 0 0 0 2 5 0 0 0 0 0 1 Firmicutes Streptococcus Streptococcaceae 0 0 0 0 0 0 0 0 3 3 9 0 0 0 0 0 0 5 0 0 0 4 0 0 0 1 0 1 Acinetobacter Moraxellaceae Gammaproteobacteria

Figure 3-4 Log-transformed heatmap of top ten OTUs from each sample. The relative abundance (in percentage) is indicated for each OTU in each sample. Color ranges from cool (blue) to warm (red) colors to represent the intensity of domination of each OTU with highest values in red. AAB1 is indicated with an asterisk.

# 51#

Chao1 Number of Phylotypes

PC PF PC PF

Niwot Ridge Horseshoe Meadows

Figure 3-5 Chao richness measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA.

# 52# #

PD Phylogenetic Distance Phylogenetic

PC PF PC PF

Niwot Ridge Horseshoe Meadows

Figure 3-6 Phylogenetic diversity measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Horseshoe Meadows samples generally had a higher phylogenetic diversity.

# 53# #

Shannon Shannon Diversity

PC PF PC PF

Niwot Ridge Horseshoe Meadows

Figure 3-7 Shannon diversity index measurements based on 16S rRNA gene sequences of endophytic bacterial communities of P. contorta (PC) and P. flexilis (PF) in Niwot Ridge, CO and Horseshoe Meadows, CA. Shannon diversity was greater in Horseshoe Meadow samples.

# 54# #

Species

0.2 PC PF 0.1 0.0 PC2: PC2: 7% -0.1 -0.2

-0.2 -0.1 0.0 0.1 0.2

PC1: 25%

Figure 3-8 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to tree species: red, P. contorta; blue, P. flexilis. Elliipses represent 95% confidence intervals for clustering based on tree species.

# 55# #

Location 0.2 Niwot Ridge Horseshoe Meadows 0.1 0.0 PC2: PC2: 7% -0.1 -0.2

-0.2 -0.1 0.0 0.1 0.2

PC1: 25%

Figure 3-9 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to sampling site: red, Niwot Ridge, CO; blue, Horseshoe Meadows, CA. Elliipses represent 95% confidence intervals for clustering based on site.

# 56# #

Species and Location 0.2 PC PF PC PF 0.1 0.0 PC2: PC2: 7% -0.1 -0.2

-0.2 -0.1 0.0 0.1 0.2

PC1: 25%

Figure 3-10 PCoA of the unweighted UniFrac distance matrix of needle endophytic bacterical communities of P. contorta (PC) and P.flexilis (PF). Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and shape correseponds to sample site: diamond, Horseshoe Meadows, CA; triangle, Niwot Ridge, CO. Color of points and ellipses correspond with species within site. Elliipses represent 95% confidence intervals for clustering based on location and site.

# #

4 Enterobacteraceae dominate the bud and needle

endophytic communities of Pinus contorta in a

subalpine meadow

4.1 Abstract

Conifers in growing in subalpine ecosystems have been found to host simple and consistent bacterial endophyte communities in their needle tissues. One dominant phylotype, a potential nitrogen (N) fixing species in the acetic acid bacteria, was found to consistently dominate the needle endophyte community (14-54%) across three species (Pinus flexilis, Pinus contorta, and Picea engelmannii) and two sites, one in the Rocky Mountains, CO and one in the Sierra Nevada, CA. To further investigate the distribution of this potential symbiont taxon (here called AAB1) in high-elevation pines, we used Illumina sequencing of 16S rRNA to examine the endophytic communities of needle and bud tissue in P. contorta individuals in an additional site in the Sierra Nevada, CA. We found that AAB1 was present in low relative abundance (0.01-8%), in the needles at this site; instead, an Enterobacter and a Citrobacter phylotype dominated the samples. There was some overlap in major phylotypes between needle and bud tissue, potentially indicating that new growth is colonized by major endophytes in adjacent needle tissue.

4.2 Introduction

Endophytic bacteria have been detected in all plants studied and can have roles in plant growth, nutrient acquisition and protection against disease and abiotic stress. Despite these important roles, the diversity and structure of endophytic communities in natural plant populations are not well documented. Recently, foliage of Pinus flexilis (limber

57# 58# pine) and co-occurring tree species Pinus contorta (lodgepole pine) (Chapter 2) and Picea engelmannii (Engelmann spruce) was found to be dominated by an acetic acid bacterial phylotype (here called AAB1), which was 97% similar to an endophytic diazotroph previously isolated from sugarcane and other crops (Carrell and Frank 2014) . The AAB1 dominated the needle endophyte community in all P. flexilis, P. engelmannii, (Carrell and Frank 2014) and P. contorta (Chapter 2) individuals at Niwot Ridge, CO, and all samples of P. flexilis and P. contorta growing at another location, Horseshoe Meadows, CA (Chapter 2).

The persistent association across three host individuals, species and two geographic locations suggests a symbiotic relationship, most likely based on endophytic N2 fixation. Such an association could be essential for the tree and therefore obligate, or alternatively, it could be facultative, with presence of dominant symbiotic taxa depending on environmental variables such as soil moisture or ecosystem N supply, and thus vary by site. The distribution of dominant endophyte phylotypes, if determined partly by the local environment, could depend on the timing of establishment of the symbiosis, which is currently unknown. Transmission of endophyte could be horizontal, from the environment (e.g. soil, dust or litter), or vertical (i.e. from parent to offspring via seed or pollen). In general, although horizontal transmission is often assumed (Rosenblueth and Martinez-Romero 2006, Hardoim et al. 2008), very little is known about how bacterial endophytes move between individual plants, but vertical transmission from parent to offspring is possible; presence of bacteria in the seeds and pollen of pine and other species has been reported (Madmony et al. 2005, Cankar et al. 2005, Hardoim et al. 2012, Rout et al. 2013).

A better understanding of the association between high-elevation pines and dominant bacterial taxa inside needles, including establishment of the symbiosis, requires characterization of the endophyte community across more sites and tissues. Because of its broad climatic and geographic range, P. contorta is a suitable species for examining whether presence/absence and/or relative abundance of the dominant phylotypes varies across location. Whereas P. flexilis prefers dry sites and is most often found on rocky ridges and steep rocky slopes, P. contorta has a larger elevation range and thrives on a variety of soils types, from water-logged organic to well-drained glacial outwashes. It co-occurs with P. flexilis at lower tree line and on dry sites in the montane forests (e.g. Niwot Ridge and Horseshow Meadows where previous studies were performed). To date, the bacterial endophyte communities of P. contorta have been explored with cultivation dependent techniques identifying isolates with plant growth promotion properties (Chanway and Holl 1991, Bal et al. 2012) but to our knowledge the endophytic communities have yet to be explored with culture-independent methods.

Differences in endophytic community structure have not been examined across tissue types in P. contorta or other subalpine conifer species. In other plants, different tissues have been found to share taxa suggesting a common source (Bodenhausen et al. 2013). Variability in community composition and structure across tissue types have also been

# 59# detected (Ma et al. 2013). Thus AAB1 may be present in different tissues as a result of a common source or may not be shared across tissue types as a result of horizontal transmission.

In the current study we use 16S rRNA Illumina sequencing to examine the relative abundance of AAB1 inside needles of P. contorta individuals growing at Tuolumne Meadows, a site in the Sierra Nevada, CA. We also examined adjacent bud tissue to study the colonization of new tissue by AAB1, and to test if the diversity and identity of endophytes in newly colonized tissue is more compatible with horizontal endophyte transmission (from the environment) or transmission from adjacent plant tissue.

4.3 Methods

4.3.1 Sample collection and sterilization Tissue samples were collected from 24 P. contorta individuals growing in Tuolumne Meadows, Yosemite National Park, CA. Buds and needles were sampled from each tree with autoclaved razorblades, placed in sterile bags, and brought on ice to University of California, Merced for immediate surface sterilization. Tissue samples were sterilized through submersion in H2O2 followed by three rinses with sterile de-ionized H20 (Izumi et al. 2008). Sterility was confirmed with negative PCR of the final rinse with 16S rRNA primers 799f and 1492r.

4.3.2 DNA extraction and amplification

We ground 0.6 g plant tissue in a sterile mortar and pestle with liquid nitrogen. DNA was extracted using a modified CTAB extraction with bead beating previously used to extract bacterial endophytic bacteria from P. contorta and other conifers (Chapter 2). DNA was precipitated with sodium acetate and isopropanol overnight followed by 30-minute centrifugation and an ethanol rinse. DNA was resuspended with 30 µl of DNA resuspension fluid (1.0 M Tris-HCL, 0.1 M EDTA) and stored at –20°C.

We amplified 16S rRNA genes of endophytic bacterial using nested PCR to reduce the occurrence of plastid sequences (Chelius and Triplett 2001) and improve consistency (Hanshew et al. 2013). Plant DNA amplification was suppressed with the primer pair 16S 799f (AACMGGATTAGATACCCKG) and 16S 1492r (TACGGHTACCTTGTTACGACT) in the first PCR reaction (PCR1). Amplification with 16S 799f and 16S 1492r typically results in mitochondrial amplicon of about 1000 bp and bacterial amplicon of about 750 bp (Chelius and Triplett 2001, Carrell and Frank 2014, Hanshew et al. 2013). In the final round of PCR (PCR2), an appropriate amplicon length for Illumina sequences was achieved from PCR1 amplicons with the Illumina adapted, Golay-barcoded primer pair 799f and 1115r (AGGGTTGCGCTCGTTG), an optimized primer set for phylogenetic analysis of short reads ( Redford et al., 2010). We reduced primer bias by decreasing the number of cycles (Jiao et al. 2006) with the following thermocycle profile used for PCR1 and PCR2: one cycle of 3 min at 95°C; 20 cycles of 40 s at 95°C, 40 s at 50°C, 1.5 min at 72°C; and a final 10 min of elongation at

# 60#

72°C. The 50 µl PCR1 reaction contained 5 ul of DNA extract, 20 µL 5 PRIME Hot Master Mix (5 PRIME, Inc.), 0.5 µg/µl Bovine Serum Albumin (Thermo Scientific), 21.5 µL PCR grade water (Fischer BioReagents), and 0.2 µM of forward and reverse primers. The 25 µL PCR2 reaction contained 3 µL of PCR1 product, 10 µL 5 PRIME Hot Master Mix, 0.5 µg/µl Bovine Serum Albumin (Thermo Scientific), 8.75µL PCR grade water (Fischer BioReagents), and 0.2 µM of forward and reverse primers. To achieve adequate concentration of amplicon product for sequencing, we completed PCR reactions in triplicate. PCR2 amplicons were then cleaned, pooled, and gel extracted (QIAquick Gel Extraction Kit) to ensure selection of the correct band size and to remove most mitochondrial products. Pooled samples were then submitted for Illumina sequencing at the University of California, Davis Genome Center.

4.3.3 Sequence Analysis

Sequences were processed and analyzed with the QIIME (1.8.0) package. Forward and reverse paired-end reads were joined with fastq-join; if the forward and reverse read did not overlap the sequences were removed from analysis (Aronesty 2011). Joined pair-end reads were quality filtered with QIIME defaults settings. Specifically, reads were excluded if > 3 consecutive bases had low quality base calls, if <75% of the read length was consecutive high quality base calls, if the Phred quality score <3, or if there were any ambiguous bases (Bokulich et al. 2012). Samples with <1000 sequences were removed from analysis (7 trees). The remaining sequences were clustered into operational taxonimc units (OTU) at 97% similarity using open-reference OTU picking: sequences were clustered against greengenes reference sequence collection, and any reads that did not match a reference sequences were clustered against each other without reference using uclust 1.2.22q (Edgar 2010). A representative sequence was chosen for each phylotype by selecting the longest sequence that had the highest number of hits to other sequences of that particular phylotype. Representative sequences were aligned using PyNAST (Caporaso et al. 2010) against the Greengenes core set (DeSantis et al. 2006). Taxonomic assignments were made using uclust with greengenes representative set of sequences as reference. Sequences classified as “Chloroplast” or “Mitochondria” were removed from the alignment. An approximately maximum-likelihood tree was constructed from the aligned of bacterial representative sequences, using FastTree (Price et al., 2009). A heatmap of the top 10 OTUs from each sample was generated using QIIME.

Prior to conducting diversity analyses, OTUs were rarefied to 1000 reads per sample. Alpha diversity was measured using different diversity metrics: Chao1 richness and Faith’s phylogenetic diversity. Diversity measurements were compared between tree stands using analysis of variance (ANOVA, α=0.05). Student's t-tests were used to compare diversity between tissue types (bud vs needle). Beta diversity was visualized using principle coordinate analysis (PCoA) from weighted UniFrac distances. Analysis of similarities (ANOSIM) and permutational multivariate analysis of variance (PERMANOVA), each with 999 permutations, were used to determine if beta diversity differed significantly among tissues.

# 61#

4.4 Results

With Illumina sequencing of the 16S rRNA of the V5-V6 region, we obtained a total of 1,050,978 quality sequences representing 257 OTUs at 97% similarity: 103 for buds and 197 for needles. Both bud and needle endophyte communities were dominated by Enterobacteriales (Gammaproteobacteria): 95% of total sequences in buds and 55% of total sequences in needles. Rhodospirillales and Rhizobiales were abundant in needles (8% and 7% of sequences) but not in buds (Figure 4-1). A heatmap generated from OTUs with a relative abundance >1% contained a total of 34 OTUs (Figure 4-2). A Gluconacterobacter phylotype, 100% similar to the characteristic AAB1 phylotype previously detected in P. flexilis and co-occurring tree species, was found present at low relative abundance in both buds (<0.01% average relative abundance) and needles (2.5% average relative abundance). Instead, a phylotype most closely related to an Enterobacter sp. dominated most needle samples (0-67%), and a second phylotype related to Citrobacter freundii also dominated most needle samples (0-24%). A single phylotype most related to Pantoea agglomerans dominated all except two bud samples at high relative abundance (0-89%). Bud and needle tissues shared some bacterial families but only needles were colonized by members of the families Caulobacteraceae, Streptococcaceae, Lactobacillaceae, Enterococcaceae, Staphylocaccaceae, and Chitinophagaceae.

Endophytic bacterial community diversity varied by tissue type. The number of observed species (Figure 4-3), richness (Figure 4-4), and phylogentic diversity (Figure 4-5) were greater in needles than buds (p=0.001). Principle coordinate analysis of the UniFrac distances demonstrated clustering by tissue type (Figure 4-6) and was confirmed with ANOSIM (R = 0.63, P = 0.001) and PERMANOVA (Pseudo-F statistic = 2.54, P = 0.001).

4.5 Discussion

4.5.1 Enterobacter and Citrobacter species, not Gluconacetobacter species, dominate the needles of P. contorta at Tuolomne Meadows, CA

A single phylotype in the acetic acid bacteria, most closely related to Gluconacetobacter diazotrophicus was previously found to consistently dominate the endophyte community in P. contorta and P. flexilis growing at two separate geographic sites (Niwot Ridge, CO and Horseshow Meadows, CA). This characteristic phylotype, here called AAB1, was also found in high relative abundance in all P. engelmannii individuals at Niwot Ridge, CO. In the current study we used Illumina sequencing of 16S rRNA to investigate the endophyte community structure, specifically in regards to the presence and abundance of AAB1 in needle and bud tissue in P. contorta growing in Tuolumne Meadows, CA. A Gluconacterobacter phylotype, 100% similar to the AAB1 phylotype previously detected

# 62# in P. flexilis and co-occurring tree species, was found to be consistently present in needle samples, although not dominant (the average relative abundance was very low at 2.5%). Instead, the needles were dominated by Citrobacter (0-24%) and Enterobacter (0-48%) phylotypes, with variable relative abundance across trees.

At a low relative abundance, AAB1 is not likely responsible for a critical role inside the needles of P. contorta growing in Tuolumne Meadows, CA. Moreover, the association between AAB1 and P. contorta, if mutualistic, must be facultative. If AAB1 observed in pines growing in previously studied sites are mutualistic N2 fixers, there are two possibilities for why this taxon would be absent in some sites as observed here. First, the establishment of the association may depend on ecosystem N supply, in which case we would not expect dominance of N2 fixing endophytes to be established under high N conditions. Second, different dominant endophyte taxa could perform the N2 fixing role in some parts of the P. contorta range.

Under the first scenario, differences in dominating taxa would arise due to difference in ecosystem N status between Tuolumne and Horseshoe Meadows. Such differences could arise due to different soil N content, bedrock N (Morford et al. 2011) and antropogenic N. Both Tuolumne and Horseshoe Meadows receive visitors and hikers, however, Tuolumne Meadows is located next to the heavily trafficked Tioga Rd, and likely receives more N deposition, with potential implications for the establishment of an N2-fixing association. N deposition is known to affect other plant-bacteria associations; for example, the N2 fixing activity in the association between feather moss and cyanobacteria is significantly inhibited close to busy roads (Ackermann et al. 2012).

Under the second scenario (i.e. dominant N2 fixing endophyte vary across the P. contorta range) other major phylotypes would be N2 fixing endophytes in P. contorta growing at Tuolumne Meadows. Enterobacter sp., which dominate most needle samples are frequently detected as plant-associated microbes, and have been found as N2 fixing endophytes of sugarcane (Zhu et al. 2012, 2013). The relative abundance of this phylotype ranged between 0 and 67%, and was less consistent than previously observed for AAB1 (Carrell and Frank 2014).

4.5.2 Little overlap in the major phylotypes in buds and needles

Principal coordinate analysis of weighted UniFrac distances of our endophytic community indicated tissue type structured bacterial endophytic communities. This is consistent with other endophyte studies that find community structure vary between tissue types (Bodenhausen et al. 2013, Ma et al. 2013). Diversity and richness was lower in buds than needles, which is expected for buds in the process of being colonized by an endophytic community, whether from the environment or from nearby plant tissue. If needles were colonized by major needle tissue taxa, we would expect the buds to host a subset of the taxa detected in needles. The amount of overlap in major phylotypes between bud and needle communities varied by phylotype. For example, the Pantoea phylotype that dominated buds was found only at very low relative abundance in the

# 63# needles (0-3%). Thus, Pantoea species may colonize the buds from the environment, or alternatively come from a small source community of relatively rare bacterial species in needles and grow to higher relative abundance in buds. Likewise, an Erwinia phylotype was consistently present in buds (>0.1-33%) and present at low relative abundance in all except one needle sample. In contrast, the other two major phylotypes found in the buds were two of the Enterobacter phylotypes that were also found to dominate needles. These phylotypes did not consistently dominate the bud samples, but were found at high relative abundance in a few samples. Thus, it is possible that bud colonization of the major needle phylotype had just begun when samples were taken, and that only a subset of the buds were colonized at high relative abundance. Other phylotypes present at low relative abundance in needles but virtually absent from buds (e.g. Bacillales and Lactobacillales) could be horizontally transmitted over the 10-13 year lifetime of individual P. contorta needles (Schoettle 1990). While needle communities were dominated by Enterobacter, bud communities were mostly dominated by Pantoea agglomerans. P. agglomerans has been detected as an endophyte with a role in sugarcane growth promotion (Quecine et al. 2012). In eucalyptus, P. agglomerans strains are vertically transferred from seeds to seedlings (Ferreira et al. 2008). It is possible that P. agglomerans promotes growth of actively growing tissue, similar to Methylobacterium extorquens in the bud meristem of Pinus sylvestris (scots pine) (Pirttilä et al. 2000, 2004, 2005). If transmitted vertically via pollen or seeds, bacteria would be likely to reside in bud tissue.

To conclude the current study provides evidence that AAB1 is not universally associated with P. contorta. The phylotype was not present at high relative abundance in the needle endophyte community on P. contorta growing at Tuolumne Meadows, CA. There was some overlap between bud and needle tissue in the community of major endophytic phylotypes, consistent with transfer both from the environment and from adjacent tissue. In particular, the major needle endophyte phylotypes were present in a subset of bud samples, indicating that some buds had been colonized by needle bacteria while others were not, possibly due to difference in bud age. Our results suggest that the relationship between P. contorta and AAB1 is facultative, possibly with establishment determined by local environmental factors such as N supply. Alternatively, other major taxa (e.g. Enterobacter sp.) may be the main symbionts in some parts of the P. contorta range. Geographic variation in source populations, climate, and ecosystem N status (soil and/or N deposition) could explain the variation in endophyte phylotypes across the host range. Difference in dominant endophyte taxa may reflect varied tolerance to environmental conditions (temperature, soil moisture) across sites. To identify the host- and environmental factors that structure the diversity of dominant endophyte taxa across locations in the P. contorta range, sampling across environmental gradients (soil N and/or N deposition), along with measurements of N-fixation activity will be required.

4.6 References

Ackermann,# K.,# O.# Zackrisson,# J.# Rousk,# D.# L.# Jones,# and# T.# H.# DeLuca.# 2012.# N2# Fixation#in#Feather#Mosses#is#a#Sensitive#Indicator#of#N#Deposition#in#Boreal# Forests.#Ecosystems#15:986–998.#

# 64#

Aronesty,#E.#2011.#eaUutils:#CommandUline#tools#for#processing#biological#sequencing# data.#

Bal,#A.,#R.#Anand,#O.#Berge,#and#C.#P.#Chanway.#2012.#Isolation#and#identification#of# diazotrophic# bacteria# from# internal# tissues# of# Pinus# contorta# and# Thuja# plicata.#Can#J#Forest#Res#42:807–813.#

Bodenhausen,# N.,# M.# W.# Horton,# and# J.# Bergelson.# 2013.# Bacterial# Communities# Associated#with#the#Leaves#and#the#Roots#of#Arabidopsis#thaliana.#PLoS#ONE# 8:e56329.#

Bokulich,#N.#A.,#S.#Subramanian,#J.#J.#Faith,#D.#Gevers,#J.#I.#Gordon,#R.#Knight,#D.#A.#Mills,# and#J.#G.#Caporaso.#2012.#QualityUfiltering#vastly#improves#diversity#estimates# from#Illumina#amplicon#sequencing.#Nature#Methods#10:57–59.#

Cankar,# K.,# H.# Kraigher,# M.# Ravnikar,# and# M.# Rupnik.# 2005.# Bacterial# endophytes# from# seeds# of# Norway# spruce# (Picea# abies# L.# Karst).# FEMS# Microbiol# Lett# 244:341–5.#

Caporaso,#J.#G.,#J.#Kuczynski,#J.#Stombaugh,#K.#Bittinger,#F.#D.#Bushman,#E.#K.#Costello,# N.# Fierer,# A.# G.# Pena,# J.# K.# Goodrich,# J.# I.# Gordon,# and# others.# 2010.# QIIME# allows# analysis# of# highUthroughput# community# sequencing# data.# Nature# methods#7:335–336.#

Carrell,# A.# A.,# and# A.# C.# Frank.# 2014.# Pinus# flexilis# and# Piceae# engelmannii# share# a# simple#and#consistent#needle#endophyte#microbiota#with#a#potential#role#in# nitrogen#fixation.#Frontiers#in#Microbiology#5.#

Chanway,# C.,# and# F.# B.# Holl.# 1991.# Biomass# increase# and# associative# nitroU#gen# fixation# of# mycorrhizal# Pinus# contorta# seedlings# inoculated# with# a# plant# growth#promoting#Bacillus#strain.#Can#J#Bot#69:507–511.#

Chelius,# M.# K.,# and# E.# W.# Triplett.# 2001.# The# Diversity# of# Archaea# and# Bacteria# in# Association#with#the#Roots#of#Zea#mays#L.#Microb#Ecol#41:252–263.#

DeSantis,#T.#Z.,#P.#Hugenholtz,#N.#Larsen,#M.#Rojas,#E.#L.#Brodie,#K.#Keller,#T.#Huber,#D.# Dalevi,#P.#Hu,#and#G.#L.#Andersen.#2006.#Greengenes,#a#chimeraUchecked#16S# rRNA# gene# database# and# workbench# compatible# with# ARB.# Applied# and# environmental#microbiology#72:5069–5072.#

Edgar,# R.# C.# 2010.# Search# and# clustering# orders# of# magnitude# faster# than# BLAST.# Bioinformatics#(Oxford,#England)#26:2460–2461.#

Ferreira,#A.,#M.#C.#Quecine,#P.#T.#Lacava,#S.#Oda,#J.#L.#Azevedo,#and#W.#L.#Araujo.#2008.# Diversity# of# endophytic# bacteria# from# Eucalyptus# species# seeds# and#

# 65#

colonization# of# seedlings# by# Pantoea# agglomerans.# FEMS# Microbiol# Lett# 287:8–14.#

Hanshew,# A.# S.,# C.# J.# Mason,# K.# F.# Raffa,# and# C.# R.# Currie.# 2013.# Minimization# of# chloroplast# contamination# in# 16S# rRNA# gene# pyrosequencing# of# insect# herbivore#bacterial#communities.#J#Microbiol#Methods#95:149–55.#

Hardoim,#P.#R.,#C.#C.#Hardoim,#L.#S.#van#Overbeek,#and#J.#D.#van#Elsas.#2012.#Dynamics# of# seedUborne# rice# endophytes# on# early# plant# growth# stages.# PLoS# ONE# 7:e30438.#

Hardoim,# P.# R.,# L.# S.# van# Overbeek,# and# J.# D.# Elsas.# 2008.# Properties# of# bacterial# endophytes# and# their# proposed# role# in# plant# growth.# Trends# Microbiol# 16:463–71.#

Ma,# B.,# X.# Lv,# A.# Warren,# and# J.# Gong.# 2013.# Shifts# in# diversity# and# community# structure# of# endophytic# bacteria# and# archaea# across# root,# stem# and# leaf# tissues#in#the#common#reed,#Phragmites#australis,#along#a#salinity#gradient#in# a# marine# tidal# wetland# of# northern# China.# Antonie# van# Leeuwenhoek# 104:759–768.#

Madmony,#A.,#L.#Chernin,#S.#Pleban,#E.#Peleg,#and#J.#Riov.#2005.#Enterobacter#cloacae,# an# obligatory# endophyte# of# pollen# grains# of# Mediterranean# pines.# Folia# Microbiol#(Praha)#50:209–16.#

Morford,#S.#L.,#B.#Z.#Houlton,#and#R.#A.#Dahlgren.#2011.#Increased#forest#ecosystem# carbon#and#nitrogen#storage#from#nitrogen#rich#bedrock.#Nature#477:78–81.#

Pirttilä,# A.# M.,# P.# Joensuu,# H.# Pospiech,# J.# Jalonen,# and# A.# Hohtola.# 2004.# Bud# endophytes#of#Scots#pine#produce#adenine#derivatives#and#other#compounds# that# affect# morphology# and# mitigate# browning# of# callus# cultures.# Physiol# Plant#121:305–312.#

Pirttilä,#A.#M.,#H.#Laukkanen,#H.#Pospiech,#R.#Myllyla,#and#A.#Hohtola.#2000.#Detection# of#intracellular#bacteria#in#the#buds#of#Scotch#pine#(Pinus#sylvestris#L.)#by#in# situ#hybridization.#Appl#Environ#Microbiol#66:3073–7.#

Pirttilä,#A.#M.,#H.#Pospiech,#H.#Laukkanen,#R.#Myllyla,#and#A.#Hohtola.#2005.#Seasonal# variations# in# location# and# population# structure# of# endophytes# in# buds# of# Scots#pine.#Tree#Physiol#25:289–97.#

Quecine,# M.# C.,# W.# L.# Araújo,# P.# B.# Rossetto,# A.# Ferreira,# S.# Tsui,# P.# T.# Lacava,# M.# Mondin,# J.# L.# Azevedo,# and# A.# A.# PizziraniUKleiner.# 2012.# Sugarcane# growth# promotion#by#the#endophytic#bacterium#Pantoea#agglomerans#33.1.#Applied# and#Environmental#Microbiology#78:7511–7518.#

# 66#

Redford,#A.#J.,#R.#M.#Bowers,#R.#Knight,#Y.#Linhart,#and#N.#Fierer.#2010.#The#ecology#of# the#phyllosphere:#geographic#and#phylogenetic#variability#in#the#distribution# of#bacteria#on#tree#leaves.#Environ#Microbiol#12:2885–93.#

Rosenblueth,# M.,# and# E.# MartinezURomero.# 2006.# Bacterial# endophytes# and# their# interactions#with#hosts.#Mol#Plant#Microbe#Interact#19:827–37.#

Rout,#M.#E.,#T.#H.#Chrzanowski,#T.#K.#Westlie,#T.#H.#Deluca,#R.#M.#Callaway,#and#W.#E.# Holben.#2013.#Bacterial#endophytes#enhance#competition#by#invasive#plants.# Am#J#Bot#100:1726–37.#

Schoettle,#A.#W.#1990.#The#interaction#between#leaf#longevity#and#shoot#growth#and# foliar#biomass#per#shoot#in#Pinus#contorta#at#two#elevations.#Tree#Physiology# 7:209–214.#

Zhu,# B.,# M.# Chen,# L.# Lin,# L.# Yang,# Y.# Li,# and# Q.# An.# 2012.# Genome# Sequence# of# Enterobacter# sp.# Strain# SP1,# an# Endophytic# NitrogenUFixing# Bacterium# Isolated#from#Sugarcane.#Journal#of#Bacteriology#194:6963–6964.#

Zhu,# B.,# Q.# Zhou,# L.# Lin,# C.# Hu,# P.# Shen,# L.# Yang,# Q.# An,# G.# Xie,# and# Y.# Li.# 2013.# Enterobacter# sacchari# sp.# nov.,# a# nitrogenUfixing# bacterium# associated# with# sugar# cane# (Saccharum# officinarum# L.).# International# Journal# of# Systematic# and#Evolutionary#Microbiology#63:2577–2582.#

# 67#

4.7 Figures

Order Taxonomy

1.00

Saprospirales

Bacillales

0.75 Lactobacillales

Caulobacterales

Rhizobiales

Rhodospirillales

Rickettsiales

0.50 Sphingomonadales e Frequency

v Burkholderiales

Enterobacteriales Relati Pseudomonadales

Xanthomonadales

TM7 0.25 Other

0.00

Buds Needles

Figure 4-1 Relative abundances of various bacterial classes based on 16S rRNA gene sequences at 97% similarity from buds and needles of P. contorta. Relative abundances of classes were calculated as percentage of sequences belonging to a particular class recovered from each sample.

# 68#

Buds Needles Genera Family Order 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 5 0 0 0 1 3 Chitinophaga Chitinophagaceae Saprospirales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 1 1 1 3 1 2 0 0 2 1 1 1 1 1 1 0 Bacillus Bacillaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 4 3 4 1 3 0 2 4 1 1 1 1 1 2 2 Bacillales Staphylococcus Staphylococcaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 2 1 0 1 0 2 1 1 1 1 1 1 2 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 1 1 1 0 0 1 0 0 0 0 0 2 0 Enterococcus Enterococcaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 1 1 1 0 0 Lactobacillus Lactobacillaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 0 2 4 0 4 4 4 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 1 1 2 0 1 0 0 1 1 0 1 0 0 2 1 Lactococcus Lactobacillales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 2 2 3 3 3 2 0 0 6 1 0 1 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 1 0 0 0 0 0 0 0 Streptococcaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 0 0 0 0 4 0 0 0 0 1 0 1 0 0 0 2 0 Streptococcus 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 1 1 0 0 3 0 0 1 0 2 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 0 1 0 1 0 1 0 1 0 0 0 0 1 1 Caulobacter Caulobacteraceae Caulobacterales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 2 0 0 0 0 0 0 Bosea Bradyrhizobiaceae Rhizobiales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 3 0 0 10 9 10 0 0 0 1 4 Methylobacterium Methylobacteriaceae 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 1 2 3 8 8 2 5 1 1 1 3 1 2 2 1 1 Gluconacetobacter Acetobacteraceae Rhodospirillales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 7 0 1 3 13 4 0 0 7 0 0 0 0 0 0 0 0 0 0 Rickettsia Rickettsiaceae Rickettsiales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 Wolbachia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 0 0 0 0 0 0 0 0 7 1 0 0 0 0 1 Sphingomonas Sphingomonadaceae Sphingomonadales 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 1 5 0 3 0 0 0 2 0 1 1 1 0 0 Herbaspirillium Oxalobacteraceae Burkholderiales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 0 0 0 0 0 0 1 0 0 1 1 1 0 0 Neisseria Neisseriaceae Neisseriales 1 1 6 0 0 1 0 1 1 1 0 4 1 1 1 1 14 8 4 11 7 24 13 20 0 2 18 12 20 4 5 5 22 3 Citrobacter 44 31 3 0 0 3 4 43 13 8 2 1 1 1 1 19 1 14 44 30 5 1 42 11 67 35 8 6 0 48 51 47 21 51 0 1 3 0 0 0 0 1 1 1 0 1 0 0 3 3 0 0 2 4 2 5 1 9 0 2 4 6 0 2 2 2 5 1 Enterobacter 37 26 1 0 0 1 1 35 12 5 2 0 0 0 2 3 0 47 4 2 2 0 0 1 18 33 0 6 0 8 10 9 2 4 0 9 3 8 10 10 10 1 32 33 13 8 8 10 10 4 14 2 0 0 15 0 0 0 1 0 0 0 0 0 0 0 0 0 Enterobacteriaceae Enterobacteriales Erwinia 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 4 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 15 29 89 84 80 73 1 20 36 72 71 83 60 60 22 68 2 0 0 0 0 0 1 1 0 3 0 0 0 0 1 0 Pantoea 0 0 1 1 1 1 2 0 1 1 1 1 1 1 1 0 1 0 0 0 4 0 0 0 0 0 0 0 0 0 0 0 0 0 Providencia 14 8 0 0 0 0 1 13 4 2 1 0 0 0 0 1 0 10 1 1 1 0 0 0 5 12 0 2 0 3 4 5 1 1 Rahnella 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 0 0 3 0 1 1 1 1 0 Acinetobacter Moraxellaceae Pseudomonadales 0 0 1 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 1 0 3 0 0 0 0 1 0 0 1 1 1 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 4 0 0 0 3 4 Stenotrophomonas Xanthomonadaceae Xanthomonadales 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 6 0 1 1 0 1 9 0 TM7 TM7 CW040

Figure 4-2 Log-transformed heatmap of OTUs with a relative abundance >1%. The relative abundance (in percentage) is indicated for each OTU in each sample. Color ranges from cool (blue) to warm (red) colors to represent the intensity of domination of each OTU with highest values in red. Relative abundances <0.01% are indicated with black text; >0.01% are indicated with white text.

# 69#

Observed Species

Buds Needles

Figure 4-3 The number of observed endophytic bacterial species at the 97% similarity of 16S rRNA gene sequences in needles is higher than the number of observed species in buds.

# 70#

Chao1 Richness

Buds Needles

Figure 4-4 Chao1 richness based on 16S rRNA gene sequences of endophytic bacteria in P. contorta was higher in needles than buds.

# 71#

Phylogenetic Distance

Buds Needles

Figure 4-5 Phylogenetic distance, or the sum of branches of phylogeny in each sample, based on 16S rRNA gene sequences of endophytic bacteria in P. contorta was higher in needles than buds.

# 72#

Needles Buds PC2: 20%

PC1: 36%

Figure 4-6 PCoA of the weighted UniFrac distance matrix of needle and bud endophytic bacterical communities of P. contorta. Points that are closer together on the ordination have bacterial communities that are more similar. Each point corresponds to a sample and color corresponds to tissue type: red, bud; blue, needle.

# #

5 Diversity and Structure of the Bacterial

Endophyte Communities in the Foliage of Giant

Trees

5.1 Abstract

Microbial endophytes are an integral part of the biology of all plants, with implications for agricultural and natural ecosystems. In long-lived plants such as coast redwood, endophyte communities that are heterogenic within and across individuals may protect the host against pathogen and pests. Culture–based methods have revealed constant foliage endophyte composition in coast redwood leaves along a north to south transect. Similarly, recent culture-independent work has revealed that the bacterial needle endophyte communities of pines are marked by low diversity and stability across individuals and species. The objective here was to test if bacterial endophytes communities of coast redwoods (Sequoia sempervirens) are similarly stable or, alternatively, variable within and across individuals and populations. We sampled foliage from three canopy heights of two coast redwood populations and one giant sequoia (Sequoiadendron giganteum) population. Pyrosequencing of the 16S rRNA was used to determine the diversity and structure of the bacterial endophyte communities in the samples. Bacterial endophyte communities of coast redwood and giant sequoia were considerably more diverse and variable than those previously observed in pines. No single phylotype or genus dominated all samples of a species. We found bacterial diversity and richness to be lower in the upper canopy of coast redwood, but not giant sequoia. Many dominant endophytic taxa belong to genera known for roles in abiotic stress protection. Microbial adaptation and community shifts may help a long-lived tree overcome ongoing challenges from rapidly changing pests and pathogens.

73# 74#

5.2 Introduction

The iconic coast redwood is the tallest living tree species, with a life span that can extend 2000 years. The only extant species in the genus Sequoia in the family Cupressaceae, coast redwood is restricted to a narrow belt of cloud-inundated humid areas along the coast of central and northern California. Coast redwood is shade tolerant, relies on cool, humid marine conditions, and is highly resistant to fungal disease (Sawyer, J.O. et al. 2000). As such, it is markedly different from the shade-intolerant Pinaceae species that dominate North American conifer forests: Pseudotsuga, Picea, or Pinus species are more vulnerable to insects, fire, and wood-decay fungi, and typically have shorter life spans. Because plant defensive traits are often mediated by microbes (Friesen et al. 2011), such differences may depend both on adaptations in the plant genome, and on associations with beneficial microbes such as fungal and bacterial endophytes.

Endophyte-mediated plant defense may be especially important in long-lived plants such as coast redwood (Carroll 1988). All plants studied so far are colonized by fungal and bacterial endophytes inside their tissues. In general, fungal endophytes of trees display high diversity, spatial heterogeneity within the canopy, and antagonistic ability against pathogens (Arnold et al. 2003, Ganley et al. 2004, Rodriguez et al. 2009, Raghavendra and Newcombe 2013). The investigation of fungal endophytes in coast redwood was pioneered by Carroll (Carroll and Carroll 1978), who isolated four different endophyte species. Later studies found a higher diversity, e.g. 26 different species (Espinosa-Garcia and Langenheim 1990). The most extensive study to date documented 16 different endophyte species, and found that the fungal endophyte community was stable among host individuals and along a north- to south distribution of coast redwood, with dominance of Pleuroplacoema sp., (Rollinger and Langenheim 1993).

To our knowledge, no previous study on the bacterial endophytic community of coast redwood exists. Endophytic bacteria, have been experimentally shown to benefit host plants in a number of ways, conferring increased resistance to pathogens, improved nutrient acquisition, and resistance to environmental stresses (Rosenblueth and Martinez- Romero 2006, Hardoim et al. 2008, Reinhold-Hurek and Hurek 2011) . Bacterial endophytes have been shown to protect against pathogenic fungi, protists, and bacteria, and in instances insects and nematodes; either through production of antimicrobials, or through induced resistance (Compant et al. 2005, 2010). Relatively few studies of bacterial endophytes in natural populations exist, but endophyte isolates from forest conifers have been shown to fix nitrogen (N) (Anand and Chanway 2013, Anand et al. 2013), to play a role in shoot tissue growth and development (Pirttilä et al. 2004, 2005, Pohjanen, J. et al. 2014), and to protect against the fungal symbiont of pine beetle (Adams et al. 2008). The mechanisms that influence bacterial endophyte community diversity in coast redwood are unknown. Factors such as plant host (Bonito et al. 2014) soil type (Lundberg et al. 2012, Bonito et al. 2014) and host genotype (Lundberg et al. 2012) have been demonstrated to structure bacterial endophyte community assemblage

# 75# in various plants. Recent results show that the bacterial foliage endophyte community in the subalpine species limber pine (Pinus flexilis) and Engelmann spruce (Picea engelmannii) is stable across individuals and species and dominated by the same single operational taxonomic unit (OTU) in the family Acetobacteraceae (Carrell and Frank 2014).

Here, as a first step towards understanding the role and diversity of the bacterial endophytes of coast redwood foliage, we used 16S rRNA pyrosequencing to investigate the effect of geographic location and canopy height on the structure of the endophyte community. In addition, for phylogenetic comparison, we examined the endophyte community in coast redwood’s closest extant phylogenetic relative giant sequoia in the family Cuppressaceae.

5.3 Methods

5.3.1 Sample collection and sterilization

Coast redwood needles were collected from two coastal sites (Samuel P. Taylor State Park and Big Creek UC Natural Reserve, CA). Giant sequoia needles were collected from trees growing at Freeman Creek Grove in Sequoia National Monument, CA. All samples were collected in Fall 2011. To assess the difference in endophytic communities across individuals, locations and species, three giant sequoia, and six coast redwood trees (three per location) were sampled. To contrast intra-tree variation in the endophytic community, needles were sampled from three heights (lower, middle, and upper) from each tree (Table 1). For each sample, approximately 10 g of needles were removed with a sterile razor blade, placed in a ziplock bag, and transported to the University of California, Merced at 4°C. Needles were sterilized through submersion in ethanol for one minute, 30% hydrogen peroxide for three minutes, followed by three rinses with sterile deionized water, and stored at -20°C. Surface sterility of foliage was confirmed by negative PCR amplification of the final rinse as well as lack of growth on general media plates.

5.3.2 DNA Extraction

Needles were pulverized to a fine powder in a sterile mortar in the presence of liquid nitrogen. DNA was extracted from 0.6g of pulverized tissue in a 2 ml screw cap tube containing 800µl of CTAB solution (1 ml CTAB buffer, 0.04 g of polyvinylpyrollidone, 5 µl of 2-mercaptoethanol), incubated in a dry bath at 60°C for 2 hours, and then homogenized with 0.3 g of 0.11 mm sterile glass beads with a bead beater for 3 minutes. Proteins were removed with the addition of an equal amount of chloroform and centrifuged for 10 minutes at 16 rcf. The aqueous top phase was placed in a sterile 2 ml snap cap tube with 1/10 volume of cold 3 M sodium acetate and 1/2 volume cold isopropanol and placed in a –20°C freezer overnight to precipitate the nucleic acids. The

# 76# sample was then centrifuged for 30 minutes at 16 rcf, supernatant decanted, 700 µl of 70% ethanol added, and centrifuged for 10 minutes. The air-dried pellet was resuspended with 30 µl of DNA resuspension fluid (1.0 M Tris-HCL, 0.1 M EDTA) and stored at – 20°C.

5.3.3 DNA Amplification

DNA was amplified by methods previously described by Carrell et al. (Carrell and Frank 2014). Briefly, DNA was amplified using a nested PCR using the thermocycle profile described by Jiao et al. (2006). For the initial PCR, we used the chloroplast excluding primer 16S 799f (AACMGGATTAGATACCCKG) and 16S 1492r (TACGGHTACCTTGTTACGACTT) which resulted in a mitochondrial product of about 1000 bp and a bacterial product of about 750 bp as described by Chelius and Triplett (Chelius and Triplett 2001). The bacterial product was then separated from mitochondrial product and extracted using E-Gel® SizeSelect™ Gels (Life Technologies, Carlsbad, CA, USA). The extracted bacterial product was amplified with the thermocycle profile described by Jiao et al. (Jiao et al. 2006) using the Golay-barcoded primer set, 799f and 1115r (AGGGTTGCGCTCGTTG) (Redford et al. 2010). The final product was then cleaned, quantified with Nanodrop, and equally pooled for pyrosequencing. The pooled product was sent to the Environmental Genomics Core Facility at the University of South Carolina for pyrosequencing on a 454 Life Sciences Genome Sequencer FLX machine.

5.3.4 Phylotype generation and classification

Sequences were analyzed and processed using the QIIME package (Caporaso et al., 2010b). Sequences were quality filtered (minimum quality score of 25, minimum length of 200 bp, and no ambiguity in primer sequence) and assigned to their corresponding sample by the barcode sequences. One sample (middle needles from a giant sequoia individual) was removed due to insufficient number of sequences. The remaining sequences were clustered into operational taxonomic units (OTUs) using UCLUST (Edgar 2010), with a minimum coverage of 99% and a minimum similarity of 97%. A representative sequence was chosen for each phylotype by selecting the longest sequence that had the highest number of hits to other sequences of that particular phylotype. Chimeric sequences were detected with ChimeraSlayer and removed before taxonomic analysis (Edgar et al. 2011). Representative sequences were aligned using PyNAST (Caporaso et al. 2010) against the Greengenes core set (DeSantis et al. 2006). Taxonomic assignments were made using the Ribosomal Database Project (RDP) classiﬁer (Wang et al. 2007) with greengenes representative set of sequences as reference. Sequences classified as “Chloroplast” (0.5%) or “Mitochondria” (10%) were removed from the alignment. A heatmap was generated in R with the gplots package with the relative abundance of genera with a minimal absolute abundance of 50 total sequences (R Core Team, 2014).

# 77#

5.3.5 Community Analysis

To evaluate communities at an equal sequencing depth, all samples were rarified to the lowest number of sequences occurring in a sample (620). An approximate maximum-likelihood tree was constructed from an alignment of representative sequences, using FastTree (Price et al. 2009). An unweighted UniFrac distance matrix was constructed from the phylogenetic tree to analyze dissimilarity of sample communities (Lozupone and Knight 2005). The unweighted Unifrac distances were visualized using principal coordinate analysis (PCoA) and an Unweighted Pair Group Method with Arithmetic mean (UPGMA) tree was created from the UniFrac distance matrix. Analysis of Similarity (ANOSIM) and Permutational Multivariate Analysis of Variance (PERMANOVA), each with 999 permutations, were used to calculate the significance of clustering of samples by location, height, or tree species. Species diversity and richness were assessed using two metrics: Chao1 (for species richness), and the Shannon index (accounting for abundance and evenness).

5.4 Results

5.4.1 Phylotypes Recovered

A total of 642 distinct bacterial phylotypes distributed across 342 bacterial genera were recovered from 44,522 quality sequences from the 26 samples. These sequence data have been submitted to the GenBank databases under project accession number SRP033097. Each sample yielded an average of 1900 sequences after plant DNA was removed (with a range of 626 to 2946 sequences), which is a sufficient sequencing depth for diversity analysis (Kuczynski et al. 2010). The number of phylotypes recovered varied by location, tree species, and tree height. The number of phylotypes recovered did not vary between lower (average of 297 phylotypes) and middle heights (average of 308 phylotypes) but fewer phylotypes were recovered from upper heights, with an average of 190 (Table 5-1). Rarefaction plots did not saturate, indicating that we under-sampled the bacterial communities at the 97% OTU level. (Figure 5-1).

5.4.2 Relative Abundance of Bacterial Taxa

The most abundant phyla in all samples were Proteobacteria and Firmicutes, followed by Acidobacteria, Actinobacteria, TM7 and Bacteroidetes (Figure 5-2). The relative abundance of bacterial phyla varied across samples, but Proteobacteria dominated all samples. Among the most abundant classes were Alpha-, Beta-, Delta-, and Gammaproteobacteria, and Bacilli (Figure 5-2). Bacilli dominated many giant sequoia samples, whereas only a few coast redwood samples had a high occurrence of this class. Conversely, more redwood than giant sequoia samples were dominated by Alphaproteobacteria. In all samples combined the Alphaproteobacteria consisted mostly of taxa belonging to the families Methylobacteriaceae and Acetobacteraceae (Figure 5-1B), the Bacilli by Bacillaceae (Figure 5-1C) the Deltaproteobacteria by

# 78#

Burkholderiaceae, Comamonadaceae and Oxalobacteraceae (Figure 5-1D), and the Betaproteobacteria by Enterobacteriaceae and Pseudomonadaceae (Figure 5-1C). Underlying variation in dominant classes across samples and species is variation in dominance by specific bacterial genera and OTUs. (Figure 5-3A,B). Most giant sequoia samples, as well as the coast redwood samples from Big Creek had a high relative abundance of bacteria in the genus Bacillus (Figure 5-3A). An OTU most similar to the soil bacterium Bacillus pocheonensis (Ten et al. 2007) (98%) was found in all three giant sequoia all trees, and in all but one sample, but only in one coast redwood sample (OTU 1916 in Figure 5-3B). Bacillus firmus was present in most giant sequoia samples and in coast redwood from Big Creek, where it made up a substantial part of the community (e.g. 24, 43 and 37% of sequences in three of the samples; Figure 5-3B). Many coast redwoods samples, especially those from Samuel P. Taylor SP were dominated by bacteria belonging to the genus Methylobacterium, which are commonly associated with plants, both in the phyllosphere, and as endophytes. One specific OTU that was only distantly related to known Methylobacterium spp. was present at high relative abundance in many coast redwood samples, while present below 1% in giant sequoia. Other genera present in high abundances were Herbaspirillum, Kozakia, Pseudomonas, Roseateles, Sphingomonas, and Candidate TM7. An OTU closely related to the species Sodalis glossinidus, made up half of the sequences in an upper canopy sample from a giant sequoia. No OTU was consistently present in all samples or within all samples from a species.

5.4.3 Community Structure

Alpha diversity (the diversity within single foliage sample) was estimated by Chao and Shannon indices (Figure 5-4A). We found the Chao index to be greater in the middle and lower height samples than upper height samples of coast redwood trees. Similarly, we found the Shannon index to be greater in the middle and lower height samples of coast redwood trees (Figure 5-4B). This suggests that diversity and richness decreases with height in coast redwood. We found no difference in diversity and richness across heights in giant sequoia (Figure 5-4A,B) . Chao and Shannon indexes were also greater in coastal redwoods than giant sequoia. Beta diversity (diversity among hosts, species and locations) was assessed with the unweighted UniFrac distance matrix and visualized with a PCoA plot. Clear clustering occurred by species (Permanova: Pseudo-F statistic = 3.0348, p=0.001; Anosim: R = 0.5548, P = 0.001), and by location (Permanova: Pseudo-F statistic = 2.8816, P = 0.001; Anosim: R = 0.3977, P = 0.001), with some separation between samples from the two coast redwood locations (Figure 5-5). There was no clustering based on canopy height (Permoanova: Pseudo-F statistic = 1.15, P = 0.192; Anosim: R = 0.211, P = 0.211) (Figure 5-5). Jackknife analysis of the unweighted UniFrac UPGMA tree showed support for the distinction between bacterial communities by tree species (Figure 5-5).

# 79#

5.5 Discussion

We used 16S rRNA pyrosequencing to test if bacterial needle endophytes communities of coast redwood and giant sequoia are stable or variable within and across host individuals and populations. In contrast to what we previously observed in subalpine conifers, endophytic communities were not simple and constant across individuals and species, but diverse, and varied extensively with canopy height, and across individuals, species and geographic locations. We found that the related species coast redwood and giant sequoia host distinct bacterial endophyte communities. PCoA plots depicted clustering by tree species with separation between the two coast redwood locations. We did not find canopy height to structure the endophyte community, but found that the diversity varied in redwood trees with lower diversity in the upper canopy compared to middle and lower canopy. This could be the result of differences in the environment of redwood trees, e.g. reduced gas exchange and lower carbon assimilation with increased tree height (Ambrose et al. 2010), that result in stress gradients across the canopy height. Alternatively, if endophytes colonize the upper canopy mainly by phloem and xylem, and not via the leaf surface, dispersal to the upper canopy could be limited, resulting in lower diversity.

We found our samples to be dominated by Alphaproteobacteria, a class of bacteria that are commonly associated with plants. Many genera we recovered have previously been detected as endophytes of forest trees: Pedobacter (Ulrich et al. 2008), Bacillus (Izumi et al. 2008), (Moore et al. 2006). Paenibacillus(Moore et al. 2006, Ferreira et al. 2008, Izumi et al. 2008, Ulrich et al. 2008), Methylobacterium (Pirttilä et al. 2005, Ulrich et al. 2008), Sphingomonas (Moore et al. 2006, Ulrich et al. 2008, Filteau et al. 2010), Burkholderia (Moore et al. 2006, Doty et al. 2009, Procópio et al. 2009), Herbaspirillum (Doty et al. 2009), Acinetobacter (Moore et al. 2006, Izumi et al. 2008, Doty et al. 2009, Procópio et al. 2009), and Pseudomonas (Moore et al. 2006, Ferreira et al. 2008, Ulrich et al. 2008, Doty et al. 2009, Procópio et al. 2009, Filteau et al. 2010). The dominant or highly abundant phylotypes recovered have also been found as endophytes in other plants.

A genus found in many samples was Bacillus. Bacteria from this genus have been found to produce antimicrobial compounds to aid in the defense of pathogens (Compant et al. 2010). B. firmus, which was found in high elative abundance in giant sequoia and coast redwood from Big Creek, is antagonistic against root-knot nematode pests on crop plants (Giannakou et al. 2004, Lamovšek et al. 2013). Moreover, Bacillus sphaericus and Bacillus pumilus was found to induce systemic protection on loblolly pine (Pinus taeda) against Cronartium quercuum, the causal agent of fusiform rust disease (Enebak and Carey 2000), and Bacillus subtilis colonizes trees as an endophyte where it can inhibit the growth of the pathogenic fungus Ceratocystis parasitica (Schreiber, L.R. et al. 1988, Wilhelm, E. et al. 1998). It is possible that the main way in which Bacillus and other endophytes inhibit insect pests in conifer is through antagonistic effects on the fungal symbionts of insects. For example, B. pumilus isolated from fresh, living phloem of P. contorta uncolonized by the mountain pine beetle (Dendroctonus ponderosae) showed a strong antagonistic effect against the pine beetle fungal symbiont Ophiostoma montium

# 80#

(Adams et al. 2008). Some strains of B. pumilus are known to produce proteolytic enzymes that degrade the fungal cell wall (Kumar 2002, Nielsen and Sørensen 2006).

The occurrence of Bacillus strains in our samples is variable with one OTU dominating giant sequoia samples only, and another dominating giant sequoia and coast redwood from one location, suggesting that the defensive endophyte community may change in response to pathogen and/or pest challenges.

Another genus that was prominent in many samples was Methylobacterium. (Ardanov et al. 2012). Giant sequoia samples contained <1% of Methylobacterium while many coast redwood samples were dominated by a Methylobacterium species (OTU 1726). Species from this genus have been linked to growth in conifers (Pohjanen, J. et al. 2014) and pathogen defense in potato Furthermore, a Methylobacterium sp. isolated from potato displayed antagonistic activity against Verticillum dahliae, a plant pathogen (Sessitsch et al. 2004).

Pseudomonas is another genus known to be effective in biocontrol of pathogens (Compant et al. 2010). A Pseudomonas fluorescens-like taxon (OTU 117) was present at high relative abundance in many samples from coast redwood and giant sequoia. P. fluorescens as well as other Pseudomonas species in the potato root endosphere have been found to be antagonistic to the widely distributed plant pathogenic fungi Verticillium dahliae and Rhizoctonia solani, (Berg et al. 2005). P. fluorescens also significantly reduced the incidence of American bollworm (Helicoverpa armigera) infestation in cotton (Rajendran et al. 2007). Other phylotypes present in high abundances were OTU 1403, 99% similar to Herbaspirillum rhizosphaerae, a plant-growth promoting endophyte in rice and sugarcane, and OTU 2805, which is 98% similar to Kozakia balinesis, a species that has been found as an endophyte in conifers (Carrell and Frank 2014). Finally, half of the sequences from one sample from the upper canopy of a giant sequoia belonged to the genus Sodalis. Bacteria in this genus have been found exclusively as insect symbionts, but infect diverse insect orders (Snyder et al. 2011), including Coleoptera (beetles), where it has been found in the gut microbiome (Grünwald et al. 2010). It is possible that the domination of a Sodalis sp. in an apparently healthy sample is due to nearby infestation of the sequoia bark beetle (Phloeosinus rubicundulus).

The pattern of endophyte diversity observed here is remarkably different from what recently has been observed in subalpine limber pine and Engelmann spruce (Carrell and Frank 2014). In those species, the endophyte community was found to be considerably less diverse and consistently dominated by a single OTU. Although only nine trees were examined in the present study, it is clear that coast redwood and giant sequoia do not host a similarly stable endophyte community. Therefore different mechanisms likely structure bacterial endophyte communities in the two systems. While the dominant endophytic taxa in limber pine and Engelmann spruce are closely related to known endophytic diazotrophs, the bacteria found in high relative abundances here were closely related to

# 81# bacteria known to inhibit or reduce incidences of pathogenic fungi and/or pests. High elevation pines grow well in N limited soil (Rueth and Baron 2002), and may depend on endophytic diazotrophs (Anand et al. 2013) that may fix N in the canopy (Carrell and Frank 2014), coast redwood may acquire N from fog (Ewing et al. 2009). Alternatively, the patterns observed here could arise if the giant trees are host to a majority of transiently visiting microbes with less significant roles in host biology.

Carroll (Carroll 1988) suggested that long-lived trees form mutualistic associations with endophytic fungi inside the foliage, enabling them to resist fungal pathogens and insect pests over the course of centuries. He argued that the short life cycle and rapid adaptation of endophytic fungi may help the host overcome the challenge from short-cycle pests and pathogens, and that in addition, tree hosts might accommodate different endophyte species in response to challenges from different pests and pathogens (Carroll 1988). Similarly, the role of microbes—both pathogenic and non-pathogenic—in induced resistance is increasingly recognized as an important defense mechanism in forest trees (Bonello et al. 2006, Eyles et al. 2010). It has also been suggested that somatic mutations give rise to mosaics of genetic and defense chemical variation within an individual plant. However, Hall and Langenheim found no evidence of within-tree spatial , or temporal (Hall and Langenheim 1986) variation in coast redwood monoterpenes.

Alternatively, or in addition to the hypothesis that variability fungal endophyte genomes and communities contribute to coast redwood’s high resistance to pathogens and subsequent longevity, bacterial endophytes could play similar roles. The endophyte community observed here, with variation within and among host individuals, species, and locations is consistent with Carroll’s hypothesis about microbially mediated host defense (Carroll 1988). While endophytic communities varied within and among individuals, a high relative abundance of at least one group of taxa known to produce anti-fungals and anti-pest compounds was seen across samples (e.g. Bacillus, Methylobacterium, Pseudomonas, Herbaspirillum) However, further research is needed to test if the endophytic community and giant sequoias is involved in suppressing biotic stress. For example, the heterogeneity of plant defense compounds synthesized by microbes could be investigated.

5.6 References

Adams, A. S., D. L. Six, S. M. Adams, and W. E. Holben. 2008. In vitro interactions between yeasts and bacteria and the fungal symbionts of the mountain pine beetle (Dendroctonus ponderosae). Microb Ecol 56:460–6.

Ambrose, A. R., S. C. Sillett, G. W. Koch, R. Van Pelt, M. E. Antoine, and T. E. Dawson. 2010. Effects of height on treetop transpiration and stomatal conductance in coast redwood (Sequoia sempervirens). Tree Physiology 30:1260– 1272.

# 82#

Anand, R., and C. P. Chanway. 2013. N2-fixation and growth promotion in cedar colonized by an endophytic strain of Paenibacillus polymyxa. Biol Fert Soils 49:235–239.

Anand, R., S. Grayston, and C. Chanway. 2013. N2-Fixation and Seedling Growth Promotion of Lodgepole Pine by Endophytic Paenibacillus polymyxa. Microbial Ecology 66:369–374.

Ardanov, P., A. Sessitsch, H. Häggman, N. Kozyrovska, and A. M. Pirttilä. 2012. Methylobacterium-Induced Endophyte Community Changes Correspond with Protection of Plants against Pathogen Attack. PLoS ONE 7:e46802.

Arnold, A. E., L. C. Mejia, D. Kyllo, E. I. Rojas, Z. Maynard, N. Robbins, and E. A. Herre. 2003. Fungal endophytes limit pathogen damage in a tropical tree. Proc Natl Acad Sci U S A 100:15649–54.

Berg, G., A. Krechel, M. Ditz, R. A. Sikora, A. Ulrich, and J. Hallmann. 2005. Endophytic and ectophytic potato-associated bacterial communities differ in structure and antagonistic function against plant pathogenic fungi. FEMS Microbiology Ecology 51:215–229.

Bonello, P., T. R. Gordon, D. A. Herms, D. L. Wood, and N. Erbilgin. 2006. Nature and ecological implications of pathogen-induced systemic resistance in conifers: A novel hypothesis. Physiological and Molecular Plant Pathology 68:95–104.

Caporaso, J. G., K. Bittinger, F. D. Bushman, T. Z. DeSantis, G. L. Andersen, and R. Knight. 2010. PyNAST: a flexible tool for aligning sequences to a template alignment. Bioinformatics 26:266–7.

Carroll, G. 1988. Fungal Endophytes in Stems and Leaves: From Latent Pathogen to Mutualistic Symbiont. Ecology 69:2–9.

Carroll, G. C., and F. E. Carroll. 1978. Studies on the incidence of coniferous needle endophytes in the Pacific Northwest. Canadian Journal of Botany 56:3034–3043.

Chelius, M. K., and E. W. Triplett. 2001. The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microb Ecol 41:252–263.

# 83#

Compant, S., C. Clement, and A. Sessitsch. 2010. Plant growth-promoting bacteria in the rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biology and Biochemistry 42:669–678.

Compant, S., B. Duffy, J. Nowak, C. Clement, and E. A. Barka. 2005. Use of plant growth-promoting bacteria for biocontrol of plant diseases: principles, mechanisms of action, and future prospects. Appl Environ Microbiol 71:4951–9.

Doty, S. L., B. Oakley, G. Xin, J. W. Kang, G. Singleton, Z. Khan, A. Vajzovic, and J. T. Staley. 2009. Diazotrophic endophytes of native black cottonwood and willow. Symbiosis 47:23–33.

Edgar, R. C. 2010. Search and clustering orders of magnitude faster than BLAST. Bioinformatics 26:2460–1.

Edgar, R. C., B. J. Haas, J. C. Clemente, C. Quince, and R. Knight. 2011. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics 27:2194– 200.

Enebak, S. A., and W. A. Carey. 2000. Evidence for induced systemic protection to fusiform rust in loblolly pine by plant growth-promoting rhizobacteria. Plant Disease 84:306–308.

Espinosa-Garcia, F. J., and J. H. Langenheim. 1990. The endophytic fungal community in leaves of a coastal redwood population diversity and spatial patterns. New Phytologist 116:89–97.

Ewing, H. A., K. C. Weathers, P. H. Templer, T. E. Dawson, M. K. Firestone, A. M. Elliott, and V. K. S. Boukili. 2009. Fog Water and Ecosystem Function: Heterogeneity in a California Redwood Forest. Ecosystems 12:417–433.

Eyles, A., P. Bonello, R. Ganley, and C. Mohammed. 2010. Induced resistance to pests and pathogens in trees. New Phytologist 185:893–908.

Ferreira, A., M. C. Quecine, P. T. Lacava, S. Oda, J. L. Azevedo, and W. L. Araújo. 2008. Diversity of endophytic bacteria from Eucalyptus species seeds and colonization of seedlings by Pantoea agglomerans. FEMS microbiology letters 287:8–14.

Filteau, M., L. Lagacé, G. LaPointe, and D. Roy. 2010. Seasonal and regional diversity of maple sap microbiota revealed using community PCR fingerprinting and 16S rRNA gene clone libraries. Systematic and applied microbiology 33:165–173.

# 84#

Friesen, M. L., S. S. Porter, S. C. Stark, E. J. von Wettberg, J. L. Sachs, and E. Martinez- Romero. 2011. Microbially Mediated Plant Functional Traits. Annual Review of Ecology, Evolution, and Systematics 42:23–46.

Ganley, R. J., S. J. Brunsfeld, and G. Newcombe. 2004. A community of unknown, endophytic fungi in western white pine. Proc Natl Acad Sci U S A 101:10107–12.

Giannakou, I. O., D. G. Karpouzas, and D. Prophetou-Athanasiadou. 2004. A novel non- chemical nematicide for the control of root-knot nematodes. Applied Soil Ecology 26:69–79.

Grünwald, S., M. Pilhofer, and W. Höll. 2010. Microbial associations in gut systems of wood- and bark-inhabiting longhorned beetles [Coleoptera: Cerambycidae]. Systematic and Applied Microbiology 33:25–34.

Hall, G. D., and J. H. Langenheim. 1986. Temporal changes in the leaf monoterpenes of Sequoia sempervirens. Biochemical Systematics and Ecology 14:61–69.

Hardoim, P. R., L. S. van Overbeek, and J. D. Elsas. 2008. Properties of bacterial endophytes and their proposed role in plant growth. Trends Microbiol 16:463–71.

Izumi, H., I. C. Anderson, K. Killham, and E. R. Moore. 2008. Diversity of predominant endophytic bacteria in European deciduous and coniferous trees. Canadian journal of microbiology 54:173–179.

Jiao, J. Y., H. X. Wang, Y. Zeng, and Y. M. Shen. 2006. Enrichment for microbes living in association with plant tissues. J Appl Microbiol 100:830–7.

Kuczynski, J., Z. Liu, C. Lozupone, D. McDonald, N. Fierer, and R. Knight. 2010. Microbial community resemblance methods differ in their ability to detect biologically relevant patterns. Nature Methods 7:813–819.

Kumar, C. G. 2002. Purification and characterization of a thermostable alkaline protease from alkalophilic Bacillus pumilus. Letters in Applied Microbiology 34:13–17.

Lamovšek, J., G. Urek, and S. Trdan. 2013. Biological Control of Root-Knot Nematodes (Meloidogyne spp.): Microbes against the Pests / BIOTIČNO ZATIRANJE OGORČIC KORENINSKIH ŠIŠK (Meloidogyne spp.): MIKROORGANIZMI PROTI ŠKODLJIVCEM. Acta agriculturae Slovenica 101.

Lozupone, C., and R. Knight. 2005. UniFrac: a new phylogenetic method for comparing microbial communities. Appl Environ Microbiol 71:8228–35.

# 85#

Moore, F. P., T. Barac, B. Borremans, L. Oeyen, J. Vangronsveld, D. Van der Lelie, C. D. Campbell, and E. R. Moore. 2006. Endophytic bacterial diversity in poplar trees growing on a BTEX-contaminated site: the characterisation of isolates with potential to enhance phytoremediation. Systematic and applied microbiology 29:539–556.

Nielsen, P., and J. Sørensen. 2006. Multi-target and medium-independent fungal antagonism by hydrolytic enzymes in Paenibacillus polymyxa and Bacillus pumilus strains from barley rhizosphere. FEMS Microbiology Ecology 22:183– 192.

Pirttilä, A. M., H. Pospiech, H. Laukkanen, R. Myllyla, and A. Hohtola. 2005. Seasonal variations in location and population structure of endophytes in buds of Scots pine. Tree Physiol 25:289–97.

Pohjanen, J., Koskimäki, J.J., Sutela, S., Ardanov, P., Suorsa, M., Niemi, K., Sarjala, T., Häggman, H., and Pirttilä, A.M. 2014. The interaction with ectomycorrhizal fungi and endophytic Methylobacterium affects the nutrient uptake and growth of pine seedlings in vitro:In press.

Price, M. N., P. S. Dehal, and A. P. Arkin. 2009. FastTree: computing large minimum evolution trees with profiles instead of a distance matrix. Mol Biol Evol 26:1641– 50.

Procópio, R. E. L., W. L. Araújo, W. Maccheroni Jr, and J. L. Azevedo. 2009. Characterization of an endophytic bacterial community associated with Eucalyptus spp. Genet Mol Res 8:1408–1422.

Raghavendra, A. K. H., and G. Newcombe. 2013. The contribution of foliar endophytes to quantitative resistance to Melampsora rust. The New phytologist 197:909–918.

Rajendran, L., R. Samiyappan, T. Raguchander, and D. Saravanakumar. 2007. Endophytic bacteria mediate plant resistance against cotton bollworm. Journal of Plant Interactions 2:1–10.

Reinhold-Hurek, B., and T. Hurek. 2011. Living inside plants: bacterial endophytes. Curr Opin Plant Biol 14:435–43.

# 86#

Rodriguez, R. J., J. F. White Jr., A. E. Arnold, and R. S. Redman. 2009. Fungal endophytes: diversity and functional roles. New Phytol 182:314–30.

Rollinger, J. L., and J. H. Langenheim. 1993. Geographic Survey of Fungal Endophyte Community Composition in Leaves of Coastal Redwood. Mycologia 85:149.

Rosenblueth, M., and E. Martinez-Romero. 2006. Bacterial endophytes and their interactions with hosts. Mol Plant Microbe Interact 19:827–37.

Rueth, H. M., and J. S. Baron. 2002. Differences in Englemann Spruce Forest Biogeochemistry East and West of the Continental Divide in Colorado, USA. Ecosystems 5:45–57.

Sawyer, J.O., Sillett, S.C., Popenoe, J.H., LaBanca, A., Sholars, T., Largent, D.F., Euphrat, F., R.F. Noss, and Van Pelt, R. 2000. Characteristics of redwood forests. Pages 39–79 The redwood forest: history, ecology, and conservation of the coast redwoods. Island Press, Covelo, California, USA.

Schreiber, L.R., Gregory, G.F., Krause, C.R., and Ichida, J.M. 1988. Production, partial purification, and antimicrobial activity of a novel antibiotic produced by a Bacillus subtilis isolate from Ulmus americana. Can J Bot 66:2338–2346.

Sessitsch, A., B. Reiter, and G. Berg. 2004. Endophytic bacterial communities of field- grown potato plants and their plant-growth-promoting and antagonistic abilities. Canadian Journal of Microbiology 50:239–249.

Snyder, A. K., C. M. McMillen, P. Wallenhorst, and R. V. M. Rio. 2011. The phylogeny of Sodalis-like symbionts as reconstructed using surface-encoding loci: Insect symbiont phylogeny using outer membrane genes. FEMS Microbiology Letters 317:143–151.

Ten, L. N., S.-H. Baek, W.-T. Im, L. L. Larina, J.-S. Lee, H.-M. Oh, and S.-T. Lee. 2007. Bacillus pocheonensis sp. nov., a moderately halotolerant, aerobic bacterium isolated from soil of a ginseng field. INTERNATIONAL JOURNAL OF SYSTEMATIC AND EVOLUTIONARY MICROBIOLOGY 57:2532–2537.

Ulrich, K., A. Ulrich, and D. Ewald. 2008. Diversity of endophytic bacterial communities in poplar grown under field conditions. FEMS microbiology ecology 63:169–180.

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 73:5261–7.

Wilhelm, E., Arthofer, W., Schafleitner, R., and Krebs, B. 1998. Bacillus subtilis an endophyte of chestnut (Castanea sativa) as antagonist against chestnut blight (Cryphonectria parasitica). Plant Cell Tiss Org Cult 52:105–108.

# 87#

5.7 Tables

Table 5-1 Samples successfully characterized by 16S rRNA in this study with number of sequences after quality control and removal of plant DNA.

Tree species Sampling Tree ID Canopy No. of No. of 97% Location height Sequences phylotypes Giant sequoia Freeman 695 L 2946 395 Creek U 2325 268 696 L 2540 249 M 1801 320 U 2171 388 1139 L 2315 300 M 1888 380 U 1888 337 Coast redwood Big 1202 L 1342 109 Creek M 2357 242 U 1563 120 1206 L 2641 478 M 1095 102 U 1156 140 1321 L 1254 81 M 1982 343 U 955 138 Samuel 1513 L 2807 247 P. Taylor M 1883 287 SP U 778 108 1515 L 2251 347 M 2523 384 U 1023 100 1751 L 2847 468 M 2431 407 U 626 119

# 88#

5.8 Figures

FC Middle ST Middle ST Lower FC Upper FC Lower 150

BC Middle BC Lower BC Upper

100 ST Upper OTUs 50 0 100 200 300 400 500 600 Number of Sequences

Figure 5-1 Rarefaction curves indicating the number of OTUs at 97% similarity for lower, middle, and upper heights of trees at each location. FC: Freeman Creek (giant sequoia), BC: Big Creek (coast redwood), ST: Samuel P Taylor (coast redwood). The rarefaction curves do not asymptote indicating sampling was not sufficient.

# 89#

A Acidobacteria Actinobacteria Alphaproteobacteria Bacilli Betaproteobacteria Clostridia Deltaproteobacteria Fusobacteria Gammaproteobacteria Sphingobacteriia TM7-3 Other Relative Abundance Relative 0695L 0696L 1139L 1202L 1202U 1206L 1206M 1206U 1321L 1321U 1513L 1515M 1515U 1751L 1751M 1751U 1515L 0696U 1139M 1513U 0695U 0696M 1139U 1202M 1321M 1513M giant sequoia coast redwood

B C 1% 3% D <1% E 2% 1% 6% 2% 3% 11% 26% 11% 26% 36% 26% 6%

52% 5% 4% 12% 62% 6% 3% 11% 46% 34% 5%

Acetobacteraceae Bacillaceae Burkholderiaceae Enterobacteriaceae Beijerinckiaceae Gemellaceae Comamonadaceae Moraxellaceae Caulobacteraceae Paenibacillaceae Neisseriaceae Pasteurellaceae Methylobacteriaceae Planococcaceae Oxalobacteraceae Pseudomonadaceae Sphingomonadaceae Staphylococcaceae Other Sinobacteraceae Thermoactinomycetaceae Xanthomonadaceae Other Turicibacteraceae Other Other

Figure 5-2 A: Bar chart showing relative abundance of major bacterial phyla (and classes for Proteobacteria) from coast redwood and giant sequoia trees calculated as percentage per sample. Each bar represents a tree sample. B-E: Pie charts showing the distribution of bacterial families from the four most abundant classes. B: Alphaproteobacteria, C: Bacilli, D: Betaproteobacteria, E: Gammaproteobacteria.

# 90#

Giant sequoia Coast redwood Big Creek Coast redwood Samuel P.T. Genus Class Acidobacterium A Granulicella Acidobacteria Terriglobus Corynebacterium Frankia Frondihabitans %Abund. Microbacterium Rothia Actinobacteria 60 Nocardioides Actinomycetospora Bifidobacterium

y Sediminibacterium Hymenobacter e 40 Mucilaginibacter Sphingobacteria

Pedobacter r Waddlia Chlamydiae

o Bacillus l Natronobacillus o 20 Oceanobacillus

C Paenibacillus Lynsinibacillus Solibacillus Sporosarcina Bacilli 0 Salinicoccus Staphylococcus Gemella Lactobacillus Streptococcus Turicibacter Finegoldia Clostridia Clostridium Leptotrichia Caulobacter Fusobacteria Bartonella Microvirga Methylocella Balneimonas Prosthecomicrobium Methylobacterium Rhodoblastus Acidisoma a-proteobacteria Acidisphaera Gluconacetobacter Granulibacter Kozakia Swaminathania Tanticharoenia Wolbachia Sphingomonas Burkholderia Pandoraea Roseateles Rubrivivax Rhodoferax b-proteobacteria Herbaspirillum Herminiimonas Massillia Ralstonia Neisseria Sorangium Desulfofrigus d-proteobacteria Geobacter Syntrophobacter Escherichia Providencia Sodalis Aggregatibacter g-proteobacteria Haemophilus Acinetobacter Pseudomonas Nitrocococcus TM7 TM7 0695L 0696L 1139L 1202L 1206L 1321L 1513L 1751L 1515L 0695U 0696U 1139U 1202U 1206U 1321U 1513U 1515U 1751U 0696M 1139M 1202M 1206M 1321M 1513M 1515M 1751M Giant sequoia Coast redwood Big Creek Coast redwood Samuel P.T. B ID % ID #OTU 0695L 0696L 1139L 1202L 1206L 1321L 1513L 1515L 1751L 0695U 0696U 1139U 1202U 1206U 1321U 1513U 1515U 1751U 0696M 1139M 1202M 1206M 1321M 1513M 1515M Closest match 1751M Uncultured TM7 96 0 0 0 0 0 0 0 0 0 0 0 3 0 0 0 0 0 1 8 0 0 13 0 9 1 9 74 Pseudomonas fluorescens 99 2 0 0 1 2 1 0 2 5 3 7 0 4 5 8 2 10 0 0 2 0 0 9 0 0 0 117 Syntrophobacter spp. 95 15 2 0 8 5 3 2 3 0 6 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 348 Herbaspirillum rhizosphaerae 99 0 0 6 3 0 14 2 1 3 0 3 2 1 0 2 0 2 0 1 7 0 0 1 1 3 0 1403 Methylobacterium oryzae 92 0 0 0 0 0 0 0 0 1 12 0 22 0 0 0 22 11 16 34 6 19 26 0 7 13 6 1726 Bacillus pocheonensis 98 8 6 0 7 7 1 7 6 0 0 0 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1916 Terriglobus roseus 98 0 0 1 3 0 0 0 1 0 1 0 4 0 0 0 0 1 3 3 1 4 3 0 5 4 1 2511 Kozakia baliensis 98 1 0 16 0 0 5 3 1 0 5 0 0 0 0 0 3 3 0 0 0 7 0 0 0 1 0 2805 Pandoraea pnomenusa 99 0 0 22 1 0 13 1 0 0 0 0 0 0 0 0 0 13 0 0 0 9 0 4 0 0 0 2970 Sodalis glossinidius 99 0 48 0 0 0 0 0 0 0 2 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3287 Bacillus firmus 100 6 1 0 13 10 0 1 2 24 3 43 0 37 3 11 2 1 0 0 21 1 0 0 0 0 0 3293 Roseateles terrae 98 2 0 0 3 2 1 0 1 12 4 10 0 9 5 8 2 5 0 0 6 0 0 18 1 1 0 3526

Figure 5-3A: Heatmap displaying the relative abundance of bacterial genera across all tree samples. Only genera with an abundance of 50 sequences were included. B: Heatmap displaying the 12 most dominant OTUs and their average relative abundances as percentages of all 16S rRNA gene sequences recovered in our samples. Color tones range from cool (blue) to warm (red) to indicate the lowest to highest relative abundance values.

# 91#

A Chao1 Species

Lower Middle Upper Lower Middle Upper

Coast Redwoods Giant Sequoia B Shannon Diversity Shannon Diversity

Lower Middle Upper Lower Middle Upper

Coast Redwoods Giant Sequoia

Figure 5-4 Boxplot showing (A) Chao1 and (B) Shannon Diversity indexes of our samples at the 97% similarity and rarified to 620 sequences per sample. Values greater than 1.5 times the mean are indicated with a plus sign (+).

# 92#

A B Giant Sequoia Freeman Creek Coast Redwood Big Creek Samuel P. Taylor

C D Upper Middle Lower

Giant sequoia 0.04 Coast redwood

Figure 5-5 PCoA of an unweighted UniFrac distance matrix. Points that are closer on the PCoA are more similar phylogenetically. Each point represents the bacterial community of a sample colored by (A) tree species, (B) location and (C) canopy height. (D) Hierarchial clustering analysis by UPGMA of all tree samples. Leaves are labeled by shape according to tree species: square, giant sequoia; circle, coast redwood.

# #

# 6 Conclusion

# # The overall aim of my dissertation was to determine if sub-alpine conifers host a mutualistic core of endophytic bacteria.

In my first chapter I found that sub-alpine conifers Pinus flexilis and Picea engelmannii foliage in Niwot Ridge, CO are consistently dominated by the same phylotype (here called AAB1 for acetic acid bacteria 1), which made up 19-53% and 14-39% of the sequences in P. flexilis and P. engelmannii respectively. In my second chapter I found P. flexilis and co-occurring Pinus contorta foliage to be dominated by the AAB1 phylotype in Niwot Ridge, CO and Horseshoe Meadows, CA. I also found AAB1 in P. contorta foliage in Tuolumne Meadows but not in high abundance. The presence of AAB1 in both Pinus and Picea across sites is most consistent with a long-standing association between trees in the pine family and acetic acid bacterial endophytes, predating the divergence of the genera Picea and Pinus. Consistent presence of the AAB1 phylotype indicates a potential mutualistic role in Pinus and Piceae health and its presence across sites suggests it may be co-evolved. In my last chapter, I found giant trees were not colonized by AAB1 suggesting that the AAB1 symbiosis may have originated after the divergence of Pinaceae and the Cupressaceae. Co-speciation of host and their associated microbes has been detected in animals but evidence for co-speciation has not previously been detected in the plant-microbe symbiosis. The results of my dissertation suggest the association of AAB1 and Pinaceae may be a candidate for the first described plant-microbe co-speciation event.

The 16S rRNA gene of AAB1 is most similar to a known endophytic N2 fixer of sugarcane Gluconacetobacter diazotrophicus presenting the possibility that AAB1 is a N2 fixer in subalpine forests. Endophytic N2 fixation in forest conifers could hold the answer to a long-standing ecological mystery: unknown inputs to boreal and temperate forests. These forests have few nodulated N2 fixing plants but have been found to acquire N at rapid rates, implying additional sources of N (Binkley et al. 2000, Bormann et al. 2002). It is generally presumed that conifers get their N from the soil and while uptake of amino acids and proteins has been demonstrated, the contribution of organic N to plant N nutrient is unclear (Näsholm et al. 2009). Rhizospheric N2 fixation has been suggested as an N source (Bormann et al. 1993), but the activity has been found too low to support N requirement, at least in pines (Chanway and Holl 1991). In recent experiments, conifer seedlings inoculated with Paenibacillus polymyxa P2b-2R, an N2 fixing strain isolated

93# 94# from P. contorta growing in N limited soil (Bal et al 2012) were found to acquire significant amounts of seedling foliar N from the atmosphere when compared to control seedlings (Bal and Chanway 2012, Bal et al. 2012, Anand et al. 2013). In light of this and the dominant pattern of AAB1, we may have uncovered a possible source of N in the subalpine ecosystem. By providing N to subalpine conifers, AAB1 may have a role in the adaptation to the N poor environment.

If AAB1 is a mutualistic bacterium, the relationship likely facultative since its presence was negligible in P. contorta at Tuolumne Meadows, CA; instead, an Enterobacter sp. and a Citrobacter sp. phylotype dominated the samples. Enterobacter sp. is a known N2 fixer of sugarcane and thus may replace the function of AAB1 in P. contorta at Tuolumne Meadows. Alternatively, the abundance of AAB1 may be dependent on N availability. N2 fixation is energetically expensive and symbiotic N2 fixation is often inhibited when other N sources are available. For example, addition of N leads to significant reductions or total exclusion of N2 fixation in moss-cyanobacteria associations (Ackermann et al. 2012) .

In chapter four I tested if endophytic bacterial communities of coast redwoods (Sequoia sempervirens) and giant sequoia (Sequoiadendron giganteum) were similarly stable as in sub-alpine conifers or, alternatively, variable within and across individuals and populations. Bacterial endophyte communities of coast redwood and giant sequoia were considerably more diverse and variable, and the trees did not host a core of bacterial taxa that were consistently associated with the host. No single phylotype or genus dominated all samples of a species suggesting giant trees are largely colonized by horizontally transferred bacteria. The ‘rare’ endophyte community structure (excluding AAB1) of subalpine conifers resembled the structure of coast redwood and giant sequoia endophytic communities, with less consistency across species and location had a strong influence on community structure. Perhaps that ‘rare’ community plays similar role for the host tree as the communities in the sequoias with roles such as defense against pathogens and pests. Changes in the environmental may change the endophytic community as they are structured by location, but it is not clear how these changes with impact the tree host or the symbiotic relationship. Next, the impact of environmental change on endophytic community structure needs to be examined to continue to elucidate the role of endophytic bacteria in plant adaptation and the potential impacts of environmental perturbations on the symbiotic relationship of endophytic bacteria and subalpine conifers.

Overall my dissertation found sub-alpine conifers consistently hosted a core endophytic microbial community with the remaining community structured by the environment: potentially revealing a long-standing relationship of a potential N2 fixer, AAB1, and subalpine conifers. Furthermore, I may have uncovered a new source of N (AAB1) to forests.

The presence of AAB1 may provide an ecological advantage to associated trees by providing access to an otherwise limiting nutrient. If the role of AAB1 is to fix N2, the association would influence nutrient cycling in the ecosystem. Nutrient cycles are

# 95# important as they regulate the availability of nutrients in the system and drive ecosystem processes. Without N limitations, the tree would be able to assimilate more carbon into biomass as well as increase the amount of nitrogen in the plant. When considering climate change, the ability to sequester more carbon may help mediate elevated CO2 levels in subalpine conifer systems. If the role of the ‘rare’ community is resistance to pests and pathogens and is structured by site, the ecological consequences of endophytic bacterial communities will vary across ecosystems with some ecosystems more resistant to various pests and pathogens. Given the breadth of impacts of endophytic bacteria, when considering organism to ecosystem level ecological questions, we need to include the bacterial endophytes of plants.

6.1 References

Ackermann, K., O. Zackrisson, J. Rousk, D. L. Jones, and T. H. DeLuca. 2012. N2 Fixation in Feather Mosses is a Sensitive Indicator of N Deposition in Boreal Forests. Ecosystems 15:986–998.

Anand, R., S. Grayston, and C. Chanway. 2013. N2-Fixation and Seedling Growth Promotion of Lodgepole Pine by Endophytic Paenibacillus polymyxa. Microbial Ecology 66:369–374.

Bal, A., R. Anand, O. Berge, and C. P. Chanway. 2012. Isolation and identification of diazotrophic bacteria from internal tissues of Pinus contorta and Thuja plicata. Can J Forest Res 42:807–813.

Bal, A., and C. P. Chanway. 2012. Evidence of nitrogen fixation in lodgepole pine inoculated with diazotrophic Paenibacillus polymyxa. Botany 90:891–896.

Binkley, D., S. Yowhan, and D. W. Valentine. 2000. Do Forests Receive Occult Inputs of Nitrogen? Ecosystems 3:321–331.

Bormann, B. T., C. K. Keller, D. Wang, and H. Bormann. 2002. Lessons from the Sandbox: Is Unexplained Nitrogen Real? Ecosystems 5:727–733.

Näsholm,#T.,#K.#Kielland,#and#U.#Ganeteg.#2009.#Uptake#of#organic#nitrogen#by#plants.# New#Phytologist#182:31–48.#