Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Examining ecological determinants of community formation and stability in the root microbiota Aleklett, Kristin Anna Eva 2015

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

Item Metadata

Download

Media
24-ubc_2016_May_aleklett_kristin.pdf [ 3.45MB ]
Metadata
JSON: 24-1.0223809.json
JSON-LD: 24-1.0223809-ld.json
RDF/XML (Pretty): 24-1.0223809-rdf.xml
RDF/JSON: 24-1.0223809-rdf.json
Turtle: 24-1.0223809-turtle.txt
N-Triples: 24-1.0223809-rdf-ntriples.txt
Original Record: 24-1.0223809-source.json
Full Text
24-1.0223809-fulltext.txt
Citation
24-1.0223809.ris

Full Text

EXAMINING ECOLOGICAL DETERMINANTS OF COMMUNITY FORMATION AND STABILITY IN THE ROOT MICROBIOTA  by  Kristin Anna Eva Aleklett  B.Sc., Lund University, 2010 M.Sc., Lund University, 2011  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILISOPHY  in  THE COLLEGE OF GRADUATE STUDIES  (Biology)  THE UNIVERSITY OF BRITISH COLUMBIA  (Okanagan)  December 2015   © Kristin Anna Eva Aleklett, 2015  ii Abstract  The root system of a plant is known to host a wide diversity of microbes that can be beneficial or detrimental to the plant. Microbial ecologists have long struggled to understand the factors influencing the composition of these communities. One overlooked aspect of microbial community assembly in root systems is the potential for individual variation among plants, and the potential effect of early colonisation events such as microbial exposure of the seed inside the parent plant and during dispersal.   In this dissertation, I relate ecological theory of community assembly to the formation of the root microbiota. I explore the extent of variation between individuals in wild plant populations, and examine the effects of historical contingency in determining bacterial and fungal community assembly and stability in the root microbiota.   The main findings in my work showed that:  - Wild plants growing in close proximity, sharing environmental conditions will still host distinct bacterial communities in their root systems based on their species identity. We also documented individual variation in root microbiota within all species examined, even the clonal plant species Pilosella aurantiaca. - Bacterial community composition varies significantly across the body of a plant, with different parts of the plant body hosting distinct communities.  - Plants are able to form new microbial associations throughout development, but the timing of microbial exposure affects the composition of the microbial community in mature plants.  - Microbial community stability fluctuates within weeks during early plant development, with one week-old plants hosting communities most likely to change in composition.      iii Preface Chapter 1 (The root microbiota – a fingerprint in the soil?)  is an adaptation of a review that was published in the journal Plant and Soil (Aleklett and Hart 2013). The review was entirely drafted by me, with editorial inputs from Dr. Miranda Hart.  Chapter 2 (Wild plant species growing closely connected in a subalpine meadow host distinct root-associated bacterial communities) was published as an original research paper in PeerJ (Aleklett et al. 2015). The experiment was designed by me in collaboration with Dr. Hart and fieldwork was carried out by me with assistance from Monika Gorzelak. DNA extractions were performed in the Hart lab at UBC Okanagan by myself, and then further processed for sequence amplification and 454-sequencing by Jonathan Leff and Dr. Noah Fierer at the University of Colorado, Boulder. Post-sequencing data processing was done by me in collaboration with the Fierer lab. Statistical analysis as well as drafting of the manuscript for Chapter 2 was done by me. All authors of the manuscript contributed to the editing process of the text.  Chapter 3 (Bacterial communities across the body of a single plant) is based on additional data that was collected and analysed from the field study described in Chapter 2. The manuscript was drafted and edited by myself and Dr. Hart, and figure 3.3 was created by Jonathan Leff. The experiment presented in Chapter 4 (First come, first served – is there a window of opportunity for microbial colonisation in the root microbiota?) is currently being submitted for publication. I designed this experiment in collaboration with Dr. Hart, and conducted the growth-chamber experiments, sampled, processed all samples and performed all molecular analyses. Sequencing was done by the Vancouver Prostate Centre at the MJS lab of UBC. Post-sequencing data processing was mainly done by me, with specific assistance from Dr. Brian J. Pickles in transforming the data through DESeq2 (methods described in Chapter 4). Statistical analysis and manuscript drafting was done by me, with editorial assistance from Drs. Hart and Pickles.  Chapter 5 (Community stability in the root microbiota during early plant development) was designed in collaboration with Dr. Hart. All experiments were conducted, and analysed for microbial community composition by myself. I drafted the manuscript, with editorial assistance from Dr. Hart.  iv Table of Contents Abstract .................................................................................................................................... ii Preface ..................................................................................................................................... iii Table of Contents ................................................................................................................... iv List of Tables ......................................................................................................................... vii List of Figures ...................................................................................................................... viix Acknowledgements ............................................................................................................ xviii Dedication .......................................................................................................................... xviiii Chapter 1: Introduction ......................................................................................................... 1 1.1 Root microbiota and the human gut ............................................................................................. 2 1.2 Community assembly in the root microbiota ............................................................................... 3 1.2.1 Dispersal limitations ......................................................................................................... 5 1.2.2 Environmental constraints ................................................................................................ 6 1.2.2.1 Abiotic environmental constraints ........................................................................... 6 1.2.2.2 Biotic environmental constrains ............................................................................... 8 1.2.3 The realized microbiota .................................................................................................. 10 1.2.3.1 Historical contingency ........................................................................................... 10 1.3 The life cycle of a plant and its effect on the root microbiota ................................................... 11 1.3.1 Inside the flower/fruit of the parent plant (Fig.1.3 – Stage 1) ........................................ 12 1.3.2 Seed dispersal (Fig.1.3 – Stage 2)................................................................................... 15 1.3.3 Germination in the soil (Fig.1.3 – Stage 3)..................................................................... 16 1.3.4 Plant development and maturation (Fig.1.3 – Stage 4) ................................................... 17 1.3.5 Inflorescence and reproduction (Fig.1.3 – Stage5) ......................................................... 17 1.3.6 Senescence (Fig.1.3 – Stage 6) ....................................................................................... 18 1.4 Stability in the root microbiota .................................................................................................. 18 1.5 Function of the root microbiota .................................................................................................. 19 1.6 Summary and research objectives: ............................................................................................. 21 Chapter 2: Wild plant species growing closely connected in a subalpine meadow host distinct root-associated bacterial communities .................................................................. 23 2.1 Background ................................................................................................................................ 23  v 2.2 Methods ...................................................................................................................................... 25 2.2.1 Field site and target plant ................................................................................................ 25 2.2.2 Experimental design ....................................................................................................... 27 2.2.3 Bacterial community analysis ......................................................................................... 28 2.2.3.1 Amplification and sequencing of target gene......................................................... 28 2.2.3.2 Processing raw sequence data ................................................................................ 29 2.2.4 Statistics .......................................................................................................................... 29 2.2.5 Spatial structuring of plant-host-associated bacterial communities in the field ............. 31 2.3 Results ........................................................................................................................................ 31 2.3.2 Variation between host species ....................................................................................... 31 2.3.3 Variation within host species .......................................................................................... 32 2.3.4 Relative abundance of taxa across hosts ......................................................................... 33 2.3.5 Variation due to spatial distance ..................................................................................... 34 2.4 Discussion .................................................................................................................................. 38 2.4.1 Host specificity ............................................................................................................... 38 2.4.2 Individual variation in root microbiota ........................................................................... 39 2.4.3 Bacterial community composition .................................................................................. 40 2.4.4 Variation in relative abundances of bacterial taxa across plant species ......................... 42 2.5 Conclusions ................................................................................................................................ 42 Chapter 3: Bacterial communities across the body of a single plant ............................... 43 3.1 Background ................................................................................................................................ 43 3.2 Methods ...................................................................................................................................... 45 3.3 Results ........................................................................................................................................ 45 3.4 Discussion .................................................................................................................................. 50 Chapter 4: Timing of soil exposure affects community composition in the root microbiota of mature plants. ................................................................................................ 53 4.1 Background ................................................................................................................................ 53 4.2 Materials and Methods ............................................................................................................... 56 4.2.1 Host plant and experimental setup .................................................................................. 56 4.2.2 DNA extraction and amplification .................................................................................. 57 4.2.3 Sequence analysis and statistics ...................................................................................... 58 4.3 Results ........................................................................................................................................ 60 4.3.1 Are plants able to take up new microbial associations throughout their lives? .............. 60  vi 4.3.2 Does plant developmental stage at inoculation determine the composition of the root microbiota? .................................................................................................................................. 62 4.4 Discussion .................................................................................................................................. 68 4.4.1 Order of microbial exposure ........................................................................................... 68 4.4.2 Plant developmental stage at inoculation........................................................................ 72 4.4.3 Relating back to natural systems .................................................................................... 74 4.4.4 Conclusions and future directions .................................................................................. 75 Chapter 5: Community stability in the root microbiota during early plant development................................................................................................................................................. 76 5.1 Background ................................................................................................................................ 76 5.2 Methods ...................................................................................................................................... 80 5.2.1 Host plant and experimental setup .................................................................................. 80 5.2.2 Harvest, DNA extraction, amplification and sequencing ............................................... 80 5.2.3 Bioinformatics and statistical analysis ............................................................................ 82 5.3 Results ........................................................................................................................................ 83 5.3.1 Perturbation effects ......................................................................................................... 83 5.3.2 Differences between plants of different ages .................................................................. 85 5.4 Discussion .................................................................................................................................. 89 5.4.1 Microbial community stability and plant age ................................................................. 89 5.4.2 Differences between responses in bacterial and fungal communities ............................ 91 5.4.3 Application and future directions ................................................................................... 92 Chapter 6: Conclusion .......................................................................................................... 94 6.1 Main findings ............................................................................................................................. 94 6.2 Strengths and limitations of the dissertation work ..................................................................... 96 6.2.1 Experimental design ....................................................................................................... 96 6.2.2 Sequencing ...................................................................................................................... 98 6.3 Future directions ........................................................................................................................ 99 Appendices ........................................................................................................................... 123 Appendix A Supporting material for Chapter 4 ............................................................... 123 Appendix B Supporting material for Chapter 5 ............................................................. 1266   vii List of Tables   Table 1.1 Variance in microbial abundance across environments encountered throughout the life stages of the plant. .............................................................. 14 Table 2.1 PERMANOVA results, comparing bacterial community resemblance between plant species and transects using different diversity metrics. ............. 30 Table 2.2 PermDISP results showing the average spread from centroid and standard error (SE) for samples of each species. The PERMANOVA (P(perm)) values assess whether there is a significant difference between species in sample dispersion, using different diversity metrics. .............................................................................................................. 33 Table 4.1 Comparison of PERMANOVA results generated from a rarefied and log-transformed dataset (bacteria: 2000 sequences/sample, fungi: 1000 sequences/sample) and a non-rarefied dataset that was rlog-transformed. The data compares bacterial and fungal communities from plants harvested before inoculation (Harvest 1), plants inoculated at different ages but exposed to the inocula for the same length of time (Harvest 2), and plants inoculated at different ages but harvested at the same age (Harvest 3). Results show that for bacteria, results remain significant independently of whether the data set was rarefied or rlog transformed. For fungi, the rlog transformed data set no longer showed significant differences at the α=0.05 level between communities from plants exposed to soil at different ages but harvested at the same age (Timing of inoculation, Harvest 3) as well as between plants harvested before and after inoculation (Effect of inoculation, Harvest 1 v.s. Harvest 2). .............................................................. 63 Table 5.1 Table showing an explanation of treatments and results of PERMANOVA analysis on log-transformed Bray-Curtis dissimilarities. Plants were either only inoculated with a resident soil (A1, B1, C1) or perturbed at different ages (A2, B2, C2). In order to  viii determine the effect of the perturbation, pair-wise comparisons were made between perturbed and non-perturbed samples of the same age. ............ 83 Table A.1. Number of samples with successful sequence amplification available for analysis from each treatment and harvest, comparing the rarefied dataset (bacteria: 2000 sequences/sample, fungi: 1000 sequences/sample) with the non-rarefied dataset. ........................................... 123 Table A.2. Survival rates in plants inoculated at different ages out of 10 plants for each treatment that were dedicated for sampling at Harvest 3, when the plants had reached the age of 12 weeks ........................................... 123 Table A.3. Results from running a 2-way PERMANOVA, comparing log transformed Bray Curtis distances between bacterial and fungal communities from Harvest 1 and Harvest 2 based on the factors Harvest and Age at harvest. Because of sample loss due to low amplification, we were not able to assess the interaction between harvest and age at harvest for fungal communities as several age classes were represented by too low number of samples (see table A.2. for numbers) ............................................................................................ 124 Table B.1. Additional information about the origin and properties of the resident and exogenous soils used in the experiment ...................................... 126 Table B.2. Climate data for the two regions where the soils were collected ................... 127 Table B.3. Results from running a 2-way PERMANOVA, comparing log transformed Bray Curtis distances between bacterial and fungal communities. Results show that for bacteria, there was a significant difference between root communities both based on when they were harvested and what soil they were exposed to, but that there was no significant interaction between harvest and exposure. For fungi, results showed that there was a significant effect of harvest time, but not soil exposure .............................................................................................. 127     ix List of Figures  Figure 1.1 Plant roots are analogous to the human gut in that they are the primary absorptive organs of the host, they interact directly with the environment and mediate important immune/hormonal pathways. ...................... 2 Figure 1.2 Factors limiting the species pool of a plant’s realized root microbiota. Out of all microbes in the environment, only a limited amount of them will be able to disperse to the roots. Out of those microbes, only a subset will have the capacity to survive and thrive under the prevailing environmental conditions in the root/rhizosphere. A selection of these microbes, able to disperse and survive the environmental conditions, will ultimately end up in the realized root microbiota of the plant. This selection will be governed by forces such as the historical contingency of the community (redrawn from Belyea and Lancaster 1999) .......................................................... 4 Figure 1.3 Microbial community assembly throughout the life of a plant. The colour of the bands and pie charts show the origin of the microbial species pool available at the various life stages. Bacterial abundance and common taxa found at the various life stages is documented in Table 1. The hypothetical community compositions presented in a, b and c illustrates what the communities would look like if structured mainly by the force of a) dispersal constrains, b) environmental constrains or c) the historical contingency (timing and order of arrival) of the community members. ................................................................... 13 Figure 2.1 This picture shows the layout of the field site at Chute Lake, BC, Canada, inhabited by a mixed plant community of herbs including Orange hawkweed (Pilosella aurantiaca). Oxeye daisy (Leucanthemum vulgare) and Alsike clover (Trifolium hybridum,) which were sampled in this experiment. ............................................................. 26 Figure 2.2 The target plant, Pilosella aurantiaca is known to be a clonal species with very little geratic variation within North America. ........................ 27  x Figure 2.3 Principal coordinates analysis plot illustrating the phylogenetic overlap in root prokaryotic community composition among samples from three different plant species. Phylogenetic overlap between communities was assessed using weighted UniFrac, and PERMANOVA results showed that community composition was significantly different among plant species (p <0.001). ...................................... 32 Figure 2.4 Comparison of the average bacterial community composition and relative abundances, at the phylum level (Proteobacteria divided into class) in root samples from three different plant species. Results show a strong dominance of sequences belonging to Betaproteobacteria in all three plant species, but especially in T. hybridum (51%) and L. vulgare (50%). The phyla representing less than 1% out of the total community were grouped as “Other” and consisted of: NKB19, Nitrospirae, PAUC34f, Cyanobacteria, Elusimicrobia, Fibrobacteres, Chlamydiae, SC4, Spirochaetes and Thermi. Sequences not matching the database were recorded as “No blast hit.” ............................................................................................................. 35 Figure 2.5 The average relative abundance of sequences belonging to the Betaproteobacteria families and Oxalobacteraceae genera found in root samples of the three plant species. The table shows a more detailed overview of which families within the Betaproteobacteria class that sequences were most commonly assigned to. At the bottom of the table, the total percent of sequences belonging to Betaproteobacteria across all samples of each plant species is listed, and above that, how these percentages are divided between different families. Values are given as the percentage of sequences belonging to a certain taxa out of the total average bacterial community for each of the three plant species (rarefied at 400 sequences/sample). The heat map is colour coded from blue (low abundance) to red (high abundance) to facilitate the overview. To the right, we have also included an even more detailed overview of how the total  xi percentages of sequences belonging to the family Oxalobacteraceae from each plant species were divided at a genus-level. By doing that, we could see that these sequences were mainly classified as Herbaspirillum. .................................................................................................... 37 Figure 3.1 Image displaying the complete plant specimen of the species Pilosella aurantiaca  which was sampled in 93 pieces  in order to examine bacterial community composition across the body. .............................. 44 Figure 3.2 Comparison of the average OTU richness in samples collected from different parts of the plant body. Results showed that there was a significant difference in OUT richness between body parts (p=0.04, F=2.56) and Duncan analysis indicated that samples from the runner hosted significantly fewer OTUs than the inflorescence and root samples (results from the Duncan test are listed as letters in the graph, where different letters indicate significantly different results at a 0.95 confidence level). .................................................................................. 46 Figure 3.3 Variations in OTU richness across the body of the P. aurantiaca plant. Samples collected are indicated by dividing lines, and 300 sequences were sub-sampled from each sample in order to be able to compare as many samples as possible while controlling for sampling effort. Samples that failed to amplify bacterial sequences are coloured white, and the rest range on a colour scale from blue (~ 50 OTUs) to red (~ 240 OTUs) showing how OTU richness varies across the plant body. .......................................................................................... 47 Figure 3.4 Principal coordinates analysis of un-weighted (a) and weighted (b) UniFrac dissimilarities between samples from different parts of the plant body. Each dot represents a sample, and samples are labelled based on where on the plant body they were collected (inflorescence, leaves, runners, bulb, roots or stem of the plant). Samples far apart from each other are considered to host bacterial communities less related to each other, and results showed that each compartment of the plant hosted distinct bacterial communities more similar to each  xii other than any other samples from the plant body. These trends were also confirmed in PERMANOVA results comparing the communities through both un-weighted (Pseudo-F=1.5645, p=0.0001) and weighted (Pseudo-F=3.3629, p=0.0001) UniFrac distances. ............................................................................................................. 48 Figure 3.5 Comparison of the average proportion of sequences belonging to different bacterial orders, among samples taken from different parts of the plant body. Sequences belonging to orders that made up less than 1% of the total community have been grouped as “Other” in order to get a better overview of general patterns in community composition. The graph illustrates the differences in bacterial community composition found within a single plant body, a trend that was confirmed in the PERMANOVA results showing a significant difference between body parts (p=0.0001). ....................................... 49 Figure 4.1 This image shows the seed head of the species Setaria viridis which was used as a host plant in the experiment. The plants used in the experiment were all of the same genotype, minimizing genetic variation between individual plants. .................................................................... 55 Figure 4.2 Schematic overview of the experimental design. At each harvest, root samples from five plants from each age treatment were collected. Harvest 1 was collected to control for community composition prior to inoculation and included plants of different ages (9, 8, 7, 1 and 0 weeks old) representing different developmental stages (flowering, budding, non-reproducing mature plant, seedling, seed) in the life cycle of S. viridis. Harvest 2 was collected two weeks after inoculation to control for length of exposure to the inoculum, and Harvest 3 was collected to control for age of plant. Harvest 3 occurred when plants had all reached maturity (12 weeks), but before senescence. The age of the plants at the time of harvest is listed in the diagram. ........................................................................................... 55  xiii Figure 4.3 α-diversity measures of bacterial (a) and fungal (b) communities in samples harvested prior to (Harvest 1), or two weeks after (Harvest 2), inoculation with soil. Each dot represents a sample, and the variation among samples was calculated using observed species richness. Results from t-tests comparing bacterial and fungal richness pre and post inoculationshow that there was indeed a significant increase in bacterial (p=0.0001, t=-4.29) but not fungal (p=0.26 t=1.15) richness post inoculation. .......................................................... 64 Figure 4.4 Principal co-ordinates analysis (PCoA) plots of log-transformed Bray Curtis dissimilarities, showing compositional dissimilarity between samples, where two points closer together host more similar communities. These plots show bacterial (a) and fungal (b) communities in samples from plants harvested immediately prior to (Harvest 1), and two weeks after (Harvest 2), inoculation and indicate a shift in bacterial community composition where we see little overlap between samples harvested before and after inoclation. For bacterial communities, PERMANOVA results confirmed that there was a significant distinction between communities in plant roots harvested before and after inoculation (p=0.0001, Pseudo-F=2.89). Though less apparent in the PCoA, PERMANOVA results also revealed that there was a significant difference between fungal communities in plant roots harvested before and after inoculation (p=0.03, Pseudo-F=1.57). .................................................................................... 65 Figure 4.5 Bar graphs showing the average relative abundance of sequences belonging to bacterial (a) and fungal (b) orders, compared between samples harvested before inoculation (Harvest 1), after inoculation (Harvest 2), and at 12 weeks (Harvest 3).  For Harvest 1 and Harvest 2, the bars show an average community composition based on a combination of all samples from that harvest. Among samples from Harvest 3, we separated out and compared the average community composition of plants inoculated at 0 versus 9 weeks. Orders  xiv representing less than 1% of the community have been grouped as “Other”. For bacterial communities we see that Actinomycetales made up a larger portion of the average community prior to soil inoculation whereas Burkholderiales became more dominant in the root microbiota after the introduction of soil. For fungal communities, we see less prominent changes in the community after soil inoculation, but that plants inoculated from seeds form very distinct fungal communities, heavily dominated by the order Xylariales............................................................................................................. 66 Figure 4.6 α-diversity measures of bacterial (a) and fungal (b) communities in samples from plants that were exposed to the soil inoculum at different developmental stages (Age_at_inoculation) but harvested when at 12 weeks (Harvest 3). Each dot represents a sample, and the variation among samples is calculated using observed species richness. ANOVA results comparing the different treatments (excluding treatments with less than 3 samples)confirmed that there was no significant effect of timing of inoculation on bacterial (p= 0.59 F=0.66) or fungal (p=0.26, F=1.55) community richness. .......................... 70 Figure 4.7 Principal co-ordinates analysis (PCoA) plots of  Bray Curtis dissimilarities between (a) bacterial and (b) fungal communities in root samples harvested from plants at 12 weeks old. The plots show a trend of plants inoculated at the same age hosting more similar communities, with plants inoculated as seeds standing out as hosting the most different bacterial and fungal communities compared to plants inoculated at other ages. PERMANOVA results confirmed that plants inoculated at different developmental stages hosted distinct bacterial (p=0.008 Pseudo-F=1.45) and fungal communities (p=0.03, Pseudo-F= 1.38). ................................................................................... 71 Figure 5.1 Depiction of potential community development (with dots of different colours representing different microbial taxa) based on whether the initial community is resistant or sensitive to perturbation  xv (indicated by the red box and the introduction of a new microbial community). In a resistant community, the progression of community development would be expected to continue in the same direction disregarding of whether the plants were exposed to a resident or exogenous soil  (A and B would host similar communities). In a sensitive community, the introduction of an exogenous soil would be expected to cause a change in community composition (C and D would host significantly different communities). ...................................................................................................... 78 Figure 5.2 Comparison of bacterial (a) and fungal (b) richness (Observed species richness) between treatments. Each dot represents a sample, and for each age class (A, B, C) there are plants that were perturbed either with the resident soil (A1, B1, C1) or the exogenous soil (A2, B2, C2). This allows us to compare between plants that got perturbed as seeds (A), one-week-old seedlings (B) or 2-weeks-old seedlings (C). The plots show an overall decrease in bacterial richness over the course of the experiment, but no significant difference in richness between plants of the same age but with different soil exposure. For fungal communities, results show no significant difference between plants with different soil exposure. These observations were in accordance with Duncan test results. .................................................................. 86 Figure 5.3 Principal Coordinates Analysis (PCoA) plots showing differences in bacterial (a) and fungal (b) community composition between treatments based on Bray Curtis dissimilarities. Each dot represents a sample, and for each age class (seeds (A), one-week-old seedlings (B) or 2-weeks-old seedlings (C)) there are plants that were exposed to either exclusively the resident soil (A1, B1, B2) or a combination of the resident and the exogenous soil (A2, B2, C2).  For bacteria, the plot shows a separation between communities in plants harvested at different ages, but no clear distinction between plants with different soil exposure harvested at the same age. For fungi, there is  xvi no clear separation between samples with different soil exposure of any age. These trends were confirmed in PERMANOVA comparisons of the Bray Curtis dissimilarities (Table). ...................................... 87 Figure 5.4 Comparison of bacterial (a) and fungal (b) richness (Observed species richness) between harvests. Each dot represents a sample, and each harvest is a combination of 14 plants of the same age (3-weeks-old (Harest1), 4-weeks-old (Harvest 2), 5-weeks-old (Harvest 3), that were either exposed toexclusively the resident soil (7 plants) or a combination of the resident and  exogenous soil (7 plants). For bacteria we can see a clear decrease in richness between Harvest 1 and 2. This change was confirmed when results were compared through a Duncan test which indicated that there was a significant difference between samples from Harvest 1 and samples from Harvest 2 and 3. For fungi there was no significant change in richness between harvests. .................................................................................. 88 Figure A.1. A comparison of the average community composition in bacterial and fungal communities of plants harvested at the same age prior to or two weeks after soil inoculation.................................................................... 125 Figure B.1. Comparison of the average proportion of bacterial orders found in plants from different treatments. Orders that made up less than 1% of the total community were grouped as “Other”. The plants compared were perturbed either with a resident- (A1, B1, C1) or an exogenous soil (A2, B2, C2) as seeds (A), one-week-old seedlings (B) or 2-week-old seedlings (C) ...................................................................................... 128 Figure B.2. Comparison of the average proportion of fungal orders found in plants from different treatments. Orders that made up less than 1% of the total community were grouped as “Other”. The plants compared were perturbed either with a resident- (A1, B1, C1) or an exogenous soil (A2, B2, C2) as seeds (A), one-week-old seedlings (B) or 2-week-old seedlings (C). ..................................................................................... 129   xvii Acknowledgements  To my supervisor Dr Miranda Hart, for taking a chance on me, letting me be such a big part of developing this project, and pushing me to be a better scientist.   To my committee members, Drs. Melanie Jones and Deanna Gibson for their advice and support throughout my degree.  To Noah Fierer, Jonathan Leff and Brian Pickles for contributing to my thesis work and providing good advice and editorial assistance.    To my family, who inspired me to take on science, explored nature with me and always encouraged me, believed in me, and made me understand myself better.   To my love, my rock, my biggest supporter and closest friend, David Kadish, who has helped me celebrate the good times and fight through the hard times, always with a smile and a pun, some tech advise and a big hug. Without you, I don’t know if I could have made it.  To my fellow academics and close friends, Jen Forsythe, Sepideh Pakpour, Monika Gorzelak and Christina Turi who have been there along my side throughout this process, providing good advice, a helping hand, a listening ear or a kick in the butt when I’ve needed it.  To the rest of the Hart-Klironomos lab: Thanks for everything. You will all go on to do amazing things. I hope our roads will cross again.     xviii Dedication          “The world shimmers, a pointillist landscape made of tiny living beings.”      - Lynn Margulis   1  Chapter 1: Introduction  Plants live in close association with microbial communities. The root system in particular hosts a diverse community of microbes. This assemblage of fungal and bacterial taxa is commonly referred to as the root microbiota (referring to the collection of microbial taxa) or microbiome (referring to the genomes of the microbes in that collection).  The composition of the root microbiota can affect important plant traits such as stress tolerance (Latch 1993; Rodriguez and Redman 2008), productivity (Schnitzer et al., 2011; van der Heijden, Bardgett, & Van Straalen, 2008; van der Heijden et al., 2006) and fitness (Smith 2001; Lau and Lennon 2011).  Despite such important effects on plants, it is still not fully understood how root microbiota communities are assembled. There is evidence that the composition of the root microbiota is governed by both abiotic (non-living chemical and physical factors in the soil) and biotic (living components of the ecosystem) factors. It has been suggested that host plants play an important role in shaping their own microbiota by directly mediating environmental factors such as pH (Hartmann et al. 2008) and soil nutrient content (Jones et al. 2004), and root architecture (Hodge et al. 2009; Marschner et al. 2011). To this end, variation in root microbiota has been shown to occur at the level of the plant family (Wang and Qiu 2006), species (Wieland et al. 2001; Pivato et al. 2007) and even genotypes (Micallef et al. 2009; Lundberg et al. 2012), suggesting some degree of plant control over microbial community assembly.  In this introduction I explore the idea that historical events in the plant life also play an important role in shaping the composition of the root microbiota through their effects on dispersal limitations, environmental constraints and the historical contingency of the root microbiota. I consider how plant phenotype has the potential to create specific root microbial communities and aim to address unanswered questions in the field: Are there key events in the plant life cycle that shape its microbiota? How stable is it over time? And like human gut microbiota, do plants have significant variation at the level of the individual? I review the literature on root microbiota while presenting the concept of historical contingency in order to contribute to the dialogue on this emerging topic. 2  1.1 Root microbiota and the human gut Recently, there has been considerable progress in describing microbial communities associated with humans, and results show variation among individuals at a host phenotype level, where host properties such as age, body mass and ethnicity were correlated with the composition of the microbiota (Huttenhower et al. 2012). Similarities between the role of the root system in plants and the digestive system in humans allow for comparisons between microbial community assembly in the these two systems (Fig.1.1) (Ramírez-Puebla et al. 2013).  Plant roots are analogous to the human gut in that they are the primary absorptive organs of the host, they interact directly with the environment and mediate important immune/hormonal pathways (Bais et al. 2006; Berendsen et al. 2012). Like the animal gut, roots are also closely related to individual health and survival of the host (Raaijmakers et al. 2009; Berg and Smalla 2009; Sekirov et al. 2010; Huttenhower et al. 2012) (Further discussion on how the microbiota affects plant health, can be found in the recent virtual issue no.2, 2012 in Plant and Soil).   Given this analogy, do plants exhibit similar levels of individual variation in microbiota as has been shown in the human gut (Spor et al. 2011; Gonzalez et al. 2011)? Humans travel   Figure 1.1 Plant roots are analogous to the human gut in that they are the primary absorptive organs of the host, they interact directly with the environment and mediate important immune/hormonal pathways. 3  and intermingle with each other in a way that could easily homogenize the microbiota within and between populations, yet, the human microbiota allows for distinction between individuals as closely related as identical twins (Turnbaugh et al. 2009). One could argue that because plants are not as mobile as humans, they should exhibit even more pronounced levels of individual variation, as a consequence of their sessile growth habit. At the same time, dispersal limitations among microbes might also result in plants growing close to each other sharing more microbial taxa. Nevertheless, hyperdiversity in soil microbial communities may present plants with sufficient variation in the species pool of root microbiota to create individual patterns, even at a local scale. Recent research on humans suggests that the community composition of gut microbiota shows significant variation as a consequence of changes in environment, diet, interactions and development of the host (Benson et al. 2010; Spor et al. 2011; Agans et al. 2011). This indicates that the community is dynamic in time and results from interactions among genotype, lifestyle and experiences (Gonzalez et al. 2011). The term ‘phenotype’ refers to the genotype combined with environmental effects and, in terms of plants, is usually associated with traits such as height and biomass that can vary between genetically identical individuals due to environmental conditions. In the same sense, the microbiota of an individual plant or animal, is an extension of the host phenotype, since its community composition will be influenced by phenotypic traits in the host. I argue that plant phenotype can exert a strong effect on microbial community assembly and that the factors determining plant phenotype (genetic and environmental) therefore also should be considered as structural forces in assembly of the root microbiota  1.2 Community assembly in the root microbiota Which microbes end up in the root microbiota (and in what abundance) depends on multiple ecological ‘filters’ as well as chance events. These ecological ‘filters’ are often discussed as niche-based or neutral processes. Niche-based theories predict that coexistence between species requires niche differentiation (Chesson 2000; Leibold and McPeek 2006) and that shifts in community composition will be related to environmental changes.  Neutral theories suggest that variation in community composition is independent of adaptation to the 4  environment and results from distance between the communities (dispersal limitations), speciation/extinction and stochastic events (Hubbell 2001). It has been suggested that for microbial communities, root-associated communities of arbuscular mycorrhizal (AM) fungi for example (Lekberg et al. 2007; Dumbrell et al. 2010), these two theories might not be exclusive,.  For microbes, niche-based processes can be split into abiotic and biotic environmental constraints whereas neutral constraints are most commonly thought of as dispersal limitations (determining the regional species pool of available colonisers).  Among those microbes able to overcome both dispersal limitations and environmental constraints in order to colonise the root (Fig. 1.2), other processes (i.e. historical contingency of the community, random events during colonisation, and speciation among the microbes) may determine the assembly of the realized root microbiota (illustrated in Fig.1.3).   Figure 1.2 Factors limiting the species pool of a plant’s realized root microbiota. Out of all microbes in the environment, only a limited amount of them will be able to disperse to the roots. Out of those microbes, only a subset will have the capacity to survive and thrive under the prevailing environmental conditions in the root/rhizosphere. A selection of these microbes, able to disperse and survive the environmental conditions, will ultimately end up in the realized root microbiota of the plant. This selection will be governed by forces such as the historical contingency of the community (redrawn from Belyea and Lancaster 1999) 5  In this introduction, I will focus on how dispersal limitations, environmental constraints and historical contingency throughout the life of the plant may affect the assembly of the root microbiota (Fig.1.2).  1.2.1 Dispersal limitations A general concept in dispersal theory is that organisms with a large number of individuals per unit area also should disperse widely and be able to colonize remote locations by chance (Finlay 2002; Martiny et al. 2006). Thus microbes should experience few constraints to dispersal. Yet, current research continues to reveal biogeographic patterns and endemic populations among microbes, indicating that microbial dispersal might not be limitless (Fulthorpe 1998; Whitaker et al. 2003; Norros et al. 2012). Whether these biogeographical patterns are a consequence of dispersal limitations or environmental selection, however, is far from resolved (Horner-Devine et al. 2004; Martiny et al. 2006; Ramette and Tiedje 2007; Eisenlord et al. 2012). Describing dispersal in microbial communities is not an easy task since microbial movements are difficult to track in the environment. Microbes are known to disperse through passive transportation in the atmosphere (Smith et al. 2012), water currents, or by movements of plant and animal vectors, where dispersal distances might be affected by which type of dispersal the various microbial taxa are able to withstand (Fierer 2008).  Since microbes span multiple kingdoms and have widely divergent life history strategies, determining a common model for dispersal is too simplistic. In general, there is little empirical evidence to inform our knowledge about microbial dispersal.  It has been found that active dispersers travel further than passive dispersers, and that the size of the organism can affect dispersal distance among active dispersers but not passive dispersers (Jenkins et al. 2007). Some microbial taxa might be limited to one or a few specific modes of dispersal while others exploit a wide range of dispersal modes. It is therefore possible that microbes with few modes of dispersal could become more concentrated in certain regions and environments while microbes able to disperse in multiple substrates and through various vectors would have a more global distribution.  Even among a relatively conserved microbial lineage in terms of growth forms (i.e. fungi), there is no one strategy of dispersal. For example, many ectomycorrhizal fungi 6  disperse by producing fungal fruiting bodies that release spores to the wind, while AM fungi often remain underground for the entirety of their life cycle and, therefore, are dependent on hyphal growth to new areas, dispersal through a vector, movement of soil or for air currents to pick up spores from the surface of the soil in order to spread to a new area (McIlveen and Cole Jr. 1976; Warner et al. 1987; Friese and Allen 1991). In terms of the root microbiota, dispersal constraints may simply limit the pool of microbes that are able to come in contact with the root area. In our model, a hypothetical root microbiota determined solely by dispersal constraints should be expected to host all microbes that encounter the root system (Fig. 1.3 a).   1.2.2 Environmental constraints We are just beginning to understand how microbial communities vary across habitat-types and locations over large and small distances (Green and Bohannan 2006; Fierer 2008; Gilbert and Meyer 2012; Hazard et al. 2012). Studies mapping the geographical distribution of bacteria in soil have found that soils from different parts of the world can harbour distinct communities, but that soils from similar biomes do not always contain similar bacterial communities (Fierer and Jackson 2006; Lauber et al. 2009). These biogeographical patterns have largely been explained by abiotic conditions such as soil type (Singh et al. 2007; Oehl et al. 2010; Bulgarelli et al. 2012; Lundberg et al. 2012) and pH (Lauber et al. 2009; Rousk et al. 2010). For the root microbiota, influential abiotic factors can also include soil nutrient content and seasonality. 1.2.2.1  Abiotic environmental constraints 1.2.2.1.1 Soil properties Soil pH is one of the most widely argued determinants of microbial communities in the soil. Studies have shown that bacterial-fungal ratios and community compositions vary across a pH gradient (Bååth and Anderson 2003; Rousk et al. 2010), that soils of different pH host distinct microbial communities (Lauber et al. 2009; Rousk et al. 2010) and that changes in pH can lower the microbial tolerance to compounds such as cadmium in the soil (Babich and Stotzky 1977). 7  Another major environmental factor influencing microbial community structure is soil nutrient content. Microbes actively seek out nutrient sources in soil to fulfil nutritional needs (Hodge et al. 2010). Among mycorrhizal fungi it is also common that the resources sought out in the soil are traded for carbohydrates with surrounding plants (Smith and Read 2008). Consequently, fluctuations in soil nutrient content could affect not only the ability of the microbes to satisfy their own needs, but also their relations to their host plants (Johnson 2010). Increased nutrient levels in the soil do not always induce increased microbial diversity or productivity. For example, nitrogen fertilization reduced microbial biomass of both fungi and bacteria in a grassland experiment (Rousk et al. 2011) where soil cores from fields with different levels of nitrogen fertilization were compared. Theoretically, the plants growing in the field should be able to provide resources for more microbes with an increased carbon production under conditions of high nitrogen content in the soil, but these results suggest that the relationship might be more complicated.  Phosphorus is an important soil nutrient readily acquired by AM fungi and used in nutrient exchange with plants (Smith and Read 2008). Increased phosphorus levels in soils could potentially reduce a plant’s dependency on this symbiosis, thereby impacting the community structure of these organisms in the soil community (Johnson 2010). Phosphorus fertilization has been shown to affect community composition of bacteria and AM-fungi (Toljander et al. 2008) and AM-fungal taxa respond differently to increased levels of phosphorus in the soil, where some respond more strongly to shifts in soil nutrient content than others (Treseder and Allen 2002). 1.2.2.1.2 Seasonality Shifts in microbial community structure have been documented throughout the changing of the seasons. This phenomenon could be a consequence of fluctuations in the amount of nutrients readily available to the microbes throughout the year.  For example, bacterial and fungal communities from alpine tundra soils have been found to correlate with snow cover dynamics (Zinger et al. 2009), and marine microbial community dynamics have been shown to correlate with seasonal fluctuations in daylight (Gilbert et al. 2012). For microbial communities associated with plants (i.e. root microbiota), seasonality will have a compound effect displayed both through changes in temperature and moisture of the surrounding soil 8  and seasonal changes in the physiology of the host plant that when winter comes will either experience senescence or relocate its resources to the root system to await the next growing season. For example, seasonal effects have been documented in microbial communities of AM fungi in the rhizosphere (Merryweather and Fitter 1998; Dumbrell et al. 2011). 1.2.2.2  Biotic environmental constrains 1.2.2.2.1 Host effects Plants influence the soil in their immediate surroundings in ways that can affect the assembly of the root microbiota (for an extensive review on this see Hartmann et al. 2008) and are also thought to interact directly with the microbes. The effects of the host plant can therefore be considered as both biotic (direct) and abiotic (indirect).  The host effect is perhaps the single most important ‘filter’ for community assembly in the root microbiota, since it affects the root environment in so many ways. For example, root growth creates plant-specific root architectures for microbes to colonise while also loosening soil particles and attracting surrounding soil fauna (Hodge et al. 2009). Other environmental conditions such as soil nutrient availability, pH and aggregation, can be mediated by plants (Hartmann et al. 2008) and as mentioned earlier, even seasonality is influenced by plant life strategies and seasonal patterns. One of the main pathways in which plants interact with microbes is through root exudates (Bais et al. 2006; Broeckling et al. 2008; Doornbos et al. 2012). Variation in root exudation (timing, amount and/or constituents) provides a mechanism by which plants can manipulate composition and microbial abundances of their root microbiota (Bakker et al. 2012). Exudates are thought to consist mainly of sugars, amino acids and organic acids that are present at high concentrations in the cytoplasm of the plant, but also include smaller amounts of complex secondary metabolites such as flavonoids that can attract specific microbes in the rhizosphere (Jones et al. 2004; Bais et al. 2006).  They can alter soil conditions such as the pH (Jones et al. 2004) which, as mentioned earlier, is thought to be a major abiotic determinant of bacterial community composition (Lauber et al. 2009). It has also been suggested that exudations in the rhizosphere of the signalling molecules jasmonic acid and salicylic acid can be involved in the interplay between roots and microbes at the initial events of colonisation (Gutjahr and Paszkowski 2009; Doornbos et al. 2011). 9  Production of root exudates can be highly variable among individual plants, depending on plant developmental stage, its growth conditions and biotic interactions (Mougel et al. 2006; Houlden et al. 2008; Badri and Vivanco 2009; Micallef et al. 2009). Root exudation is known to vary between plant species (Gransee and Wittenmayer 2000; Broeckling et al. 2008), be genetically regulated (Broeckling et al. 2008; Badri and Vivanco 2009) and even be able to shape distinct rhizo-bacterial communities for different genotypes (Micallef et al. 2009).  Small genetic differences in the plant can result in alterations to root exudation amounts, patterns and constituents (Broeckling et al. 2008; Schweitzer et al. 2008; Micallef et al. 2009) potentially resulting in differences among root microbiota.  In addition, plant genotypes create differences in root and shoot morphology and architecture (Hodge et al. 2009; Badri and Vivanco 2009). Any of these differences could affect which microbial taxa are able to colonise the roots. The concept of community ecosystem phenotypes (Whitham et al. 2006) suggests that genes in one species (i.e. a host plant) can determine the composition of species in an entire ecosystem (Bailey et al. 2005; Shuster et al. 2006; Schweitzer et al. 2008). For example, microbial communities associated with Populus were linked to variation in tannin production among Populus genotypes, which created a cascade of downstream ecosystem level effects (Bailey et al. 2005). In the case of root microbiota, unique genetic traits in a host plant could ‘regulate’ the species composition of the root microbiota and determine its expressed ecosystem phenotype.  1.2.2.2.2 Competition and coexistence Besides a constant interaction with the host plant, microbes of the root microbiota actively interact with the surrounding soil fauna and each other, competing for space and resources. Microbial communities therefore experience dynamics of competition and succession, even though their interactions might be harder to detect and interpret (Fierer et al. 2010). Most studies trying to disentangle interactions between microbes analyze their communities through mapping which species are found under specific conditions or in co-occurrence with other species. The balance between bacterial and fungal dominance in soils has been investigated on several occasions (De Boer et al. 2005; Rousk et al. 2008; Rousk et al. 2010) where a decrease in bacterial growth has shown a corresponding increase in fungal growth 10  (Rousk et al. 2008). In contrast to that, it has also been shown that there exist specialized relationships between the two kingdoms. For example, so called ‘mycorrhiza helper bacteria’ are thought to facilitate AM fungal infection in plant roots (Pivato et al. 2009). It has even been suggested that the phylogeny of AM fungal isolates can act as a predictor of bacterial community composition (Rillig et al. 2006). As a whole, environmental conditions (whether mediated by the plant or not) exert strong selective power on root associated microbes on a small spatial scale. In our model, a microbial community determined by environmental constraints should consist of a subset of the microbes able to disperse to the community – the ones best adapted to survive under the local environmental conditions (Fig.1.3 b).  1.2.3 The realized microbiota While dispersal and environmental constraints limit which microbes are able to form the root microbiota, they do not determine the realized root microbiota. Not all microbes able to disperse to the roots will do so and not all microbes able to survive in the root environment will colonise it. The realized microbiota will therefore consist of a subset of microbes able to both disperse and survive in the root (Fig.1.2). The composition of this realized microbiota will be determined through the processes of colonisation and succession within the community (Fig.1.3) and shaped throughout the plant’s life. 1.2.3.1  Historical contingency  The process of colonisation (timing and order in which species colonise a new habitat) is commonly referred to as the historical contingency of the community (Fukami and Nakajima 2011) and is believed to affect community assembly through priority effects. Priority effects describe a situation whereby the first taxon to arrive in a new habitat gains an advantage in colonisation (Fukami et al. 2005). This pattern has been shown in root colonisation by ectomycorrhizal fungi, where Rhizopogon occidentalis gained an advantage over three other Rhizopogon species by being first to colonise the roots (Kennedy and Bruns 2005; Kennedy et al. 2009). In our case, the historical contingency of the root microbiota may refer to the timing and the order of exposure of different microbial taxa to the root system as well as succession of taxa within the community. 11  The idea of historical contingency in reference to plant microbes is not entirely new; it has been known for some time that plants use root exudates to communicate with root symbionts, and changes in the composition and pattern of exudation provide a direct mechanism by which plants can attract microbes in a desired order. In addition to these chemical signals between plant hosts and microbes, certain events over the course of a plant’s life can also affect the timing of plant root exposure to potential root microbes and their ability to successfully colonise. Further, microbial communities can affect the historical contingency of plant succession. In a well-known example, microbial pathogens increased density dependent selection for diverse communities due to high loads near conspecifics (i.e. Janzen-Connell hypothesis, shown in the field by Packer & Clay (2000)). The establishment of early succession plants has also been shown to induce changes in the soil microbial community, causing a plant-soil feedback, affecting the historical contingency of the plant community assembly (Kardol et al. 2007).  The importance of historical contingency in wood decomposing microbial communities has been shown by Fukami et al. (2010) and historical contingency has also been suggested as a shaping factor of the microbiota in human hosts (Costello et al. 2012). Considering these emerging patterns of community assembly in other systems, it is intriguing to consider what could affect the historical contingency of the root microbiota in plant hosts. In our model, a root microbiota influenced by historical contingency should represent taxa encountered as a result of dispersal that were able to manage the environmental constraints, but will also show a partitioning in the community that reflect differences among taxa in their timing and order of arrival to the root (Fig.1.3 c).  1.3 The life cycle of a plant and its effect on the root microbiota The combined effects of dispersal and environmental constraints of microbial community assembly are not static over the lifetime of a plant. Some life events of the plant may present it with new species pools of microbes (by facilitating microbial dispersal) while others can affect the community dynamics in the root microbiota through changing exudation patterns (i.e. during plant reproduction) and providing new niche space for microbes to colonise (i.e. 12  during root growth and development). Thus, it is important to look at how these factors (dispersal and environmental constraints) change over the life span of an individual plant.  The life of a plant can be split into six major stages that ultimately will affect the community composition of its root microbiota: 1) the seed (inside the flower or fruit of the parent plant), 2) seed dispersal 3) seed germination, 4) plant development, 5) plant reproduction and finally, 6) plant senescence (Fig.1.3). Out of these six major life events, the first three represent the mobile phase of a plant’s life and should present the plant with the bulk of the species pool that make up its microbiota (Fig.1.3). Stages 4, 5 and 6 might still experience new species introduction (i.e. through disturbances or root growth into previously unexplored soil) but will mainly impose environmental constraints on microbial community dynamics. Disturbances in the environment such as water and salt stress, herbivory and nutrient limitations will also contribute to whether the host plant will associate with a specific microbe or not. Estimates of actual microbial abundance/diversity for these different events are given in Table 1.1 Microbial dispersal and environmental limitations will be an overarching factor that carries through all the different life events. In the section below, I describe a potential role for these key events in a plant life cycle that may create variation in their root microbiota.    1.3.1  Inside the flower/fruit of the parent plant (Fig.1.3 – Stage 1) A plant’s first exposure to microbes occurs while it is still a seed within a parental inflorescence. Seeds are exposed to vertical transmission of microbes inside the flower that colonise the seed coat (van Overbeek et al. 2011). The composition of microbial communities in flowers have shown spatial and temporal patterns (Belisle et al. 2012; Shade et al. 2013) and are affected by interaction with insects and other animals that act not only as pollinators but also as hosts and dispersers of microbes to the parental inflorescence, and ultimately, the seed coat of their offspring (Fürnkranz et al. 2012). Whether this manifests differences in the root microbiota remains to be determined, but there are some indications that seed coat microbiota acquired in the parent plant gets carried on to the soil where it germinates. For example, van Overbeek et al. (2011) found that seeds acquired from seed banks in the soil carried a bacterial microbiota more similar to that of seeds still in the flower head than those of the surrounding soil communities.  13    Figure 1.3 Microbial community assembly throughout the life of a plant. The colour of the bands and pie charts show the origin of the microbial species pool available at the various life stages. Bacterial abundance and common taxa found at the various life stages is documented in Table 1. The hypothetical community compositions presented in a, b and c illustrates what the communities would look like if structured mainly by the force of a) dispersal constrains, b) environmental constrains or c) the historical contingency (timing and order of arrival) of the community members. 14  Table 1.1 Variance in microbial abundance across environments encountered throughout the life stages of the plant. Environment Bacterial abundance and diversity (if available) Commonly found  bacterial taxa  Anthosphere (flower) 105 CFU/blossom in stigma a  102-3 CFU/hypanthium a  ~107 CFU/g b 1,677 OTUs/50,865 tag-sequences c  Deinococcus-Thermus, TM7, Bacteroidetes, Firmicutes, and Proteobacteria c Gammaproteobacteria on pollen and in pistils b Carposphere (fruit) 102-3CFU/g d ~103CFU/g b Firmicutes d Atmosphere (air) 104-6cells/m3 e  ~300 phylotypes/600 sequences e Pseudomonadales, Burkholderiales, Rhizobiales, Sphingomonadales (all proteobacteria) e Soil (bulk soil) 4.2 × 106 CFU/g soil f  49,944 phylotypes/132,090 sequences g Acidobacteria, Alphaproteobacteria, Actinobacteria, Bacteroidetes, and Beta/Gammaproteobacteria g Spermosphere  (surrounding germinating seed) < 102 CFU/g b  Pseudomonas (Gammaproteobacteria) and Burkholderia (Betaproteobacteria)h Rhizosphere  (soil surrounding roots) 107-109CFU /g d  1000-1500 OTUs/20,000 sequences i Acidobacteria, Proteobacteria i Proteobacteria and Actinobacteria j Rhizoplane (surface of plant roots) 105-107 CFU/g d Betaproteobacteria, Bacteroidetes and Actinobacteria i a Stockwell et al. 1999, b Fürnkranz et al. 2012, c Shade, McManus and Handelsman 2013, d Compant, Clément and Sessitsch 2010, e Bowers et al. 2011, f Torsvik and Øvreås 2002, g Lauber et al. 2009,     h Liu et al. 2012, i Bulgarelli et al. 2012, j Reviewed by Buée et al. 2009  15  For seeds of fruit-bearing plants, the carposphere (fruit) can also provide an important source of microbial inoculation for the seed (Fürnkranz et al. 2012). Studies of microbial communities in the fruits of Styrian oil pumpkin (Cucurbita pepo L. subsp. pepo var. styriaca Greb.) (Fürnkranz et al. 2012) and grapevine (Vitis vinifera L.) (Compant et al. 2011) have shown that seeds are exposed to a specific set of microbes inside the fruits that are thought to be derived either from the anthosphere (flower) (Fürnkranz et al. 2012) or endosymbionts transported through the xylem to the reproductive organs from the root system (Compant et al. 2011).  1.3.2 Seed dispersal (Fig.1.3 – Stage 2) Once the seed leaves its parent plant, whether it gets eaten or dispersed, the seed now enters its most mobile phase. There are multitudes of ways that a seed could be dispersed and consequently a large variety of potential microbial inocula it may encounter on its journey to germination. Seeds can be dispersed over long or short distances, carried by water, wind or animal vectors. The specific conditions a seed experiences during dispersal will expose it to a unique cohort of microbes. This exposure provides the seed with a community of microbes before it reaches its germination site.  This early exposure may be of importance to the ultimate microbiota of the plant through affecting the historical contingency of the community.  For mycorrhizal fungi, seed-dispersing animals have been shown to aid in plant colonisation, not only by depositing seeds in spaces where spores are present, but also through dispersing the spores of the mycorrhizal fungi (Theimer and Gehring 2007). Considering the highly specified co-evolution between plants and their seed dispersers, it is likely that other microbes are dispersed, in part, through actions of animal vectors. For example, it is known that seed germination rates are influenced by the passage through guts of specific animal vectors (Traveset, 1998). This phenomenon is often explained by gut scarification of the seeds (Traveset and Verdú 2002), but it remains to be tested if it also influences the microbial community in the seed coat. Furthermore, exiting the animal vector together with excrement can provide the plant with a distinct microbial community and unique environmental conditions in which to germinate (Traveset and Verdú 2002). For example, many ectomycorrhizal fungi are known to be dispersed through faeces of small 16  mammals (Maser et al. 1978; Kotter and Farentinos 1984) and marsupials (Claridge et al. 1992) that also consume and cache seeds of various plants on a regular basis.  1.3.3  Germination in the soil (Fig.1.3 – Stage 3) Once the seed arrives at its soil destination, it encounters a new, highly diverse, community of microbes. The unique composition of the microbial community in the local soil that the seed encounters at this stage creates a strong source of variation among individual plants that could be reflected in their root microbiota (Fig.1.3). Normander and Prosser (2000) compared the composition of bacterial seed coat communities in cucumber with the bacterial microbiota of its initial roots and found that the root community resembled the surrounding soil at the time of germination more than the seed coat. Little is known of how much of the earlier seed coat community is carried on at this stage (Green et al. 2006; van Overbeek et al. 2011), and exactly what determines the order and timing of colonisation in the emerging root system.  When roots enter the bulk soil, microbial communities experience what is called a ‘rhizosphere effect’ creating distinct microbial communities in the section of the soil that is under influence of rhizodeposition of cells and excretions from the root system (Nelson 2004; Doornbos et al. 2012). Part of the rhizosphere effect includes fluctuations of water content in the soil. During the day, water is drawn into the root and transported up to be released in the rhizosphere soil at night-time when transpiration in the plant is low. Increased rates of oxygen consumption as a consequence of root respiration also contribute to rhizosphere effects as well as changes in soil pH caused by proton emission through H+-ATPase in epidermal cells (Hawkes et al. 2007). As mentioned earlier, exudations of organic compounds from the roots also play an important part in creating the conditions of the rhizosphere soil by providing the microbes with resources such as carbon and salts that are scarce in the surrounding bulk soil (Doornbos et al. 2012). Together all these plant-induced changes contribute to creating a unique environment in the soil surrounding the root, attracting a distinct community of microbes.   17  1.3.4 Plant development and maturation (Fig.1.3 – Stage 4) Microbial communities associated with plant roots have shown to be dynamic over the lifespan of the host plant (Mougel et al. 2006; Houlden et al. 2008).   These succession patterns could be associated with the fact that root exudation rates gradually increase over the life of a plant until inflorescence (Badri and Vivanco 2009). Even within a root system, root exudation patterns can be dynamic, with actively growing regions secreting higher rates of exudates that stimulate microbial growth (Marschner et al. 2011). As the plant develops and matures, root morphology and exudation shift, creating new space for microbial niches and temporal dynamics in the distribution of the microbiota across the root axis. An older plant should therefore present a larger variety of microbial niches within their root systems compared with a young plant. Not only amounts, but also the composition, of exudates experience temporal and spatial dynamics (Marschner et al. 2002). This pattern was shown for exudation of organic carbon in an experiment with varieties of rice by Aulakh et al. (2001), who also noted that the exudation of sugars was substituted with exudation of organic acids as the plant grew older. This variation in root exudates ultimately affects the microbial communities in the surrounding soil, and contributes to the variation in microbial community composition at specific developmental stages (Marschner et al. 2002).   1.3.5  Inflorescence and reproduction (Fig.1.3 – Stage5) When a plant enters its reproductive state, it experiences a higher demand for resources (Dunne and Fitter 1989). This may temporarily affect the composition of the plant microbiota either by nutrient limitations or changed allocation of nutrients. For example, it has been shown that colonisation by AM fungi increases immediately prior to, or during, reproduction (Dodd and Jeffries 1986). This shift is mainly thought to aid the plant in acquiring phosphorus (Dunne and Fitter 1989). As a consequence, increased colonisation of mycorrhizal fungi has shown to augment phosphorus content in seeds and the number of seeds produced per fruit (Smith 2001). These effects of mycorrhizal colonisation on seed production have been found mainly in phosphorus-deficient soils and could be transient changes in the composition of the root microbiota (Smith 2001).  18  1.3.6 Senescence (Fig.1.3 – Stage 6) When a plant begins to die, its active resource allocation to the root microbiota decreases (Badri and Vivanco 2009). Within the root system, diminishing exudation rates in older root tissues lead to lower microbial growth rates in those zones (Marschner et al. 2002). At the same time, the senescing plant tissue provides a new source of nutrients to the part of the microbiota that exploit a saprotrophic lifestyle. Senescence of plant tissue can activate a switch in some microbes, making them convert from being passive members of the endophytic microbial community (inside plant tissues) to a saprotrophic lifestyle (Porras-Alfaro and Bayman 2011). Endophytes that exhibit this lifestyle gain an advantage when it comes to being first in place to break down the senescing tissues, which also attract other saprotrophic microbes from the surrounding bulk soil. When studying the effect of senescence on the root microbiota, it is also important to consider differences between annuals and perennials in terms of resource allocation and death of root tissue since they may exhibit completely different microbial community dynamics given the type of resources that are offered from the root system (Thomas 2013).  1.4 Stability in the root microbiota Microbial communities experience spatial distribution but are also dynamic in time. Dynamics within the microbial community could change through microbial adaptation and local selection within the community (Martiny et al. 2006), shifts in root exudation patterns, for example, in response to herbivory or as the plant ages (Marschner et al. 2002; Badri and Vivanco 2009) or by introduction or loss of habitat as roots grow and age within the system (Marschner et al. 2011) It is not clear how stable the root microbiota is throughout the lifetime of a plant. Studies of microbial community stability across developmental stages of the host suggest that the level of stability could depend on the plant species. For example, clover (Medicago truncatula) and sugar beets (Beta vulgaris var. Amythyst) show changes in microbiota associated with different developmental stages of the plants (Mougel et al. 2006; Houlden et al. 2008) whereas the microbiota of pea (Pisum satvium var. quincy) and wheat (Triticum aestivum var. pena wawa) remain stable (Houlden et al. 2008). In the case of Arabidopsis 19  thaliana, plant developmental stage appears to affect only certain bacterial taxa (Lundberg et al. 2012).  A recent study in human gut microbiota suggests that microbial communities in guts of babies start off highly unstable, but at the age of 27 months have developed a more stable community structure, much resembling that of a healthy adult (Koenig et al. 2011). For early succession of microbial communities in plants, a study looking at dynamics of the bacterial community surrounding germinating maize-seeds (the spermosphere) gives us an idea of the time frame that it may occur within. Their study showed temporal dynamics with successional shifts in the community, just hours apart (12, 24 and 36 hours into germination) (Liu et al. 2012). For plants, it is yet to be determined exactly how community dynamics shift compositionally in these early life stages and whether those events are critical for the establishment of a healthy and functional root microbiota (Green et al. 2006). By monitoring microbial communities over time it is possible to investigate not only patterns of microbial succession but also how microbial communities respond to disturbances. This information could prove important for applications in both human and environmental contexts, where the robustness of microbial communities influence personal health, success of agricultural practices and the maintenance of natural ecosystems. Robust communities are classified as either resistant (changes minimally) or resilient (recovers to the pre-disturbance state) to disturbance (Allison & Martiny 2008). Though there have been a few studies investigating the robustness of microbial communities in humans, aquatic environments and soil, a lot still remains unknown (Gonzalez et al. 2011; Costello et al. 2012; Larsen et al. 2012).  1.5 Function of the root microbiota While our aim is to show how microbial community assembly is influenced by plant phenotype, there exists considerable evidence for the reverse; that is, plant phenotype can also be influenced by the microbial community composition. For example, plants express phenotypic variance in presence or absence of mycorrhizal fungal inoculum (Wilson and Hartnett 1998; Hoeksema et al. 2010); furthermore, soils with specific microbial communities can confer benefits in growth as well as disease suppressive properties to plants 20  (Raaijmakers et al. 2009; Mendes et al. 2011). Lau and Lennon (2011) tested whether Brassica rapa grown in simplified microbial communities showed similar phenotypes to plants of the same species grown in association with more complex soil microbial communities. Their results showed that plants grown in simplified microbial communities developed phenotypes with a lower above ground biomass, reduced chlorophyll content and which produced fewer flowers and seeds. The effect of pathogenic microbes on the functioning of their host plants is well-documented. For example, pathogenic microbes are known to trigger immunological responses in plants (Jones and Dangl 2006). Understanding the interaction between crops and their microbiota is economically important to minimize loss in agricultural systems (Bakker et al. 2012) and recent research suggests that plant health can be closely related to the community assembly of its rhizosphere microbiota (Berendsen et al. 2012). Other microbes, in contrast, can benefit the host plants by increasing nutrient uptake and augment the plant’s defence capacity against other, pathogenic, microbes (Zamioudis and Pieterse 2012). This induced systemic resistance (ISR) trains the immune system of the plant to activate its defence more rapidly when attacked by a pathogen (Zamioudis and Pieterse 2012; Berendsen et al. 2012).  ISR induction has been found to be associated with multiple different bacterial factors such as flagella, antibiotics, N-acylhomoserine lactones, salicylic acid, jasmonic acid, siderophores, volatiles, and lipopolysaccharides. The bacterial groups most commonly known to induce ISR in plants belong to the genera Pseudomonas and Bacillus (Hardoim et al. 2015). Endophytic fungi are not known to be involved in ISR, but commonly produce compounds that have growth-inhibitory activities toward plant pathogens and herbivores such as alkaloids, steroids, terpenoids, peptides, polyketones, flavonoids, quinols, phenols, and chlorinated compounds (Hardoim et al. 2015). The fact that plants can be primed by beneficial microbes to defend themselves better against pathogens suggests that plants with access to beneficial microbes will develop a different immune response than those that do not encounter them and potentially experience increased fitness. The historical contingency of root microbiota could in this way lead to individual level differences in immune responses and fitness within a population of plants.  21  1.6 Summary and research objectives: Distinct microbial communities associated with individual plants may reflect their life experience, starting from conditions in the parental flower, during its dispersal and germination, to changes in the environment throughout its development and senescence. That two seeds share the same genetic makeup, originate in the same flower, or germinate a millimetre apart does not guarantee that they will host equivalent root microbiota. Part of the variation can be ascribed to genetically determined traits in the plants, affecting root morphology, root exudation and life history traits. Another important source of variation comes from environmental conditions throughout the plant’s life that shape its phenotype, such as mode of dispersal, soil type where it germinates as well as disturbances and competition with surrounding plants. Ultimately, the assembly and stability of the root microbiota could depend on many factors: the phenotype of the host plant, the stability of the soil conditions surrounding the root, the historical contingency of microbial colonisation and interactions between the microbes themselves. One could even argue that the root microbiota is an extended phenotype of the plant – a fingerprint in the soil. The research of my PhD dissertation is focused on the question of variation in the root microbiota, and which factors have the potential to cause variation in root microbiota between individual plants. In order to address these questions, I developed four research objectives that will be examined in this thesis.  Objective 1: Determine how much individual variation in root microbiota there is within and between natural populations of wild plant species. To make progress in understanding variation among microbial communities in plants, I wanted to determine the extent of variation among root microbiota of individual plants in nature. This information could prove important for applications in both human and environmental contexts, where the robustness of root microbiota may be crucial to the success of agricultural practices and the maintenance of natural ecosystems. Because plant species are known to differ in root morphology, exudation and chemical signalling, I hypothesised that plants of the same species would host more similar microbial communities than plants of different species. Because correlations have been observed between plant genotypes and their microbial community composition, I also hypothesised that individuals 22  of a plant species with a known lower genetic diversity would show less individual variation in root microbiota than individuals of out-crossing plant species.  Objective 2: Determine how much microbial community composition varies within the body of an individual plant. This was an exploratory study designed to understand the extent of microbial community variation within a single plant. I also wanted to assess whether root systems hosted unique microbial communities within the plant body, rich enough to pursue further study of individual variation and patterns of historical contingency. I hypothesised that the roots, leaves, stem and inflorescence of the plant host distinct microbial communities.  Objective 3: Examine whether historical contingency of the root microbiota contributes towards individual variation among plants. We still do not fully understand temporal dynamics of the microbiota in plants. Does the historical contingency in a plant seedling affect the structure of its microbial community as an adult plant? Will a plant with an established microbial community, once introduced to a new microbe, change associations or keep the ones already formed? I considered the idea that not only plant genetics and environment, but also historical contingency contributes towards creating individual root microbiota. Based on the idea that early colonisers are thought to gain an advantage in community assembly, and that the root system is thought to change continuously throughout plant development, I further hypothesised that plants exposed to an inoculum at the same developmental stage would form more similar root microbiota than plants exposed to the inoculum at different developmental stages. Objective 4: Determine if microbial community stability is equal during all stages of early plant development. I wanted to assess the robustness of bacterial and fungal root communities at various stages of early plant development. Based on what is known about microbial community stability in host systems, I hypothesised that the root microbiota in plants perturbed earlier in development would be more likely to change in microbial community composition, creating more variation between individual plants. 23  Chapter 2: Wild plant species growing closely connected in a subalpine meadow host distinct root-associated bacterial communities  2.1  Background Plant roots function as distinct habitats within the soil and bacterial communities in root systems have repeatedly been shown to differ from those of the surrounding bulk soil (Smalla et al. 2001; Haichar et al. 2008; Gottel et al. 2011; Lundberg et al. 2012). Even though root-associated bacterial communities (both rhizospheric – in the soil surrounding the roots, epiphytic – living at the surface of roots and endophytic – living inside root tissues) have been under investigation for many years, there is still little consensus about how these communities are formed and what determines their composition (Berg and Smalla 2009; Aleklett and Hart 2013; Bulgarelli et al. 2013).  Traditionally, the composition of bacterial communities living in association with plants has been attributed to environmental factors. For example, soil type has been suggested as the strongest determinant of community structure in root-associated microbial communities (De Ridder-Duine et al. 2005; Singh et al. 2007; Lundberg et al. 2012; Bulgarelli et al. 2013). At the same time, it has also been argued that the host plant may play an equally large role in determining the composition of its root microbiota (Marschner et al. 2005; Costa et al. 2006; Hartmann et al. 2008; Doornbos et al. 2012), especially endophytic bacterial communities (Haichar et al. 2008).  Recent work has demonstrated that hosts can alter their root microbiota by regulating soil conditions in the vicinity of the root system through root exudation of sugars, phenolics and amino acids, which could also function as signalling molecules with the microbes in the surrounding soil (Chaparro et al. 2013). Since root exudation patterns and composition can be associated with plant gene expression, variation in host genetics has the potential to create large differences in the chemical profile of plants and, consequently, the composition of microbes able to inhabit the root system. Several studies have found that different plant species or genotypes of the same species host distinct microbial communities (Marschner et 24  al. 2005; Bailey et al. 2005; Schweitzer et al. 2008; Micallef et al. 2009; Manter et al. 2010; Becklin et al. 2012; Peiffer et al. 2013). Even in studies of endophytic and rhizosphere soil bacteria, where soil type was considered to have the strongest effect on structuring the root microbiota, differences in bacterial community composition between plant genotypes was still detected (Bulgarelli et al. 2012; Lundberg et al. 2012). The root environment varies greatly among plant species (Bardgett et al. 2014); these differences may lead to the selection of distinct bacterial communities. Plants can differ in terms of root lifespan (Roumet et al. 2006), root architecture (Hodge et al. 2009), root surface structure, and components and patterns of root exudation (Bais et al. 2006). Root exudates are known to provide a food source for the microbes (Farrar et al. 2003), stimulate  symbiotic associations (such as mycorrhizal infection or nodule formation) (Bais et al. 2004), and defend the plant against pathogens (Doornbos et al. 2012). All these plant characteristics could contribute to shaping root systems of different plants into local habitats and potentially distinct niches for microbial colonizers. The role of intra-species variation among root-associated microbial communities has been overlooked, but might represent a significant proportion of variation in natural systems (Bell et al. 2014).  Since we know that natural populations exhibit variation in root exudation patterns and root morphology, one would expect there to be variation among individual plants in their root microbiota as well (Micallef et al. 2009). But variation among plants might also be driven by environmental heterogeneity because we know that small-scale environmental heterogeneity exists in soil systems. Is this variation static across plant taxa, or do different taxa exhibit more variation than others? If plant genetics determine bacterial community composition, then certainly, populations with low genetic diversity (i.e. asexually reproducing, or metapopulations), would be expected to have less variation than sexually reproducing populations with high levels of gene flow. In this study, we sampled Pilosella aurantiaca, a plant species known to reproduce clonally through stolons and apomixis, and then examined whether individuals within that species had less dispersion in their microbial community composition than individuals in two other, sexually reproducing, plant species, Trifolium hybridum and Leucanthemum vulgare.    The majority of studies characterizing bacterial communities in the root microbiota have been conducted with model plants in artificial greenhouse settings or agricultural 25  contexts (Marschner et al. 2001; Garbeva et al. 2004; van Overbeek and van Elsas 2008; Micallef et al. 2009; Manter et al. 2010; Doornbos et al. 2012; Bulgarelli et al. 2012; Lundberg et al. 2012; Chaparro et al. 2013) where the study of genetically modified plants have been especially informative when it comes to understanding slight differences between plant genotypes (van Overbeek and van Elsas 2008; Weinert et al. 2009; Inceoğlu et al. 2010). While these studies are crucial for understanding the mechanistic basis of plant:microbe interactions, they do not reflect how natural environmental conditions contribute to variation in bacterial community composition across individual plants, particularly in complex environments where a wide diversity of plants and biota are interacting. In this study, we explored variation in bacterial community composition between individual root systems of neighbouring plants in a common field in order to determine how much variation exists within and between plant taxa. We sampled the root microbiota of three plant species growing within 10 meters from each other in a field and asked – are bacterial root communities distinct among plant species growing in a common location, close enough to be exposed to the same environmental conditions? And – do certain plant species contain more intra-species variance in bacterial communities than others?  2.2 Methods 2.2.1 Field site and target plant  Samples were collected in August, 2011, from a subalpine meadow near Chute Lake, British Columbia, Canada (49.698859N,-119.533133W). The sampling area has not been used for agriculture or forestry but is in proximity to a forestry road as well as a camp site. Since it contains a high number of invasive plant species, it could also be considered a disturbed site. The soil at the site was determined to be a sandy loam, and the site is within the Interior Douglas Fir, dry warm (IDFdw) biogeoclimatic zone (Biogeoclimatic Ecosystem Classification (BEC) and Ecology Research program of the British Columbia) (Meidinger and Pojar 1991). This zone is characterised by cold winters and dry, warm summers with low amounts of precipitation due to the rain shadow created by the Coast, Cascade and Columbia mountains. The plant community sampled for this experiment was growing up on a rock 26  plateau, away from tree cover (Fig.2.1). The dominant vegetation at the field site consisted of Orange hawkweed (Pilosella aurantiaca (L.) F.W. Schultz and Schultz-Bip), Hairy vetch (Vicia villosa, Roth), Oxeye daisy (Leucanthemum vulgare, Lam.), Wild strawberry (Fragaria virginiana, Duchesne), Timothy (Phleum pratense, L.) and Alsike clover (Trifolium hybridum, L.) growing homogenously, but at different abundances across the field (Fig.2.1).    Figure 2.1 This picture shows the layout of the field site at Chute Lake, BC, Canada, inhabited by a mixed plant community of herbs including Orange hawkweed (Pilosella aurantiaca). Oxeye daisy (Leucanthemum vulgare) and Alsike clover (Trifolium hybridum,) which were sampled in this experiment.  Our target plant was P. aurantiaca (formerly known as Hieracium aurantiacum) (Fig.2.2), which is native to Europe and invasive in North America. Because of P. aurantiaca’s stoloniferous reproduction, the species is able to form dense mats in the landscape, making it hard for other plant species to compete for space through seed germination in the field (Giroday and Baker 2006). This is one of the features that have made the species a successful invasive plant within British Columbia (Giroday and Baker 2006). Genetic diversity within P. aurantiaca has previously been examined across 48 locations in North America, and results showed that there were only three genotypes, of which two were found only in 27  isolated locations (one in Alaska and one in Oregon) (Loomis and Fishman 2009).  By choosing to work with a plant expressing this low diversity in wild populations, we hoped to minimize genetic variation within the population that we sampled. To clarify the role of host identity and intra-species variance in bacterial root microbiota, we additionally sampled two of the co-occurring plant species, L. vulgare and T. hybridum , which were in the same developmental stage (flowering) as P. aurantiaca.    Figure 2.2 The target plant, Pilosella aurantiaca is known to be a clonal species with very little geratic variation within North America.  2.2.2 Experimental design Root systems of P. aurantiaca were collected 1 m apart along two 10 m transects (n=20) in order to explore within-species variance of this clonal species. To be able to compare bacterial community composition among species, additional samples of T. hybridum (n=10) and L. vulgare (n=10) were also collected, where present, along the transects, several of which were growing within centimetres of sampled P. aurantiaca. Because T. hybridum and L. vulgare were not as abundant within the field, their original sample size was lower. For 28  statistical comparisons between species however, an equal sample size of 8 plants/species was used throughout analyses. Each root system was rinsed from surrounding rhizospheric soil in de-ionized water in order to separate it from roots of neighbouring plants. Root systems were then cut up in pieces and a subsample of root tissue, representative of the whole root system, including young fresh roots as well as older root tissues (with no exclusion of nodules in T. hybridum), was collected and further used for classification of bacterial community composition. Since no further treatment was performed in order to remove rhizoplane microbes, we assume that the communities extracted could be of either endophytic or epiphytic origin.  2.2.3 Bacterial community analysis 2.2.3.1  Amplification and sequencing of target gene DNA from all collected plant tissues (0.25g/sample) was extracted using a PowerSoil DNA Isolation Kit (MoBio Laboratories Inc., USA) according to the manufacturer’s protocol. Bacterial diversity and the relative abundances of individual taxa were assessed by barcoded pyrosequencing of a portion of the 16S rRNA gene. Each DNA sample was amplified in triplicate through PCR reactions using the protocol described in Fierer et al. (2008) except with a different primer pair. The forward primer contained the 454 Life Sciences primer B sequence, the bacterial primer 799f (Chelius and Triplett 2001) and a two-base linker sequence (‘AG’). The reverse primer contained the 454 Life Sciences primer A sequence, a unique 12 bp error-correcting Golay barcode (Fierer et al. 2008), a ‘GT’ linker sequence, and the ‘universal’ bacterial primer 1115r (Reysenbach and Pace 1995).  The targeted gene region has been shown to be appropriate for accurate taxonomic classification of bacterial sequences and the primers are designed to exclude chloroplasts from plant tissues in the samples (Redford et al. 2010). Amplicons were visualized via gel electrophoresis, purified and quantified. Amplicons from all samples were then combined in equimolar ratios into a single tube. Samples were sequenced at Engencore (University of South Carolina) on a Roche GS-FLX sequencer running the Titanium chemistry.     29  2.2.3.2  Processing raw sequence data All sequences were de-multiplexed and further analyzed using the Quantitative Insights Into Microbial Ecology (QIIME) toolkit (Caporaso et al. 2010). Sequences that shared 97% sequence identity were brought together in clusters to represent what we refer to as Operational Taxonomic Units (OTUs). Sequence identity was determined by calculating the number of matching (identical) letters divided by the length of the shorter sequence. A representative sequence was then chosen from each OTU cluster and used to assign taxonomy to the OTU by comparing it to  the Greengenes reference database (February 14th 2012 version) (DeSantis et al. 2006) using the Basic Local Alignment Search Tool (BLAST) classifier (Altschul et al. 1990). In order to correct for differences in the number of sequences analyzed per sample, a randomly-selected subset of 400 sequences per root sample was used to compare relative differences in taxonomic diversity. Only samples from which we obtained a minimum of 400 bacterial sequences per sample or more were considered in the study, eliminating 3 samples from the study (one P. aurantiaca and two L. vulgare). Though 400 sequences cannot fully capture the rare biosphere, it allowed us to compare samples while still maintaining as many samples as possible. It has previously been shown that studies of bacterial communities show similar results even at a lower rarefaction (Hamady and Knight 2009; Kuczynski et al. 2010). In fact, re-analyzing our data set with a higher rarefaction limit showed the same general trends but drastically lowered our number of samples available to analyse.  2.2.4 Statistics Differences in community composition between samples were calculated using a phylogenetic metric (UniFrac), where weighted UniFrac shows an emphasis on the more abundant taxa in samples and un-weighted UniFrac treats all taxa the same (Lozupone et al. 2007; Hamady et al. 2010). As a comparison, we also included a taxonomic metric (Bray-Curtis distance) to explore whether dissimilarity patterns were the same in terms of presence/absence of taxa. Before calculating Bray-Curtis distances, all relative abundances were log-transformed. Principal Coordinate Analysis (PCoA) is used to compare groups of samples based on distance metrics (in our case UniFrac or Bray Curtis distances). The distances between samples are projected into Euclidean space in a large number of 30  dimensions. Pairs of principle coordinate (PC) axes can then be compared plotting samples in 2-dimensional scatterplots in order to assess how samples relate to each other. The percent variation (spread of the data values) explained by each PC will then also be listed for each axis. 2-D scatterplots of Principal Coordinates Analysis (PCoA) were generated in PRIMER-E (Clarke and Gorley, 2006) and used to visualize the greatest amount of variability in the pair-wise distances between samples.  We tested for variation among host plants in their root microbiota using a 2-way PERMANOVA (Anderson 2005) with host species and transect as factors and weighted and un-weighted UniFrac as well as Bray Curtis as our dissimilarity metrics. Sequential sums of squares from permutations of the raw data were used to provide Pseudo F-ratios, and the probability of observing the Pseudo-F ratios was calculated by comparing our results to the null distribution of 9999 permutations. To minimise un-equal sampling efforts (Anderson and Walsh 2013), we subsampled 8 samples from each species, which were further used for both PERMANOVA and PermDISP analyses.    Table 2.1 PERMANOVA results, comparing bacterial community resemblance between plant species and transects using different diversity metrics. Factor Diversity metric Pseudo-F P (perm) Species Weighted UniFrac 6.15 0.0001  Un-weighted UniFrac 1.43 0.0001  Bray Curtis  1.88 0.0001 Transect Weighted  UniFrac 1.22 0.23  Un-weighted  UniFrac 1.06 0.22  Bray Curtis  1.07 0.24 SpXTr Weighted  UniFrac 1.54 0.09  Un-weighted  UniFrac 1.03 0.28  Bray Curtis 1.06 0.20  31  Variability in community composition within each of the three species was analyzed through PermDISP (Anderson 2004) (9999 permutations), creating a centroid for each  species and measuring the average spread of samples belonging to that species from the centroid. A large spread (high average) would indicate a high variability in community composition among individuals within the species (Anderson 2004).  2.2.5 Spatial structuring of plant-host-associated bacterial communities in the field Since plants were collected along two transects, the geographic distance between all samples within each transect was recorded and used to assess whether root systems growing in closer proximity to each other hosted more similar bacterial communities. To determine the effect of geographic distance within the field on variation in root microbiota, we compared the distances between roots in terms of both phylogenetic similarity (UniFrac) and spatial distance (Euclidean distance) using Spearman’s test in RELATE. This analysis determines the degree of correlation between two distance matrices (PRIMER-E) (Clarke and Gorley, 2006).  2.3 Results 2.3.2 Variation between host species When comparing the phylogenetic overlap between bacterial root microbiota (UniFrac) across three different species of plant hosts growing in a common field, bacterial communities from samples of the same plant species were significantly more similar to each other than to bacterial communities sampled from plants of the two other species (Table 2.1). This was true for both weighted (Pseudo-F=8.54 p=0.0001) and un-weighted UniFrac (Pseudo-F=1.66 p=0.0001) as well as Bray Curtis dissimilarities (Pseudo-F=2.27 p=0.0001) (Table 2.1). These patterns were also evident from the principal coordinates analyses, which showed little overlap between samples of different plant species (Fig.2.3). 32   Figure 2.3 Principal coordinates analysis plot illustrating the phylogenetic overlap in root prokaryotic community composition among samples from three different plant species. Phylogenetic overlap between communities was assessed using weighted UniFrac, and PERMANOVA results showed that community composition was significantly different among plant species (p <0.001).  2.3.3 Variation within host species Plant species differed in how much variance there was among bacterial communities of individual root samples. There was a significant difference between plant species in the amount of compositional dissimilarity of bacterial taxa between individual plants (Bray Curtis: F=9.30 p=0.005) (Table 2.2). That is, the amount of dispersion of individual plants from the centroid differed among plant species.  In this case, P. aurantiaca exhibited the highest average dispersion among plant individuals, while T. hybridum showed the least dispersion (Table 2.2). This was not the case when the same data was analysed using phylogenetic measures, since we could not detect any significant difference in dispersion among plant taxa (UniFrac: weighted F=3.10 p=0.11; un-weighted F=1.04 p=0.37) (Table 2.2).  33   Table 2.2 PermDISP results showing the average spread from centroid and standard error (SE) for samples of each species. The PERMANOVA (P(perm)) values assess whether there is a significant difference between species in sample dispersion, using different diversity metrics. Diversity metric F P (perm) Species Average SE Weighted UniFrac 3.10 0.11 P. aurantiaca 9.96 E-2 4.28 E-3    T. hybridum 8.57 E-2 5.70 E-3    L. vulgare  0.12 1.29 E-2 Un-weighted UniFrac 1.04 0.37 P. aurantiaca 0.47 3.39 E-3    T. hybridum 0.46 5.42 E-3    L. vulgare  0.47 7.25 E-3 Bray Curtis 9.30 0.005 P. aurantiaca 51.43 0.73    T. hybridum 47.13 0.44    L. vulgare  51.05 1.05  2.3.4 Relative abundance of taxa across hosts A total number of 4384 unique OTUs were analyzed within the rarefied data set. A taxonomic summary, showing the average abundance of bacterial phyla in P. aurantiaca, T. hybridum and L. vulgare samples, illustrates the compositional differences between root systems of different plant species (Fig.2.4). In P. aurantiaca, the most abundant phylum was Betaproteobacteria which made up, on average, 29% out of all sequences found in P. aurantiaca samples, followed by Bacteroidetes (19%), Alphaproteobacteria (16%) and Actinobacteria (12%). In T. hybridum, Betaproteobacteria made up, on average, 51% of the all bacterial sequences found in the species, followed by Alphaproteobacteria (21%) and Bacteroidetes (16%). Bacterial communities in L. vulgare samples were, similarly to T. hybridum, dominated by Betaproteobacteria (50%), followed by Bacteroidetes (18%) and Alphaproteobacteria (12%). An overview of the 14 most abundant OTUs across all samples 34  as well as their relative abundance within samples of the different plant species is given in Table 2.3. A closer examination of the relative abundances of all Betaproteobacteria found in samples showed that while Burkholderiales was the predominant order across all three species, T. hybridum and L. vulgare samples were heavily dominated by bacteria of the family Oxalobacteriaceae – especially bacteria of the genus Herbaspirillum (11% of the total bacterial community in T. hybridum and 18% in L. vulgare) (Fig.2.5).  2.3.5 Variation due to spatial distance Results of the RELATE test, exploring the potential correlation between geographic distance (measured in the field within transects) and bacterial community relatedness, showed mixed results depending on which metric was used and which transect was investigated (Table 3). Among samples from the west transect, results from none of the metrics were significantly correlated to geographic distance (UniFrac: weighted p=0.69; un-weighted p=0.66).  For the samples from the east transect, the only condition under which we could detect an effect of spatial distance on root community similarity was using un-weighted UniFrac values (p=0.05). Values of weighed UniFrac (p=0.19) were not correlated to geographic distance (p=0.86). 35   Figure 2.4 Comparison of the average bacterial community composition and relative abundances, at the phylum level (Proteobacteria divided into class) in root samples from three different plant species. Results show a strong dominance of sequences belonging to Betaproteobacteria in all three plant species, but especially in T. hybridum (51%) and L. vulgare (50%). The phyla representing less than 1% out of the total community were grouped as “Other” and consisted of: NKB19, Nitrospirae, PAUC34f, Cyanobacteria, Elusimicrobia, Fibrobacteres, Chlamydiae, SC4, Spirochaetes and Thermi. Sequences not matching the database were recorded as “No blast hit.” 36  Table 2.3 The core root microbiota represented by the fourteen OTUs with the highest abundances across all samples. Values are calculated as the average percent out of the 400 sequences recorded for each sample, across all species (Total %). As a comparison, data is also included for what percentage (on average) the fourteen OTUs make up within the bacterial communities of the three plant species examined (P. aurantiaca, T. hybridum and L. vulgare).  # OTU ID  Total %   P. aurantiaca  %  T.  hybridum  %  L. vulgare %  Phylum  Class  Order  Family  Genus  Species 1537 7.6 2.8 9.4 16.8 Proteo- bacteria Beta- proteobacteria Burk-holderiales Oxalo-bacteraceae Herba- spirillum  19032 3.1 3.2 3.9 2.2 Proteo- bacteria Beta- proteobacteria Burk-holderiales    22328 2.7 1.0 5.1 3.8 Proteo- bacteria Beta- proteobacteria Burk-holderiales Coma-monadaceae Limno- habitans  30435 1.5 1.7 1.3 1.2 Proteo- bacteria Beta- proteobacteria Burk-holderiales  Methy-libium Methylibium petroleiphilum 4453 1.3 0.8 1.7 2.1 Proteo- bacteria Beta- proteobacteria Burk-holderiales Oxalo-bacteraceae Janthino-bacterium Janthinobacterium lividum 1231 1.2 1.9 0.5 0.4 Bacteroi-detes Sphingo- Bacteria Sphingo-bacteriales Flexi-bacteraceae Cytophaga  22285 1.0 1.5 1.0 0.4 Proteo- bacteria Alpha- proteobacteria Rhizobiales Brady-rhizobiaceae Brady-rhizobium  20009 1.0 1.5 0.2 0.6 Actino- bacteria Actinobacteria Actino-mycetales Thermomon-osporaceae Actino-corallia Actinocorallia longicatena 25072 0.9 1.0 1.0 0.7 Proteo- bacteria Gamma- proteobacteria Xantho-monadales Xantho- monadaceae Rhodano- bacter Rhodanobacter lindaniclasticus 1340 0.9 1.0 1.0 0.6 Bacteroi-detes Sphingo- Bacteria Sphingo-bacteriales Sphingo- bacteriaceae Sphingo- bacterium Sphingobacterium faecium 5213 0.8 0.6 0.9 1.3 Bacteroi-detes Sphingo- Bacteria Sphingobacteriales Sphingo- bacteriaceae   29492 0.8 1.3 0.1 0.6 Proteo- bacteria Gamma- proteobacteria Chro-matiales Sino-bacteraceae   26917 0.8 1.1 0.3 0.7 Chloroflexi Chloroflexi Rosei-flexales Kouleo-thrixaceae Kouleo-thrix  1184 0.8 0.4 1.3 1.1 Bacteroi-detes Flavobacteria Flavo-bacteriales Flavo-bacteriaceae Chryseo-bacterium  37   Figure 2.5 The average relative abundance of sequences belonging to the Betaproteobacteria families and Oxalobacteraceae genera found in root samples of the three plant species. The table shows a more detailed overview of which families within the Betaproteobacteria class that sequences were most commonly assigned to. At the bottom of the table, the total percent of sequences belonging to Betaproteobacteria across all samples of each plant species is listed, and above that, how these percentages are divided between different families. Values are given as the percentage of sequences belonging to a certain taxa out of the total average bacterial community for each of the three plant species (rarefied at 400 sequences/sample). The heat map is colour coded from blue (low abundance) to red (high abundance) to facilitate the overview. To the right, we have also included an even more detailed overview of how the total percentages of sequences belonging to the family Oxalobacteraceae from each plant species were divided at a genus-level. By doing that, we could see that these sequences were mainly classified as Herbaspirillum. 38  2.4 Discussion  2.4.1 Host specificity Our study shows that root bacterial communities vary significantly between plants belonging to three different species, growing in close proximity to each other in natural plant communities. Because the plant roots collected in our study were growing in such close proximity to each other, it is unlikely that variation in soil conditions (usually thought to be one of the main drivers of microbial community structure) would be a significant source of variation among our root samples. Our results also confirmed that there was no overall correlation between geographic distance between samples and bacterial community relatedness. While these results would need to be further confirmed across multiple field sites in order to draw general conclusions about whether this is just a local trend or a general pattern across plant species and soil types, our results support previous work showing bacterial host plant specificity in roots of agricultural crops (Marschner et al. 2001; Wieland et al. 2001; Haichar et al. 2008) and wild grass species (Kuske et al. 2002; Osanai et al. 2013).  Although all plant species investigated in this study (P. aurantiaca, T. hybridum and L. vulgare), are perennial, there are significant differences in root morphology between the species. For example, P. aurantiaca and L. vulgare (both belonging to the family Asteraceae) have creeping root stocks and produce fibrous root systems whereas T. hybridum (family Fabaceae) grows a branching tap root system that is known to form nodules with nitrogen fixing bacteria. This variation in root morphology could contribute to the differences in abundance and composition of bacteria in our results. For example, roots that penetrate deeper soil may encounter different microbes than those in shallow layers (Fierer et al. 2003). Similarly, the thickness and/or texture of the root surface (i.e. woody, fibrous) may be more or less penetrable to colonizing bacteria. Part of the variation seen in bacterial community composition between the three plant species could also be caused by species-specific root exudation patterns. For example, several members of the Asteraceae are known to produce allelochemicals that could affect the bacterial community as well as surrounding plants (Alford et al. 2009). However, these differences are difficult to assess in wild plant communities, especially when roots of 39  different plant species grow in close proximity to each other with entangled root systems.  In our study, the roots grew so intimately that exudation from one plant species could have influenced root systems of neighbouring plants.  2.4.2 Individual variation in root microbiota We know that genetic differences between plants, even at the genotype level, can affect the composition of the root microbiota (Bailey et al. 2005; Schweitzer et al. 2008; Peiffer et al. 2013).  Thus, we would expect variation in bacterial community composition among individuals within a population of plants, even when they are growing in a common environment, due to genetic variation in the population in terms of root traits and exudation chemistry, among other factors. Though there is a potential for clonality in T. hybridum, we still predicted less individual variance in the root microbiota among individuals from P. aurantiaca (thought to consist of mainly one genotype across all of North America (Loomis and Fishman 2009)) than within the two co-occurring out-crossing plant species with presumed higher genetic diversity (L. vulgare and T. hybridum).  Though our data show a significant difference in compositional turnover within different plant species, it rejects the hypothesis that P. aurantiaca had the most similar root communities across individuals. Comparing the average dispersion of bacterial community composition for the three plant species, there was no indication that P. aurantiaca had a smaller dispersion than the two other plant species (Table 2.2). Instead, it shows that P. aurantiaca had the highest variation within a species comparing dispersion based on taxonomic differences (Table 2.2). The fact that we could not detect any differences in dispersion when using phylogenetic metrics suggests that individual root systems differ more in terms of which taxa are present or absent than how related they are, or that there is little phylogenetic conservatism at the individual level. Overall, this study shows that the extent of individual variation seen in root microbiota varies between species, but that a plant species thought to be more genetically homogenous does not necessarily host more homogeneous root communities. It also indicates that individual variation in bacterial community composition in root systems is determined, not only by plant genetics, but also by small-scale variation in the surrounding environment (soil 40  chemistry, humidity, biotic interactions etc.) and potentially, events throughout the plant’s life that could affect root colonisation (Aleklett and Hart 2013).  2.4.3 Bacterial community composition Similar patterns of bacterial community composition to what we found in our plants, growing in a subalpine meadow in Canada, have been reported in rhizosphere samples of other studies.  For example, roots tissues of the plant species that we sampled were mainly dominated by Betaproteobacteria, (Fig. 2.4), especially members of the order Burkholderiales and the family Oxalobacteriaceae, which represented as much as 32% of the total bacterial community in L. vulgare (Fig. 2.5). Seed- and root-colonizing populations of Oxalobacteriaceae have previously shown to be responsive to plant species (Green et al. 2007), supporting our data of plant species hosting distinct bacterial communities. Dominance by these taxa in root systems has also been reported in other studies. For example, roots of Arabidopsis thaliana, examined at the same taxonomic resolution by Lundberg et al. (2012), were dominated by Betaproteobacteria and Oxalobacteriaceae both in samples of rhizosphere soil as well as in the endophytic root compartment. In sphagnum mosses, Burkholderiales has also been documented as one of the  dominant bacterial groups, thought to be behind the production of anti-fungal compounds and anti-microbial properties of the sphagnum mosses (Opelt et al. 2007). Other studies have found Actinobacteria to dominate in root tissues of plants (e.g. Ottesen et al. 2013), especially in communities of the endophytic compartment (Bodenhausen et al. 2013). In our study, Actinobacteria represented at most 12% out of the total bacterial community in the plant species that we sampled (Fig. 2.4) and was mainly found in samples of P. aurantiaca that, in general, were less dominated by Betaproteobacteria. The dominance of sequences belonging to the genus Herbaspirillum was further emphasized when we examined the fourteen most abundant OTUs across all samples (Table 3). Herbaspirillum spp. are known to colonize apoplastic or intracellular spaces of plant tissues and several species have shown the ablility to fix nitrogen (Schmid et al. 2006). While it is believed that this nitrogen-fixing ability could be beneficial to their plant host, it has also been documented that certain Herbaspirillum strains are mild pathogens and a causative 41  agent of “mottled stripe disease” in crops such as sugar-cane (Schmid et al. 2006). Besides Herbaspirillum, we also saw high abundances of sequences belonging to Limnohabitans and Cytophaga (Table 2.3), two genera more commonly associated with bacterial communities in fresh water (Kirchman 2002; Šimek et al. 2010) as well as the species Methylibium petroleiphilum, a recognized methylotroph (Kane et al. 2007) and Janthinobacterium lividum, known to thrive in soils (Shivaji et al. 1991) and produce antibiotics (Johnson et al. 1990). The high presence of these groups in our samples could be due to the inclusion of epiphytic members of the root microbiota, where bacteria associated with water films and soil particles of the root surface would be expected.  In comparison, Bodenhausen and colleagues (2013) found that a Flavobacterium of the phylum Bacteroidetes stood out as the single most abundant OTU in endophytic root samples, making up 10.15% of the total community. Though bacteria of the phylum Bacteroidetes represented a significant part of the community in root samples of the three plant species sampled in our study (Fig.2.4), they were by no means the most dominant taxonomic group in any of the species (Table 2.3). As the genus Trifolium are known to be hosts of nitrogen-fixing bacteria that form nodules in their roots, T. hybridum samples were expected to host larger populations of Alphaproteobacteria, specifically belonging to the order Rhizobiales which is a  common symbiont of legumes (Masson-Boivin et al. 2009). This expected pattern was not evident in our results though. The only OTU belonging to the order Rhizobiales detected in notable abudances in our study was a Bradyrhizobium taxa which made up 1% of the collective community of T. hybridum samples and 1.5% in P. aurantiaca samples (Table 3). Instead, it was evident that the T. hybridum community was dominated by the family Oxalobacteraceae (23.48%) (Fig.2.5) and specifically one OTU of the genus Herbaspirillum (10.75%) (Table 3), which is mainly known to colonize roots of non-leguminous plants, and have nitrogen-fixing properties (Baldani et al. 1997). What stands out though is that this group of bacteria was even more predominant in L. vulgare samples, where Oxalobacteraceae made up 32.47% of the community and the same Herbaspirillum OTU represented 18.26% of the total community.    42  2.4.4 Variation in relative abundances of bacterial taxa across plant species  We observed differences in the evenness of bacterial taxa across host plants. While the roots of L. vulgare and T. hybridum seemed dominated by a few select groups of microbes, samples of P. aurantiaca supported communities with abundances more evenly distributed among bacterial taxa (Fig. 2.4; Table 2.3). Though few studies have looked specifically at variation in bacterial evenness between plant species, it could be an important source of variation. For example, dominance of single taxa may indicate specialized plant/bacterial associations whereas high evenness in community composition could reflect generalist associations among plants and bacteria. Alternatively, differences in evenness may result from microbial interactions within the plant, not driven by the plant but microbial competition for plant resources.  2.5 Conclusions In this study, we showed that plant identity plays a major role in explaining the variation seen in root microbiota both between and within plant species growing under natural conditions. Further studies across a larger set of wild plant species and natural sites as well as more detailed investigations of the effect of plant genetics versus plant phenotypic traits on bacterial community assembly could help resolve the relative contribution of host identity at an individual level in shaping the root microbiota. It would also allow us to draw further conclusions as to whether plats more related to each other actually host more similar bacterial communities across plant species and families. The results of our study speak of how intimately related bacterial communities are with their host plants. Root systems of wild plants are never alone; they are constantly surrounded by the roots of other plants, entangled in the soil, competing for resources and space. Yet, our results show that bacterial communities associated with roots of plants growing in a common field are distinct between plant species. Ultimately, we are not able to tell exactly why these three plant species have such distinct bacterial root communities, but further studies linking metabolomics of wild plants with bacterial community composition would be useful for better understanding how plants affect bacterial community assembly.  43  Chapter 3: Bacterial communities across the body of a single plant  3.1 Background Plants are known to be hosts of diverse microbial communities, both in their aboveground and belowground structures (Lindow and Brandl 2003; Buée et al. 2009; Vorholt 2012). Since plant bodies can be considered as ecosystems, containing a specific set of habitats for microbes to colonise (Andrews and Harris 2000; Aleklett et al. 2014) it is further important to understand the spatial heterogeneity in microbial distributions across the plant body in order to gain a more complete picture of the diversity and composition of bacterial communities living in association with plants. The fact that the composition of these microbial communities has been further linked to plant health and productivity (Berendsen et al. 2012; Bakker et al. 2012), makes it even more important to understand how these intimate relationships between microbes and hosts are formed, and what factors are crucial to maintain their co-existence.  Distinct microbial habitats within the plant could be created by various factors, as microbial community formation would be influenced by: (1) Plant exudates: It is known that certain parts of the plant more frequently exude energy rich sugars that could function as a nutrient source for the microbes (Heslop-Harrison and Heslop-Harrison 1985; Fourie and Holz 1998; Badri and Vivanco 2009), as well as antimicrobial volatiles, known to deter specific groups of bacteria (Junker et al. 2011). (2) Tissue surface structure: Variation in epidermal cell structure between plant parts could affect the ability for microbes to colonise different surfaces epiphytically as well as endophytically, for example by affecting their ability to penetrate cell walls (Ngugi and Scherm 2006). (3) Surrounding source environment: The various parts of the plant body are exposed to very different surroundings, as most plants grow both above and below ground, in contact with both microbial communities of the surrounding air and soil (Bulgarelli et al. 2013). Besides providing the various plant tissues with very different source communities for colonisation, the surrounding environment will also influence the abiotic conditions present in different plant body parts.  44  All this variation contributes towards creating micro-habitats within the plant body, containing more or less rich nutrient sources, and with varying abiotic conditions. It could, therefore, be expected that this variation also would be reflected in the bacterial community composition across the plant body, creating communities specialised in the living conditions presented by different parts of the plant.  In this experiment I explored the bacterial diversity contained within the body of a single plant, sampled from roots to flowers, and examined whether micro-habitats within a single plant determined by morphological features select for specific bacterial communities.    Figure 3.1 Image displaying the complete plant specimen of the species Pilosella aurantiaca  which was sampled in 93 pieces  in order to examine bacterial community composition across the body. 45  3.2 Methods We sampled an individual plant of P. aurantiaca growing in the wild (for description of the field site, see details in Methods Chapter 2) with a 15 cm deep soil core surrounding the plant. The plant was handled with sterile gloves and transported to UBC-Okanagan where it was dissected into 93 separate pieces collected from different compartments of the plant (19 inflorescence samples, 19 stem samples; 19 leaf samples; 4 samples from the slightly thickened section where the stem and the root intersects (referred to as ‘bulb’ in this chapter) ; 28 samples from roots and 4 samples from a reproductive runner). For an overview of how the plant was divided into pieces, see Fig.3.3 where every piece of the plant sampled is indicated. Instruments were sterilised in between each dissection cut and plant pieces were immediately transferred into the MoBio PowerSoil kit for DNA extraction. Extraction was carried out according to manufacturer’s protocol, and DNA amplification, sequencing and data processing was performed according to methods described in Chapter 2.  To control for sampling effort, the data set was rarefied at 400 sequences per sample. α-diversity was compared between different sections of the plant body as average OTU richness, and differences were assessed through a type III ANOVA, with an additional Duncan test. Analyses were done in R using the packages ‘vegan’ (Oksanen et al., 2014) and ‘laercio’ (Da Silva 2015). Observed number of OTUs (out of a sub sample of 300 sequences) were further mapped across the plant samples from the individual plant, and visualised in SitePainter (Gonzalez et al. 2012). Phylogenetic differences between bacterial communities in samples from different plant parts were visualised in PCoAs, and β-diversity was further assessed through a 1-way type III PERMANOVA (9999 permutations) in PRIMER-E (Anderson 2004) with UniFrac values (weighted and un-weighted) as dissimilarity metrics.   3.3 Results Out of 93 samples taken from various parts of the plant body, bacterial DNA was only successfully amplified and sequenced from 60 samples. The total OTU richness across all samples (rarefied at 400 sequences) was 5695, of which 1204 OTUs were found in samples from the inflorescence (n=7, average 253 OTUs/sample, SEM+/- 8.9), 1324 in leaf samples (n=10, average 198 OTUs/sample, SEM+/- 23.1), 669 (n=4, average 199 OTUs/sample, 46  SEM+/- 48.9) in the runner, 694 in bulb samples (n=4, average 215 OTUs/sample, SEM+/- 17.9), 2930 in the roots (n=21, average 264 OTUs/sample, SEM+/- 8.2) and 1852 in samples of the stem (n=14, average 220 OTUs/sample, SEM+/- 19.8). ANOVA results showed that there was a significant difference in OTU richness between body parts (p=0.04, F=2.56), and Duncan analysis indicated that this difference was largely due to the fact that samples from the runner hosted significantly fewer OTUs than the inflorescence and root samples (Fig.3.2). Mapping OTU richness across the plant body also showed that bacterial diversity varied across individual samples from the plant (Fig.3.3).    Figure 3.2 Comparison of the average OTU richness in samples collected from different parts of the plant body. Results showed that there was a significant difference in OUT richness between body parts (p=0.04, F=2.56) and Duncan analysis indicated that samples from the runner hosted significantly fewer OTUs than the inflorescence and root samples (results from the Duncan test are listed as letters in the graph, where different letters indicate significantly different results at a 0.95 confidence level).  aabababab050100150200250300Average total number of OTUsPlant body part47   Figure 3.3 Variations in OTU richness across the body of the P. aurantiaca plant. Samples collected are indicated by dividing lines, and 300 sequences were sub-sampled from each sample in order to be able to compare as many samples as possible while controlling for sampling effort. Samples that failed to amplify bacterial sequences are coloured white, and the rest range on a colour scale from blue (~ 50 OTUs) to red (~ 240 OTUs) showing how OTU richness varies across the plant body.  48     Figure 3.4 Principal coordinates analysis of un-weighted (a) and weighted (b) UniFrac dissimilarities between samples from different parts of the plant body. Each dot represents a sample, and samples are labelled based on where on the plant body they were collected (inflorescence, leaves, runners, bulb, roots or stem of the plant). Samples far apart from each other are considered to host bacterial communities less related to each other, and results showed that each compartment of the plant hosted distinct bacterial communities more similar to each other than any other samples from the plant body. These trends were also confirmed in PERMANOVA results comparing the communities through both un-weighted (Pseudo-F=1.5645, p=0.0001) and weighted (Pseudo-F=3.3629, p=0.0001) UniFrac distances. 49   Figure 3.5 Comparison of the average proportion of sequences belonging to different bacterial orders, among samples taken from different parts of the plant body. Sequences belonging to orders that made up less than 1% of the total community have been grouped as “Other” in order to get a better overview of general patterns in community composition. The graph illustrates the differences in bacterial community composition found within a single plant body, a trend that was confirmed in the PERMANOVA results showing a significant difference between body parts (p=0.0001).4% 3% 2% 2% 4% 2%9% 10% 11%19%21%12%3% 2% 2%4%2% 2%2%2%16%8% 8%18%11%14%5%13% 8%2%10%10%11%11% 8%6%2%12%24% 7%4%2%2%2%10%9%9%11%10%5%2%3%2%3%2%2%2%2%3%2%3%2% 2%3%2%3%3%8%36%2%5%3%5% 3%3%2%2%16% 16%11% 11%21%9%0%10%20%30%40%50%60%70%80%90%100%Inflorescence Stem Leaves Bulb Roots RunnerOtherTM7-3_EW055Gammaproteobacteria_XanthomonadalesGammaproteobacteria_PseudomonadalesGammaproteobacteria_EnterobacterialesGammaproteobacteria_ChromatialesGammaproteobacteria_AlteromonadalesDeltaproteobacteria_MyxococcalesBetaproteobacteria_RhodocyclalesBetaproteobacteria_MethylophilalesBetaproteobacteria_BurkholderialesBetaproteobacteria_unknownAlphaproteobacteria_SphingomonadalesAlphaproteobacteria_RhizobialesAlphaproteobacteria_CaulobacteralesSphingobacteria_SphingobacterialesActinobacteria_SolirubrobacteralesActinobacteria_MC47Actinobacteria_ActinomycetalesActinobacteria_Acidimicrobiales50  Looking closer at how the bacterial communities were structured across the plant body, we found a significant distinction in β-diversity among bacterial communities from various body parts, when comparing both weighted (Pseudo-F=3.3629, p=0.0001) and un-weighted (Pseudo-F=1.5645, p=0.0001) UniFrac distances. This pattern could be seen in the PCoAs, where samples from the same body part clustered closer together (Fig.3.4), and when comparing the average bacterial community composition found in samples from different body parts (Fig.3.5). In general, there was a higher abundance of bacteria belonging to the order Sphingomonadales in samples from leaves (24%) and stems (12%) compared to other body parts, and root samples showed the highest proportions of Actinomycetales (21%) in their communities (Fig.3.5). Interestingly, samples from the runner of the plant differed in community composition from samples of the stem or the roots; the difference that stood out the most was that samples of the runner showed a strong dominance of bacteria belonging to the order Enterobacteriales (36%) in the community (Fig.3.5).  3.4 Discussion While the body of a plant might be a microbial island in the landscape, it is not a homogenous one. In fact, we found that samples from different parts of the plant body hosted distinct bacterial communities. This shows that bacterial community composition varies at small scales within the plant, and that plant morphological features help define micro-habitats within the ecosystem that is the plant body. However, though bacterial communities varied compositionally, there were few stark differences seen in richness across the plant body, and though samples from roots and the inflorescence hosted the richest bacterial communities, they were only significantly richer than samples from the runner.  Though most research on phyllosphere microbial communities has been focused on specific plant organs, or comparisons between plant species, recently, a few other publications have appeared in the literature, describing bacterial communities across individual plants, and distinguishing between different plant organs (Ottesen et al. 2013; Junker and Keller 2015; Leff et al. 2015). For example, Ottesen et al. (2013) presented a baseline survey of bacterial community composition in tomato plants, where they did 51  targeted sampling of different body parts of multiple plants, and concluded that differences seen within a plant body also were consistent across multiple individuals, suggesting that there are general trends for which bacteria that colonise leaves versus stems or roots etcetera. Our study supports this idea, as we saw a clear distinction in bacterial community composition between body parts (Fig.3.5), and though we sampled a completely different plant species, our results of a strong presence of Actinomycetales in root samples and Sphingomonadales in stems and leaves was in accordance with community structures found in body parts of tomato plants (Ottesen et al. 2013). Interestingly, they recorded an increasing amount of the family Enterobacteriaceae in tomato fruits, and we found that the order Enterobacteriales was surprisingly dominant in the reproductive runners of our plant. This could indicate that these bacteria may be particularly adapted for colonising newly produced tissues in the plant body, or that they are trying to secure dispersal into the new plant bodies that will develop from these vegetative and sexually reproductive structures.  Another study by Bodenhausen et al. (2013) compared leaves and roots of multiple individuals of Arabidopsis thaliana, and found that roots and leaves hosted distinct bacterial communities, but that richness, diversity and evenness was similar in the endophytic compartments of the structures. Epiphytic bacterial communities on the other hand, were richer, more diverse and had a higher evenness in roots than in leaves. Because we sampled a combination of the endophytic and epiphytic compartment of plant tissues, we were not able to make this distinction, but our results showed that richness did not vary significantly between leaf and root samples, suggesting that we may have better captured the endophytic part of the root microbiota. A recent study by Junker and Keller (2015) (which extensively sampled different parts of the phyllosphere), however, did not distinguish between epiphytic and endophytic communities, but showed that microhabitat filtering within the plant body had a strong effect on bacterial community composition and richness that even exceeded the influence of environmental effects such as precipitation, altitude, substrate age and geographic distance (Junker and Keller 2015). With an increasing number of studies examining microbial communities within the plant body in close detail, and a constant evolution of better tools to map the distribution of these organisms and their diversity, we are continuously learning more about their life history and impact on their plant host. However, further examination of a larger set of plants, across 52  a wider range of plant species, and with more detailed sampling across the plant body would be useful in order to be able to draw conclusions about general patterns in community composition specific to body parts of plants and how they might be interrelated. For example, it would be interesting to see if microbial communities in floral structures are influenced by pollinator visitations (Aleklett et al. 2014), and if seeds pick up microbial colonisers from the inflorescence prior to dispersal, as it could be a potential mechanism for microbial inheritance through vertical transmission (Truyens et al. 2014). Ultimately, understanding the ecosystem that is the plant body and how habitats within the plant body select for distinct microbial communities is important in order to secure plant resilience against phytopathogens, something that can be applied in both agricultural and conservation practices.    53  Chapter 4: Timing of soil exposure affects community composition in the root microbiota of mature plants.  4.1 Background In the field of plant community ecology, both the timing and the order of species arrival are thought to be important factors in community formation (van de Voorde et al. 2011; Kardol et al. 2013; Leopold et al. 2015).  Whereas some of these community assembly concepts have been discussed in terms of insect colonisation of plants (e.g. Janzen, 1968), the methodology to examine microbial community assembly has only recently arisen, and despite their importance to plant health (Berendsen et al. 2012) and productivity (Clay 2001; Van Der Heijden et al. 2008), considerably less is known about what affects community formation of microbial symbionts associated with plant hosts (Nemergut et al. 2013; Bulgarelli et al. 2013; Johnson 2015). Studies investigating the formation of the human gut microbiota have identified a critical ‘window of opportunity’ for microbial colonisation (Ley et al. 2006; Sekirov et al. 2010), and it is hypothesised that the timing of colonisation may play a role in the development of host immunity (Kelly et al. 2007; Hansen et al. 2013). Like animal guts, plant roots are colonised by a diverse array of bacterial and fungal communities (Buée et al. 2009), which are thought to be closely associated with plant immunity and health  (Berendsen et al. 2012; Ramírez-Puebla et al. 2013). There is also evidence that the root microbiota varies compositionally between plants of different ages (Micallef et al. 2009) and developmental stages (Mougel et al. 2006; Houlden et al. 2008; Lundberg et al. 2012; Chaparro et al. 2013), but few studies have assessed how receptive plants are to microbial colonisers at different stages of development.  Here, we explored the hypothesis that a critical ‘window of opportunity’ exists for microbial colonisation of roots in terms of the historical contingency (timing and order of arrival) of both timing and order of microbial arrival.    Microbial dispersal to the root may affect community assembly in the root microbiota through variation in the order of microbial exposure between plants. For example, the first microbes encountered by a radicle may gain residency opportunistically, and preclude subsequent colonisers. This phenomenon, where early-arriving species in the community are thought to have a colonising advantage over late-coming species establishing in the 54  community, is known as priority effects (Drake 1991; Chase 2003). Priority effects have previously been documented in microbial communities of both wood decaying fungi (Fukami et al. 2010) and nectar yeasts (Peay et al. 2012), as well as in mycorrhizal fungi – colonising plant roots (Kennedy et al. 2009; Werner and Kiers 2014). In order to study priority effects, paired species of colonising microbes are commonly examined, but the role of priority effects in the context of whole microbial communities, where cohorts of microbes of different species may enter the community simultaneously, is not clear.  There is also reason to believe that host developmental stage may affect the recruitment of microbes to the rhizosphere: root systems change over plant development from the first fine radicles that emerge from the seed to far-branching networks of variously sized roots (Fitter 2002). Roots also change chemically with age, releasing fluctuating amounts of carbohydrates and amino acids over time (Badri and Vivanco 2009; Chaparro et al. 2013). It is not known whether these changes constitute a ‘window’ for colonisation by root microbes, but given the dramatic changes in plant development, plants may not be equally receptive to microbial colonisation throughout their lives. For example, the lack of root hairs and increasing amount of hardened surfaces of older roots may create colonisation barriers for certain microbial taxa (Watt et al. 2006) in the older parts of the root system. This would lead to a diversification of the root habitat in older plants, and potentially prevent them from hosting the same microbial community as a young plant throughout their root system.   The goal of this study was to test whether the timing of microbial exposure affected the resulting mature root microbiota. We predicted that:  If the order of microbial exposure is important, then the microbes able to colonise the root system at an early developmental stage should persist in the community at later developmental stages, regardless of subsequent exposure to microbes.  If plant developmental stage is important for determining the colonisation success of microbes entering the root microbiota, then plants exposed to microbes at different developmental stages should form distinct microbial communities.  55   Figure 4.1 This image shows the seed head of the species Setaria viridis which was used as a host plant in the experiment. The plants used in the experiment were all of the same genotype, minimizing genetic variation between individual plants.  Figure 4.2 Schematic overview of the experimental design. At each harvest, root samples from five plants from each age treatment were collected. Harvest 1 was collected to control for community composition prior to inoculation and included plants of different ages (9, 8, 7, 1 and 0 weeks old) representing different developmental stages (flowering, budding, non-reproducing mature plant, seedling, seed) in the life cycle of S. viridis. Harvest 2 was collected two weeks after inoculation to control for length of exposure to the inoculum, and Harvest 3 was collected to control for age of plant. Harvest 3 occurred when plants had all reached maturity (12 weeks), but before senescence. The age of the plants at the time of harvest is listed in the diagram.  56  4.2 Materials and Methods 4.2.1 Host plant and experimental setup Setaria viridis (L.) P.Beauv. is a drought-tolerant, C4 grass with a short lifecycle (Brutnell et al. 2010; Bennetzen et al. 2012), making it an ideal host plant for the purposes of this study (Fig.4.1). Clonal seeds of S. viridis (accession A 10.1) were donated by the Brutnell Lab at the Boyce Thompson Institute for Plant Research, New York, USA. Plants were grown under semi-sterile conditions in Sure Roots plug trays (TO38SR, Stuewe and Sons, INC.) containing rock wool plugs (Grodan A-OK 1.5 inch starter plugs), surrounded by Turface (Turface, MVP). Plants were grown in a growth chamber (Conviron CMP6010) with 12h/12h (day/night) photoperiod, 28oC/22oC (day/night). Watering was performed daily with autoclaved de-ionised water and fertilisation occurred once per week with 2.5 ml Technigro 17-5-24 plus (1:200 dilution).  Plants were established in cohorts based on developmental stage in order to be able to apply the soil inoculum at one point in time, avoiding storage effects on the inoculum. Therefore, the only difference among the groups was the age at which they received microbial inoculum (for a more detailed description of the planting and harvesting setup see Fig.4.2).  At the time of inoculation, plants represented different life stages (seeds, seedlings, adult plant, budding plant, flowering plant) and ages (0, 1, 7, 8 and 9 weeks old).  Before inoculation, one third of the plants from all age categories were harvested to determine root microbial communities before inoculation (Harvest 1) (for a detailed summary of how many samples were acquired and retained from each harvest see Table A.1). Remaining plants were inoculated with a soil-slurry collected from the rhizosphere of a single wild specimen of S. viridis growing at University of British Columbia Okanagan campus (49.939975N, -119.399264W). Briefly, collected soil was mixed with 250 ml autoclaved de-ionised water and sieved through a 2-mm mesh before 1 ml was applied to the rock wool plugs of the plants.  Two weeks after inoculation, half of the remaining plants were harvested to control for length of exposure to soil microbes (Harvest 2). Roots were collected for DNA extraction (described below). The remaining plants were grown until they reached 12 weeks (occurring from 3-12 weeks after inoculation) (Harvest 3) after which they were harvested and sampled 57  for root communities and aboveground biomass. By 12 weeks, all plants were mature plants that had not yet started to senesce. In total, 5 samples were processed for sequencing from each age class, at each harvest, given that enough samples had survived (for survival rates see Table A.2.).    4.2.2 DNA extraction and amplification Whole plants were destructively harvested and roots were rinsed with autoclaved, de-ionised water and dried with sterile filter paper. A subset of 0.1 g root tissue (wet weight) was sampled from each plant from inside the rock wool plug (controlling for immediate root environment), representing a subset of roots of various ages. Roots were frozen with liquid nitrogen, and then manually crushed using a sterile pestle. Attempts were made to extract DNA from the seeds used in the experiment (both as individual seeds and as a mix of 10 seeds), but without successful amplification of either bacteria or fungi. Samples were kept at -80oC until DNA extraction (DNeasy mini plant) (QIAGEN) according to the manufacturers protocols.  Bacteria: The V5 and V6 region of the 16S SSU rRNA gene was amplified to characterise the bacterial community, using the forward primer 799f (Chelius and Triplett 2001), and the reverse ‘universal’ bacterial primer 1115r (Reysenbach and Pace 1995). Each sample was also labelled with a unique 10 base pair identification barcode associated with the forward primer, as well as a 4 bp TCAG key, and a 21 bp adapter for 454 sequencing. The forward primer used in our study (799f) was designed to exclude chloroplast DNA and give a mitochondrial product approximately 1.5 times the size of the bacterial product (Chelius and Triplett 2001). Each Polymerase Chain Reaction (PCR) consisted of 25 µl (14.85 µl H2O, 5 µl GoTaq Flexi  buffer, 1 µl BSA, 2 µl MgCl2, 0.5 µl dNTP, 0.2 µl forward primer, 0.2 µl reverse primer, 0.25 µl GoTaq Flexi, 1 µl DNA template). Samples were initially denatured at 95 °C for 5 min and then amplified by using 30 cycles of 95 °C for 1 min, 61°C for 1 min, and 72°C for 1 min with a final extension of 7 min at 72 °C.  Fungi: The ITS2 region was amplified for fungal identification using the primers fITS7 and ITS4 (Ihrmark et al. 2012). Each sample was also labelled with a unique 10 base pair identification barcode associated with the forward primer, as well as a 4 bp TCAG key, and a 21 bp adapter for 454 sequencing. PCR was carried out where each sample consisted of 25 µl 58  (12.75 µl H2O, 5 µl GoTaq Flexi  buffer, 1 µl BSA, 3.5 µl MgCl2, 0.5 µl dNTP, 0.5 µl forward primer, 0.5 µl reverse primer, 0.25 µl GoTaq Flexi, and 1 µl DNA template). Samples were initially denatured at 94 °C for 5 min and then amplified by using 34 cycles of 94 °C for 45s, 61°C for 45s, and 72°C for 1 min. A final extension of 7 min at 72 °C was added at the end of the program to ensure complete amplification of the target region. All samples were amplified in triplicate. Negative controls (no-template) were included in all steps of the process to check for primer or sample DNA contamination. Samples were sent to the Laboratory for Advanced Genome Analysis (LAGA) at the Vancouver Prostate Centre (University of British Columbia, Vancouver) for purification and 454 sequencing using the GS-FLX Titanium sequencing platform, emulsion PCR and Lib-L chemistry for uni-directional sequencing (Roche, Branford, CT, USA).   Sequences are available at: http://dx.doi.org/106084/m9.figshare.1420638   4.2.3 Sequence analysis and statistics Sequence data was processed through the QIIME pipeline (Caporaso et al. 2010b). Operational Taxonomic Units (OTUs) were defined at 97% sequence similarity. Bacterial sequences were aligned with PyNast (Caporaso et al. 2010a) against the Greengenes database (13_8 database) (DeSantis et al. 2006) using the rdp classifier (Wang et al. 2007) and chimeric sequences were removed with ChimeraSlayer (Haas et al. 2011). Fungal sequences were identified using UNITE (12_11_otus database) (Abarenkov et al. 2010; Kõljalg et al. 2013) and the rdp classifier (Wang et al. 2007). In order to adjust for unequal numbers of sequences among samples, the bacterial dataset was rarefied at 2000 sequences/sample and the fungal data set was rarefied at 1000 sequences/sample before further analysis of α- and ß-diversity. Because the effects of rarefying have recently been debated (McMurdie and Holmes 2014), and rarefying reduced the number of samples available for analysis (Table A.1.), we confirmed the results of our ß-diversity PERMANOVA analysis by performing the same statistical procedures on non-rarefied data that was independently filtered and transformed using the regularised log (rlog) transformation on the log2 scale using DESeq2 (Love et al. 2014). The rlog-transformation is recommended if sequence depth varies widely between samples (Love et al. 2014).  59  Species richness (α-diversity) was analysed for both bacterial and fungal data using the ‘phyloseq’ (McMurdie and Holmes 2013) and ‘vegan’ (Oksanen et al. 2014) packages in R.  Observed species richness (Observed) in each sample was compared using t-tests or a Wilcoxon test (depending on whether the data was normally distributed), and ANOVAs (Type III SS). Singletons were retained for analysis of α-diversity, but removed from the data set before analysis of ß-diversity. Generally it is recommended that singletons are retained in richness estimation, as many estimators use singletons and doubletons to model species accumulation (e.g. Chao et al., 2005). However, due to concerns that next-generation sequencing of fungi produces many singletons that might be largely artifactual (Tedersoo et al. 2010; Smith and Peay 2014), we confirmed that the results obtained displayed the same general pattern with singletons removed. Dissimilarities in bacterial and fungal communities between samples were calculated using the Bray-Curtis (Bray and Curtis 1957) measure for the rarefied dataset and Euclidian distances for the rlog-transformed dataset. As the rarefied datasets are comprised of integers, they are transformed into distance matrices using dissimilarity indices. The transformed datasets are continuous variables, which have already been converted into a matrix of numerical values that preserve the patterns of variance within the dataset.  Thus, for the rarefied data, we constructed distance matrices using ß-diversity metrics, and for the transformed data we constructed distance matrices using Euclidean distance metrics.  We tested for significant differences in ß-diversity between treatments using 1-way and 2-way PERMANOVA (Type III SS) (9999 permutations) tests to account for unequal sample sizes (Anderson 2005). Pair-wise permutational comparisons between age-classes were performed for samples, comparing the effect of timing of inoculation and the percentage. Scatter plots (2-D) of Principal Coordinates Analysis (PCoA) were generated in PRIMER-E (Clarke and Gorley 2006) and used to visualise the greatest amount of variability in the pair-wise dissimilarities between samples.  Significant differences in Bray-Curtis dissimilarities between clusters observed in the PCoAs were further compared by a similarity percentage analysis (SIMPER) at the order level (using PRIMER-E), identifying which bacterial and fungal orders contributed most to the variation seen between samples. Treatments with fewer than 3 samples were included in figures but excluded from statistical analyses. 60  4.3 Results 4.3.1 Are plants able to take up new microbial associations throughout their lives?   Before and after inoculation (Harvest 1 compared to Harvest 2) Bacteria  Bacterial root communities changed significantly for plants of all ages when introduced to soil microbes. Even though a ‘background’ community was already established at the time of inoculation (Harvest 1), the introduction of soil microbes increased species richness for all time classes regardless of whether the data was examined with (p=0.0001, t=-4.29) or without (p=<0.0001, t=-4.76) singletons included (Fig.4.3) and changed the microbial community composition of the roots (p=0.0001 Pseudo-F=2.89) (Fig.4.4). This was observed when comparing samples harvested before inoculation (Harvest 1) to samples harvested two weeks after inoculation (Harvest 2). Analysis of the non-rarefied dataset confirmed the statistical difference between bacterial communities before and after inoculation (p=0.0001, Pseudo-F=2.13) (Table 4.1). A closer look at the community composition before and after inoculation showed that the most dominant bacterial order in un-inoculated roots was Actinomycetales (Fig.4.5), making up as much as 49% of the average community sampled across plants of different ages. Once plants were inoculated with the soil slurry, Burkholderiales became more prominent in the root system. In plant roots harvested two weeks after inoculation, Burkholderiales made up 38% of the total community (Fig.4.5). We saw a trend of Rhizobiales increasing post inoculation from 4% to 8% of the average bacterial community, and Sphingobacteriales, Saprospirales, Caulobacterales and Enterobacteriales also going from 0% to 1% of the average bacterial community post inoculation. The same trends of increasing amounts of Burkholderiales and Rhizobiales in the communities post inoculation was also observed when we compared only plants harvested at the same age (9 weeks) before and after inoculation (Fig.A.1.).  To see whether these trends were significant, we performed a SIMPER analysis to see which bacterial orders contributed the most to differences between root microbiota before and after inoculation. Results from the SIMPER analysis showed that overall, differences 61  between samples harvested before and after inoculation were mainly driven by differences in the orders Actinomycetales (31.65%), Burkholderiales (25.62%), Sphingomonadales (14.26%), Rhizobiales (6.25%), and Acidobacteriales (5.23%), which contributed the highest percentage (listed in brackets) to the average dissimilarity between groups.  Fungi  For fungal communities, soil inoculation had no significant effect on fungal community richness regardless of whether the data was analysed with (p=0.26 t=1.15) or without (p=0.21, W=138) inclusion of singletons (Fig.4.3), but altered the composition of the overall community significantly (Fig.4.4) (p=0.03, Pseudo-F=1.57). However, this compositional effect was not significant at the α = 0.05 level when considering the non-rarefied dataset (p=0.06, Pseudo-F=1.37) (Table 4.1). Compositionally, comparing the average fungal community in samples before and after inoculation, the biggest change was seen in the order Sporidibolales, which made up 43% of the community in un-inoculated plants, but only 28% post inoculation (Fig.4.5). In addition, an unknown order of Dothideomycetes increased from 0% to 6% post inoculation, as well as ones belonging to an unknown order of Sordariomycetes (from 0% to 4%). In general, the proportion of sequences in the average community belonging to unidentified fungal taxa or fungal orders making up less than 1% on their own (listed as “Other” in the figure) also increased post inoculation (Fig 4.5). When we compared only plants of the same age (9 weeks) harvested before and after inoculation, we saw that the amount of fungi belonging to the order Hypocreales increased post inoculation from 28% to 48% on average whereas the proportion of Sporidiobolales and unidentified fungi decreased (Fig.A.1.). Results from the SIMPER analysis showed that overall, differences between samples harvested before and after inoculation were mainly driven by the orders Sporidiobolales (28.9%), Hypocreales (28.1%), Unidentified fungi (16.7%) and Eurotiales (7.8%), which had the highest percentage contribution to the average dissimilarity between Harvest 1 and Harvest 2.  62  4.3.2 Does plant developmental stage at inoculation determine the composition of the root microbiota?   Comparison of plants harvested at the same age, but inoculated at different ages (Harvest 3). Bacteria We found no difference in species richness between plants inoculated at different developmental stages for 12-week-old plants (Harvest 3) (p= 0.59 F=0.66) (Fig.4.6), but there was a significant difference in community composition (p=0.008 Pseudo-F=1.45) (Fig.4.7). This compositional difference was confirmed by the analysis of the non-rarefied dataset (p=0.02, Pseudo-F=1.29) (Table 4.1). In general, plants inoculated at similar ages hosted bacterial communities that were more similar to each other (Fig.4.7), and pair-wise comparisons showed that this trend was driven primarily by the difference between plants inoculated as seeds (0 weeks) versus those inoculated at 7 weeks (p=0.007), 8 weeks (p=0.008) and 9 weeks (p=0.01). A difference between age groups was also seen in survival rates post inoculation, as a majority of plants inoculated at 1 week did not survive to maturity (Table A.2.). Plants inoculated at different developmental stages but exposed to the inoculum for the same length of time (Harvest 2), showed a significant difference in observed bacterial species richness (p=0.005 F=5.9), as well as a significant difference in community composition between plants inoculated at different ages (p=0.0002, Pseudo-F= 2.22). This trend was also confirmed in the analysis of the non-rarefied dataset (p=0.0001, Pseudo-F=2.26) (Table 4.1). Pair-wise comparisons of the Bray Curtis dissimilarities further revealed that the significant difference was primarily between plants inoculated far apart in age (0 weeks and 1 week compared to 7 (p=0.018), 8 (p=0.019) and 9 weeks (p=0.028)) whereas there was no significant difference between 0 weeks and 1 week or between 7, 8 and 9 weeks.  63  Table 4.1 Comparison of PERMANOVA results generated from a rarefied and log-transformed dataset (bacteria: 2000 sequences/sample, fungi: 1000 sequences/sample) and a non-rarefied dataset that was rlog-transformed. The data compares bacterial and fungal communities from plants harvested before inoculation (Harvest 1), plants inoculated at different ages but exposed to the inocula for the same length of time (Harvest 2), and plants inoculated at different ages but harvested at the same age (Harvest 3). Results show that for bacteria, results remain significant independently of whether the data set was rarefied or rlog transformed. For fungi, the rlog transformed data set no longer showed significant differences at the α=0.05 level between communities from plants exposed to soil at different ages but harvested at the same age (Timing of inoculation, Harvest 3) as well as between plants harvested before and after inoculation (Effect of inoculation, Harvest 1 v.s. Harvest 2).    Factor tested (in bold) and samples examined.  Bacteria Rarefied (log-transformed Bray Curtis dissimilarities)  Bacteria  (rlog-transformed  Euclidian distances) Fungi Rarefied (log-transformed  Bray Curtis dissimilarities) Fungi (rlog-transformed  Euclidian distances)  Pseudo-F p Pseudo-F p Pseudo-F                  p Pseudo-F p Timing of inoculation Harvest 2 Harvest 3         2.22 0.0002 2.26 0.0001 - - 0.97 0.59 1.45 0.008 1.29 0.02 1.38 0.03 1.17 0.06 Effect of inoculation Harvest 1 compared to Harvest 2         2.89 0.0001 2.13 0.0001 1.56 0.03 1.37 0.06 64   Figure 4.3 α-diversity measures of bacterial (a) and fungal (b) communities in samples harvested prior to (Harvest 1), or two weeks after (Harvest 2), inoculation with soil. Each dot represents a sample, and the variation among samples was calculated using observed species richness. Results from t-tests comparing bacterial and fungal richness pre and post inoculationshow that there was indeed a significant increase in bacterial (p=0.0001, t=-4.29) but not fungal (p=0.26 t=1.15) richness post inoculation.     65   Figure 4.4 Principal co-ordinates analysis (PCoA) plots of log-transformed Bray Curtis dissimilarities, showing compositional dissimilarity between samples, where two points closer together host more similar communities. These plots show bacterial (a) and fungal (b) communities in samples from plants harvested immediately prior to (Harvest 1), and two weeks after (Harvest 2), inoculation and indicate a shift in bacterial community composition where we see little overlap between samples harvested before and after inoclation. For bacterial communities, PERMANOVA results confirmed that there was a significant distinction between communities in plant roots harvested before and after inoculation (p=0.0001, Pseudo-F=2.89). Though less apparent in the PCoA, PERMANOVA results also revealed that there was a significant difference between fungal communities in plant roots harvested before and after inoculation (p=0.03, Pseudo-F=1.57). 66   Figure 4.5 Bar graphs showing the average relative abundance of sequences belonging to bacterial (a) and fungal (b) orders, compared between samples harvested before inoculation (Harvest 1), after inoculation (Harvest 2), and at 12 weeks (Harvest 3).  For Harvest 1 and Harvest 2, the bars show an average community composition based on a combination of all samples from that harvest. Among samples from Harvest 3, we separated out and compared the average community composition of plants inoculated at 0 versus 9 weeks. Orders representing less than 1% of the community have been grouped as “Other”. For bacterial communities we see that Actinomycetales made up a larger portion of the average community prior to soil inoculation whereas Burkholderiales became more dominant in the root microbiota after the introduction of soil. For fungal communities, we see less prominent changes in the community after soil inoculation, but that plants inoculated from seeds form very distinct fungal communities, heavily dominated by the order Xylariales. 67  A comparison of bacterial root microbiota in plants from Harvest 3 inoculated at the most distant time points (0 weeks and 9 weeks), showed that bacterial orders dominating in the average root microbiota of plants harvested pre-inoculation (Harvest 1), Actinomycetales and Sphingomonadales, also comprised a significant part of the root microbiota in plants inoculated at 9 weeks (Fig.4.5). In comparison, plants inoculated as seeds hosted communities with proportional abundances similar to the average bacterial community found in roots harvested two weeks post-inoculation, with a strong presence of Burkholderiales and Rhizobiales. Another observed difference was that Cytophagales comprised 4% of the average bacterial community for plants inoculated as seeds, but were not present in plants inoculated at 9 weeks or in samples harvested before and two weeks after inoculation (Fig.4.5). SIMPER analysis further revealed that the orders Burkholderiales (24.86%), Actinomycetales (22.5%), Rhizobiales (16.3%), Sphingomonadales (12.5%), and Cytophagales (6.0%) were the highest percentage contributors to the average dissimilarity between plants inoculated at 0 weeks and 9 weeks but harvested at the same age (12 weeks).  Fungi The effect of plant developmental stage was less pronounced for fungi compared to bacterial communities.  Like bacteria, there was no difference among fungal species richness for plants inoculated at different developmental stages but harvested at the same age (12 weeks) (p=0.26, F=1.55) (Fig.4.6). A further comparison of ß-diversity using Bray Curtis dissimilarities between roots inoculated at different developmental stages, but harvested at 12 weeks (Harvest 3) revealed that fungal communities were compositionally different (p=0.03 Pseudo-F=1.38) (Fig.4.7, Table 4.1). Pair-wise comparisons of plants inoculated at different ages, however, showed that this was primarily driven by differences between samples inoculated at week 0 and those inoculated at week 7 (p=0.10), week 8 (p=0.05) and week 9 (p=0.10).  Analysis of the non-rarefied dataset showed a similar, marginally significant difference between root systems inoculated at different developmental stages (p=0.06 Pseudo-F=1.17) (Table 4.1). For Harvest 2, only plants inoculated at 7, 8 and 9 weeks were retained in the analysis of fungal communities, as we were unable to successfully amplify fungi from the plants inoculated at weeks 0 and 1.  Further, only plants inoculated at 8 and 9 weeks contained 68  enough samples to be statistically compared post rarefaction.  There was no significant difference in observed species richness (p=0.43, t=0.82) between inoculation at week 8 or 9. There was also no significant difference in fungal community composition between these samples based on a pair-wise comparison of Bray Curtis dissimilarities (p=0.46 t=0.99), or when the data was analysed without rarefaction including more samples (Table 4.1) (p=0.59, Pseudo-F=0.97). When fungal community composition was compared in plants harvested at 12 weeks old (Harvest 3), plants that were inoculated at different developmental stages (0 and 9 weeks) hosted very different communities (Fig.4.5). Plants inoculated at 9 weeks hosted fungal communities similar to the average fungal community found in plants two weeks after inoculation, whereas plants inoculated as seeds (0 weeks) were dominated by Xylariales, an order that was not present in samples before or two weeks after inoculation. Another notable difference between plants inoculated at 0 weeks and 9 weeks was that Pleosporales, which comprised 2 % of the community before inoculation, represented a large portion of the community in plants inoculated at 9 weeks (32%), but less than 1% in plants inoculated at 0 weeks (Fig.4.5). SIMPER analysis confirmed that Xylariales (40.3%), Hypocreales (28.6%) and Pleosporales (15.6%) were the orders with the highest percentage contribution to the dissimilarity seen between samples inoculated at 0 weeks and 9 weeks.  4.4 Discussion 4.4.1 Order of microbial exposure  Our hypothesis that the original root microbiota would be resistant to change following subsequent exposure to a soil inoculum was not supported, as plants of all developmental stages were able to form new bacterial and fungal associations. However, we did see that plants inoculated later in life had formed different microbial communities than plants inoculated with soil from seeds. A comparison of the average community composition in plants indicated that plants inoculated at an older age may retain a larger portion of the ‘background’ community while gaining new community members from the inoculation, whereas plants inoculated from seeds host communities with proportions more similar to the average composition post inoculation (Fig.4.5). This trend suggests  that priority effects may 69  be in effect, since soil inoculation did not override the community established in older plants prior to inoculation.  However, to fully elucidate whether it actually was priority effects, we would need further insight into how quickly the communities formed, what the source of the different microbes in the community was and the ability to track specific microbes in a more detailed manor. Nonetheless it is interesting to consider what impact priority effects of entire microbial inocula could have in the root microbiota. Priority effects in the root microbiota could be due to microbial competition for root resources (Johnson 2015). For example, bacterial taxa have developed various competitive traits such as antimicrobial and fatty acid production that disperse competing bacteria, or glycoproteins that prevent root adherence by other bacteria (Hibbing et al. 2010). These competitive mechanisms could establish certain microbial taxa as better suited for acquiring and maintaining residence in the root, increasing the importance of order of arrival to the community. For example, Actinomycetales dominated un-inoculated plant roots and were able to persist in high numbers in the root system of plants inoculated late in life, but not plants inoculated from seeds (Fig.4.5). By contrast, the order Burkholderiales made up a smaller portion of the average root community before inoculation, but was found to dominate the community in plants inoculated from seeds. Endophytic actinomycetes are known to use the competitive strategy of producing antibiotics (Hardoim et al. 2015), while Burkholderiales have been described as scavengers that express antibiotic resistance (Dantas and Sommer 2012). This could explain why Burkholderiales were able to better establish and maintain dominance if they gained access early on, but not outcompete Actinomycetes if they had already established.  The overall effect of inoculation was less pronounced in fungal communities, but differences existed between fungi in plants harvested prior to inoculation compared to plants harvested two weeks after inoculation, suggesting that plants were receptive to new colonisation. The weaker response to inoculation seen in fungal communities could be caused by the slower growth of fungi compared to bacteria. Due to the nature of the experiment, fungi would be mainly introduced to the root system through spores, thus a lag time between inoculation and root colonisation would be expected (Swinnen et al. 2004). Bacteria, on the other hand, could arrive fully developed in the soil slurry and ready to colonise the root. Our study is limited to only knowing what community structure looked like  70   Figure 4.6 α-diversity measures of bacterial (a) and fungal (b) communities in samples from plants that were exposed to the soil inoculum at different developmental stages (Age_at_inoculation) but harvested when at 12 weeks (Harvest 3). Each dot represents a sample, and the variation among samples is calculated using observed species richness. ANOVA results comparing the different treatments (excluding treatments with less than 3 samples)confirmed that there was no significant effect of timing of inoculation on bacterial (p= 0.59 F=0.66) or fungal (p=0.26, F=1.55) community richness. 71   Figure 4.7 Principal co-ordinates analysis (PCoA) plots of  Bray Curtis dissimilarities between (a) bacterial and (b) fungal communities in root samples harvested from plants at 12 weeks old. The plots show a trend of plants inoculated at the same age hosting more similar communities, with plants inoculated as seeds standing out as hosting the most different bacterial and fungal communities compared to plants inoculated at other ages. PERMANOVA results confirmed that plants inoculated at different developmental stages hosted distinct bacterial (p=0.008 Pseudo-F=1.45) and fungal communities (p=0.03, Pseudo-F= 1.38).   72  in the roots before and after inoculation. Further studies, using mock communities with known quantities of traceable fungi and bacteria as an inoculum, would provide further information on how the assemblage dynamics affect future community composition in the root microbiota.  It could be argued that the order of arrival of microbial communities to the root was influenced by successional dynamics within the microbial communities. For example, it has been shown that fungal succession occurs within a root over the growth stages of the plant (Yu et al. 2012). These successional patterns were explained by differences in functionality, where vegetative stages selected for fungi more dependent on root-derived energy sources, and later developmental stages hosted a larger proportion of known saprotrophic fungi (Yu et al. 2012).   In our study, succession may have been a factor in the dynamics of Xylariales, which was not present in plants harvested before, and two weeks after, inoculation, but dominated the fungal communities of plants inoculated from seeds and harvested at 12 weeks (Fig.4.5). Fungi belonging to Xylariales are known to maintain an endophytic lifestyle in plant tissues, and produce fungal metabolites (Cytochalasins) that have antibiotic effects (Verma et al. 2008). It has also been suggested that they could start decomposing cellulose and lignin when plant senescence begins (Davis et al. 2003). This would explain how they are able to establish in an already existing microbial community, and why we see a greater abundance of these fungi in older plants, harvested at 12 weeks (Fig.4.5).  4.4.2 Plant developmental stage at inoculation  We found support for our hypothesis that plant developmental stage at the time of inoculation was important for determining the composition of root microbiota, as plants exposed to the inoculum at different developmental stages formed distinct bacterial and fungal communities (Fig.4.4). It has previously been argued that the developmental stage of the plant host asserts strong selective forces on the composition of its root microbial community (Mougel et al. 2006; Houlden et al. 2008; Yu et al. 2012; Chaparro et al. 2013; Chaparro et al. 2014). Conversely, it has also been argued that abiotic factors (such as soil type, soil chemistry and climate) could determine which microbes persist in the root microbiota (Buyer et al. 1999; Heuer et al. 2002; Singh et al. 2007), suggesting that plants exposed to the same soil conditions for the same amount of time would host similar microbial communities. Results 73  from this study show that neither of these theories alone fully explains the variation in root microbiota among individual plants as not all plants of the same age hosted equivalent microbial communities, and not all plants exposed to the soil inoculum for the same amount of time hosted communities with similar composition. The idea that soil and plant factors could be jointly governing the composition of the root microbiota is not new (Berg and Smalla 2009), and previous studies agree that the variation seen cannot always be explained solely by one or the other (Marschner et al. 2001; Garbeva et al. 2008; Lundberg et al. 2012). What sets our experiment apart from earlier studies is that specifically timing of microbial exposure in association with plant developmental stages was tested, and results showed that not just what a root system is exposed to, but also at what developmental stage the exposure occurs matters for root microbial community development.  In our results, plants inoculated from seeds formed distinct communities of bacteria and fungi compared to plants inoculated at later developmental stages. Previous studies have presented contradicting ideas of the relative importance of the seed microbiome versus the microbial communities of the soil in which the plant germinates (Buyer et al. 1999; Nelson 2004). It has been argued that the seed microbiome constitutes a potential reservoir for root-colonising microbes that gives them early access to colonising the developing root system (Barret et al. 2014), while others have claimed that once a seed enters the soil, the root microbiota will mainly be recruited from the bulk soil (Normander and Prosser 2000). A study by Green et al. (2006) showed that bacterial seed communities had the ability to persist in root systems, but that their persistence was taxon-dependent and variable depending on which soil the seeds were planted in.  Based on our findings, plants inoculated with soil at the developmental stage of seed (0 weeks) developed the most distinct bacterial and fungal communities compared to plants inoculated later in life. This suggests that microbial exposure during seed germination may have lasting effects on the root microbiota, creating variation among individual plants of the same age.  Put another way, soils surrounding germinating seeds could be the primary source of priority effects in root microbial communities. However, our results are not detailed enough to conclude this and therefore additional experiments could help figure out whether priority effects are in fact in play in the root mcirobiota.  74  Another interesting observation in our study was that plants inoculated at 1-week old had a lower survival rate than plants inoculated at a later stage (Table A.2.). One possible explanation to this could be that 1-week old plants host a less stable microbial community (see results Chapter 5) and therefore were potentially more sensitive to opportunistic pathogens introduced with the soil inoculum. Further experimentation is needed in order to determine whether plant performance can be linked to microbial community composition at different developmental stages.  4.4.3 Relating back to natural systems Our study shows that plants of all developmental stages took up new microbial associations if introduced to a soil inoculum (Fig.4.6, Fig.4.7). However, in natural environments, few plants would germinate in the absence of soil; therefore, development of the microbiota would be driven by the local community present when the plant germinates. A more natural scenario to examine for future studies would be to evaluate the effect of exposing soil-reared plants to different soils throughout plant development.  Our study supports the idea that historical events (inheritance, seed dispersal, soil exposures and disturbances in the environment) may have a strong effect on microbial community assembly and dynamics (Aleklett and Hart 2013).  The fact that we were able to detect historical contingency for variation in root microbiota among plants in a short-lived annual plant suggests that historical contingency may be an important factor for root community assembly in natural systems. Perennial root systems may be more or less receptive to new colonisation than the plants in our study, depending on whether the microbes are able to over-winter in the root system, or if they experience seasonality in colonisation patterns (Smalla et al. 2001; Dumbrell et al. 2011). The stability of root microbial communities over time in perennial root systems is unexplored, but is an important component of stability, since most natural systems are dominated by perennials. In the present study, microbial exposure during early stages of development in particular determined the composition of the root microbiota in mature plants, enhancing even more the role of microbial exposure during seed dispersal, as well as the local community composition and environmental conditions of the spot where the plant germinates, for creating individual root microbiota.  75  4.4.4 Conclusions and future directions This study provides unique insight into root bacterial and fungal community dynamics at different developmental stages. Plants in this study had a broad window of opportunity for colonisation by new microbes, but responded differently to inoculation at different life stages with plants inoculated as seeds appearing more receptive to new colonisation than older plants with already developed root systems.. We also found that plants exposed to soil for a longer time acquired communities different from plants only exposed for a short amount of time, however, plants of different ages exposed to the soil for the same amount of time did not host equivalent microbial communities. This tells us that both plant age and length of soil exposure could influence community formation in the root microbiota. While we saw an effect of colonisation timing on community formation in the root microbiota, the mechanisms driving these differences need more study, and should include looking at things such as fluctuations in root exudation patterns, physical changes in root surface structures during development and specific differences between roots of different ages. Knowledge about how receptive plants are to new microbial colonisers throughout their lives could be beneficial within agriculture and horticulture where plants are commonly grown in aseptic conditions prior to field or greenhouse transplant, and therefore may be deprived of important microbial interactions early in life. No studies (to our knowledge) have explicitly tested the effect of withholding soil microbes on plant performance, but studies have shown that exposure to beneficial soil microbes can prime the plant immune system, and help them withstand attack from pathogens (Conrath et al. 2006; Zamioudis and Pieterse 2012). This is a field of research that needs further investigation in order to determine the effect microbial exposure in early plant development on plant productivity and immunity, and further studies using multiple host species and more long-lived plants will reveal whether the trends observed in our study hold true across a broader spectrum of plant hosts.      76  Chapter 5: Community stability in the root microbiota during early plant development   5.1 Background Plant root microbiota are key determinants of plant health – yet little is known about how stable these communities are, nor the factors that influence their stability. The first weeks of a plant’s life are crucial for developing a root system, and acquiring nutrients. It has also become clear in the last few years that exposure to microbes during this time may have an important role in shaping the microbiota of mature plants (Green et al. 2006; Truyens et al. 2014; Barret et al. 2014) with consequences for both host productivity and health (Berendsen et al. 2012; Mendes et al. 2013). Whether the early microbial community that establishes is ‘fixed’ or fluctuating has not been investigated, but is important for understanding the processes determining mature root community composition.  For human hosts, is has been suggested that microbial communities in the gut of an infant would progress from species poor and unstable to a species-rich, stable-state community in a healthy adult (Koenig et al. 2011; Spor et al. 2011; Lozupone et al. 2012). This would suggest that community invasibility is high at a young age. For plant hosts it has been shown that the root microbiota changes continuously as the plant grows older (Mougel et al. 2006; Houlden et al. 2008; Yu et al. 2012; Chaparro et al. 2014), but a few studies have also suggested that the community might reach a stable state after the first two weeks of development (Ibekwe and Grieve 2004; Edwards et al. 2015), and that root communities increase in complexity (number of species and interactions) in older plants (Gomes et al. 2001), suggesting increased stability.  However, there are few records of how bacterial and fungal richness changes during early plant development, and to our knowledge, there are no studies that have specifically tested how microbial community stability progresses throughout the plant’s life. Do plants, like humans, go from hosting species-poor and unstable, to species-rich and stable, microbial communities throughout development?  There are reasons to believe that community stability might change during early plant development. For example, there is rapid development of new habitat within the root system, 77  which would provide microbiota with a larger amount of root surfaces and a wider diversity of habitats to colonise in older plants. Especially, older plants might provide an increasing number of root tips and proliferation of root hairs (Fitter 2002), which are known to create specifically attractive microbial habitats within the root system (Marschner et al. 2011). This diversification of microbial habitats within the root system could affect community stability by increasing the potential for a larger variety of microbes to colonise the root system (Hart and Endo 1981), potentially increasing rhizosphere competence for new-coming microbes in the community.  Another factor that could influence microbial community stability during early plant development is the changing resource availability. For example, it is known that root exudation of sugars (a known energy source for root-dwelling microbes) diminishes as the plant grows older (Badri and Vivanco 2009), with visible differences in exudation during the first two weeks of growth (Chaparro et al. 2013). In addition, it has also been shown that exudation of phenolic acids, many of which have antimicrobial properties, (Cueva et al. 2010) increases after the first weeks of germination (Chaparro et al. 2013). It has also been shown that plants increase their secretion of defence-related proteins such as chitinases, glucanases, myrosinases later in development, specifically in association with flowering (De-la-Peña et al. 2010). This suggests that younger root systems would provide resource-rich environments with higher invasibility.  Besides physical and environmental properties affecting stability in the root microbiota, biotic properties such as species richness of the community may play a role in determining resistance and invasibility. Though little is known about changes in microbial richness in roots during early plant development, a few studies have suggested that complexity would increase with plant age (Gomes et al. 2001; Ibekwe and Grieve 2004), and stabilise within the first two weeks (Ibekwe and Grieve 2004; Edwards et al. 2015). In general, a community in an older plant that was more diverse would also be expected to be more stable, and therefore more resistant to invasion by new microbes (Elton 1958; Robinson et al. 1995).  78   Figure 5.1 Depiction of potential community development (with dots of different colours representing different microbial taxa) based on whether the initial community is resistant or sensitive to perturbation (indicated by the red box and the introduction of a new microbial community). In a resistant community, the progression of community development would be expected to continue in the same direction disregarding of whether the plants were exposed to a resident or exogenous soil  (A and B would host similar communities). In a sensitive community, the introduction of an exogenous soil would be expected to cause a change in community composition (C and D would host significantly different communities).  Given the rapid changes in the root system during early development, both in terms of root proliferation and microbial community establishment, we wanted to specifically test the resistance of root microbiota during early plant development, and investigate whether the community becomes more resistant to perturbance over time. As the composition of the root microbiota is thought to change continuously throughout plant development (Yu et al. 2012; Chaparro et al. 2014), rather than looking at stability by measuring resilience (assuming that the community will return to an original state) (Pimm 1984), we studied a form of resistance 79  (the degree to which microbial composition remained unchanged) (Grimm and Wissel 1997) in the community, by examining to what degree bacterial and fungal communities were able to maintain the trajectory of their community development when faced with a soil perturbation (Fig.5.1). We perturbed plants at three different stages of early development (seeds, one-week-old seedlings and two-week-old seedlings) by either re-introducing them to the resident soil they were already exposed to, or introducing them to an exogenous soil inoculum. We then examined how bacterial and fungal community development was affected by perturbation and introduction of an exogenous soil to the system.  Bacteria and fungi are crucial members of the root microbiota (Buée et al. 2009), but the differences in life histories and colonisation strategies between them are vast (Andrews 1992; Fierer et al. 2007; Denison and Kiers 2011), and it would be ignorant to presume that they exhibit similar stability. Though there is a large variety of life history strategies among fungi themselves (Andrews 1992), overall, they are often seen as K-strategists in comparison to bacterial colonisers (De Vries and Shade 2013), exhibiting slower growth rates (Six et al. 2006) and more specific requirements for successful establishment. In the root system, bacterial communities are generally thought to be more opportunistic (Hardoim et al. 2012), whereas many fungi take longer to establish (Klein et al. 1996) but have the ability to secure a more long-term nutrient exchange with the plant, for example through mycorrhiza (Smith and Read 2008). Based on these assumptions, it could be expected that fungal colonisers would be more resistant to change once established in the root systems, whereas bacterial communities would show quicker shifts in community composition, potentially decreasing their resistance to change and increasing their invasibility (De Vries and Shade 2013). We therefore thought it important to investigate responses to perturbation in both bacterial and fungal communities of the root microbiota, in order to get a more complete view of its stability. In sum, our main goals were to examine (1) if the root microbiota in plants at different stages of early development responded differently to soil perturbation, and (2) if bacterial and fungal communities in the root microbiota had similar responses to soil perturbation. Overall, we predicted that the root microbiota in plants perturbed later in development would be more resistant to change than communities associated with younger plants. We also predicted that bacterial communities would be more sensitive to soil perturbation than fungal communities. 80  5.2 Methods 5.2.1 Host plant and experimental setup  The plant Setaria viridis (L.) P.Beauv. is a fast growing plant with known genotypes (Brutnell et al. 2010; Bennetzen et al. 2012), making it an ideal host plant for the experiment. Clonal seeds of S. viridis (accession A 10.1) were provided by the Brutnell lab at the Boyce Thompson Institute for Plant Research, New York, USA. Seeds were planted in Ray Leach cone-tainer pots (SC10R, Stuewe and Sons, INC.) in rock wool plugs (A-OK 1.5 inch starter plugs, Grodan), surrounded by Turface (Turface, MVP). All plants were grown in a growth chamber (Conviron CMP6010) under the conditions 12h/12h (day/night) photoperiod, 28oC/22oC (day/night), watered daily with autoclaved de-ionised water, and fertilised once a week with 2.5 ml Technigro 17-5-24 plus in a 1:200 dilution (Technigro).  The experiment was set up to study the effect of perturbation, created by the addition of an exogenous soil on resident root microbial communities (for details of the experimental setup, see Table 5.1). All pots were initially inoculated with a resident soil at the time of planting the seeds. The plants were further introduced to an additional soil perturbation either from the start as seeds (A2), or after one (B2), or two weeks (C2) of germinating. These ages were chosen in order to capture differences in early plant development before the root microbiota is thought to reach a stable state (Ibekwe and Grieve 2004; Edwards et al. 2015). For each age class, half of the plants were re-inoculated with the resident soil (A1, B1, C1) and half of the plants were exposed to a new exogenous soil (A2, B2, C2).  The two soils used (resident and exogenous) were collected in geographically remote areas with different climate, ecosystems and plant communities, and showed clear differences in properties such as texture, particle size and soil organic matter content (For more detailed descriptions see Table B.1. and Table B.2.). Soil inocula were prepared using 250 ml autoclaved de-ionised water, and sieved through a 2 mm mesh before being applied to the rock wool plugs of the plants. Each plant received a total of 2.5 ml soil slurry at the time of planting, and an additional 2.5 ml slurry at the time of perturbation. 5.2.2 Harvest, DNA extraction, amplification and sequencing To control for length of exposure and let the perturbation take effect, all plants were grown for an additional 3 weeks post perturbation before harvest, creating three different age classes 81  among the harvested plants (Harvest 1: 3 weeks old, Harvest 2: 4 weeks old and Harvest 3: 5 weeks old). In order to assess the effect of type of perturbation (resident or exogenous) root communities harvested from plants of the same age class were compared, and root communities in plants only exposed to the resident soil (A1, B1, B2) were used as a baseline for expected community development. Among plants that were exposed to the exogenous soil (A2, B2, C2), plants in which we saw a divergence in microbial community development were classified as sensitive to invasion, whereas plants hosting communities that continued the same community development despite the introduction of an exogenous soil were classified as resistant (Fig.5.1). This allowed us to examine questions such as: are we able to change the direction of community development in the root microbiota by introducing an exogenous soil to the system, and is the root microbiota more or less resistant to invasion at different ages during early development? Roots were harvested from seven plants of each treatment (A1, A2, B1, B2, C1, C2), with a total of 14 plants per harvest. All roots were collected from inside of the rock wool plug in order to control for differences in the immediate environment surrounding the roots. Roots were further rinsed with autoclaved de-ionised water, and dried on sterile filter paper. A sub-sample of 0.10 g (wet weight) from the root system (representing a collection of roots of different ages) was collected from each plant, which was then used for DNA extraction. DNA extraction was carried out using the FastDNA Green SPIN Kit (MP Biomedicals) according to the manufacturer’s protocol, including a spin protocol with 1.40 mm ceramic spheres and one 6.35 mm ceramic sphere/tube, to enhance root tissue lysis. Targeted DNA amplification of the V5 and V6 region of the 16S SSU rRNA gene was performed to characterise bacterial communities using the primers 799f (Chelius and Triplett 2001), and the reverse ‘universal’ bacterial primer 1115r (Reysenbach & Pace, 1995). Fungal communities were characterised by extracting the second internal transcribed spacer (ITS2) region of the rRNA gene, using the primers fITS7 and ITS4 (Ihrmark et al. 2012).  Each DNA sample was further amplified in triplicate through Polymerase Chain Reactions (PCR) using the protocol described in Chapter 4, and negative controls (no-template) were included in all steps of the process. For identification, each sample was labelled with a unique 10 base pair (bp) barcode associated with the forward primer, as well as a 4 bp TCAG key, and a 21 bp adapter.  All samples were sent to the Laboratory for 82  Advanced Genome Analysis (LAGA) at the Vancouver Prostate Centre (University of British Columbia, Vancouver) for purification and 454 sequencing using the GS-FLX Titanium sequencing platform, emulsion PCR and Lib-L chemistry for uni-directional sequencing (Roche, Branford, CT, USA).  5.2.3 Bioinformatics and statistical analysis Sequence data was processed through the QIIME pipeline (Caporaso et al. 2010b). Operational Taxonomic Units (OTUs) were defined at 97% sequence similarity. Bacterial sequences were aligned with PyNast (Caporaso et al. 2010a) against the Greengenes database (13_8 database) (DeSantis et al. 2006) using the rdp classifier (Wang et al. 2007) and chimeric sequences were removed with ChimeraSlayer (Haas et al. 2011). Fungal sequences were identified using UNITE (12_11_otus database) (Abarenkov et al. 2010; Kõljalg et al. 2013) and the rdp classifier (Wang et al. 2007). The data sets were further rarefied at 1648 (bacteria) and 1267 (fungi) sequences per sample in order to adjust of unequal sampling.  This allowed us to retain all samples for further analysis. Species richness (α-diversity) was analysed for both bacterial and fungal data using the ‘phyloseq’ (McMurdie and Holmes 2013), ‘vegan’ (Oksanen et al., 2014) and ‘laercio’ (da Silva 2015) packages in R. Observed species richness (Observed) in each sample was compared between treatments using t-tests, ANOVAs (Type III SS) and a Duncan test. Singletons were retained for analysis of α-diversity, but removed from the data set before analysis of ß-diversity (for rationale, see methods Chapter 4).  Scatter plots (2-D) of Principal Coordinates Analysis (PCoA) were generated in PRIMER-E (Clarke and Gorley, 2006) visualising variability in pair-wise dissimilarities between samples. Dissimilarities in bacterial and fungal communities between samples were calculated using the Bray Curtis (Bray and Curtis 1957) measure, and tested for significant differences in ß-diversity between treatments using PERMANOVA (Type III SS) (9999 permutations) (Anderson 2005). Pair-wise permutational comparisons between treatments were performed for samples of the same age class. Compositional differences between samples perturbed with the resident or the exogenous soil were further compared by a similarity percentage analysis (SIMPER) at the order level, identifying which bacterial orders that contributed most to the variation seen between treatments. 83  5.3 Results 5.3.1 Perturbation effects Bacteria Pair-wise comparisons (Duncan test) between samples in the same age class but exposed to the resident or exogenous soil showed that there was a no significant difference in bacterial richness (A1-A2: p=0.95, B1-B2: p=0.98, C1-C2: p=0.10) between treatments for plants perturbed at any age (Fig.5.2). However, we noted that plants exposed to the exogenous soil (A2, B2, C2) tended to have a higher β-diversity among samples (Fig.5.2) and therefore ran a t-test comparing standard deviations of observed species richness. Results showed that plants perturbed with the exogenous soil did not have significantly larger standard deviation (at the α = 0.05 level) than what was recorded among samples (p=0.06, t=-2.59) from plants only exposed to the resident soil.  Table 5.1 Table showing an explanation of treatments and results of PERMANOVA analysis on log-transformed Bray-Curtis dissimilarities. Plants were either only inoculated with a resident soil (A1, B1, C1) or perturbed at different ages (A2, B2, C2). In order to determine the effect of the perturbation, pair-wise comparisons were made between perturbed and non-perturbed samples of the same age. Treatment name Stage at perturbation Soil perturbation Age at harvest Pair-wise comparison of Bray-Curtis dissimilarities (Bacteria) Pair-wise comparison of Bray-Curtis dissimilarities (Fungi) t-value P (perm) t-value P (perm) A1 Seed  Resident soil 3 weeks 1.12 0.098 1.10 0.194 A2 Seed Exogenous soil 3 weeks B1 one-week-old Resident soil 4 weeks 1.28 0.007 0.88 0.864 B2 one-week-old Exogenous soil 4 weeks C1 2-weeks-old Resident soil 5 weeks 1.02 0.347 1.05 0.288 C2 2-weeks-old Exogenous soil 5 weeks  84  PERMANOVA results showed that for bacterial communities, the two factors Harvest (comparing between harvest 1, 2, and 3) and Exposure (comparing between plants exposed to either only the resident or a combination of the resident and the exogenous soil) both had a significant effect on community composition (based on Bray Cutis dissimilarities). However, there was no significant interaction between the two factors (Table B.2.).   Further pair-wise PERMANOVA comparisons of differences in community composition (Bray Curtis dissimilarities) between plants harvested at the same age showed a significant difference between plants with different soil exposure when comparing plants that were perturbed at 1-week-old (treatment B1 and B2) (p=0.007, t=1.28) (Table 5.1) (Fig.5.3). However, plants perturbed either from start (A1, A2), or after two weeks of germination (C1, C2) did not show a significant divergence in bacterial community composition (A1-A2: p= 0.1, t=1.12; C1-C2: p=0.35, t=1.02) (Table 5.1) (Fig.5.3). Looking closer at community composition, we could see that overall, plants exposed to the exogenous soil (A2, B2, C2) hosted a larger proportion of bacteria belonging to the order Enterobacteriales, and a smaller proportion of the orders Xanthomonadales, Sphingomonadales and Bacilliales (Fig.B.1.). SIMPER analysis showed that the orders Actinomycetales (23%), Burkholderiales (18%), Xanthomonadales (16%), Enterobacteriales (15%) and Sphingomonadales (11%) contributed to 83% of the variation seen between plants with different soil exposure (Table 5.2). In samples perturbed with the exogenous soil at one week old (A2) we saw an on average larger proportion of Burkholderiales (27%), compared to plants of the same age only exposed to the resident soil (A1) (19%) (Fig.B.1.).  Fungi For fungal communities, pair-wise comparisons showed that there was no significant difference in fungal community richness between plants harvested at the same age but with different soil exposre (A1-A2: p=0.99, B1-B2: p=0.99, C1-C2: p=0.99) (Fig.5.2). PERMANOVA results comparing community composition through Bray Curtis dissimilarities showed a significant effect of Harvest but not soil exposure on community composition, and there was also no significant interaction between the two factors examined.  Pairwise comparisons of treatments further confirmed this results (A1-A2: p=0.19, t=1.10; B1-B2: p=0.86, t=0.88; C1-C2: p=0.28, t=1.05) (Fig.5.3) (Table 5.1) as there was no 85  significant difference between root communities in plants with different soil exposure perturbed at any age examined.  Examining the fungal community composition (Fig.B.2.), it was clear that the order Hypocreales was dominating the community throughout the experiment, and across all treatments. When we examined the community at a higher taxonomic resolution, we found that this dominance was created by a high presence of the genus Fusarium.   5.3.2 Differences between plants of different ages Bacteria Comparing bacterial richness among harvests, there was a clear trend of plants harvested at an older age (Harvest 2 and Harvest 3) hosting bacterial communities with lower species richness than plants harvested at three weeks old (Harvest 1) (p=<0.0001, F=29.97) (Fig.5.4). However, pair-wise comparisons showed that there was no significant difference (at the p<0.05 level) in species richness between samples from Harvest 2 and Harvest 3 (p=0.06). Plants harvested at the same age also hosted bacterial communities significantly more similar to each other than any other samples (p=0.0001 Pseudo-F=4.27), disregarding whether the plants harvested at the same time had been exposed to the exogenous soil or not. Pair-wise comparisons of the three harvests showed that all age classes (3, 4 and 5 weeks old) were hosting bacterial communities significantly different from each other (p=0.0001). Examining the average bacterial community composition in samples harvested at the three different ages, results further showed that older plants hosted an increasing proportion of bacteria belonging to the orders Burkholderiales, Sphingomonadales and Rhizobiales, while the proportion of bacteria belonging to the orders Xanthomonadales and Actinomycetales as well as the class TM7-3 decreased in abundance in older plants (Fig. B.2.). SIMPER analysis showed that overall, differences in the bacterial order Actinomycetales contributed most to the Bray Curtis dissimilarities seen between plants of different ages.  86   Figure 5.2 Comparison of bacterial (a) and fungal (b) richness (Observed species richness) between treatments. Each dot represents a sample, and for each age class (A, B, C) there are plants that were perturbed either with the resident soil (A1, B1, C1) or the exogenous soil (A2, B2, C2). This allows us to compare between plants that got perturbed as seeds (A), one-week-old seedlings (B) or 2-weeks-old seedlings (C). The plots show an overall decrease in bacterial richness over the course of the experiment, but no significant difference in richness between plants of the same age but with different soil exposure. For fungal communities, results show no significant difference between plants with different soil exposure. These observations were in accordance with Duncan test results.87    Figure 5.3 Principal Coordinates Analysis (PCoA) plots showing differences in bacterial (a) and fungal (b) community composition between treatments based on Bray Curtis dissimilarities. Each dot represents a sample, and for each age class (seeds (A), one-week-old seedlings (B) or 2-weeks-old seedlings (C)) there are plants that were exposed to either exclusively the resident soil (A1, B1, B2) or a combination of the resident and the exogenous soil (A2, B2, C2).  For bacteria, the plot shows a separation between communities in plants harvested at different ages, but no clear distinction between plants with different soil exposure harvested at the same age. For fungi, there is no clear separation between samples with different soil exposure of any age. These trends were confirmed in PERMANOVA comparisons of the Bray Curtis dissimilarities (Table).    88   Figure 5.4 Comparison of bacterial (a) and fungal (b) richness (Observed species richness) between harvests. Each dot represents a sample, and each harvest is a combination of 14 plants of the same age (3-weeks-old (Harest1), 4-weeks-old (Harvest 2), 5-weeks-old (Harvest 3), that were either exposed toexclusively the resident soil (7 plants) or a combination of the resident and  exogenous soil (7 plants). For bacteria we can see a clear decrease in richness between Harvest 1 and 2. This change was confirmed when results were compared through a Duncan test which indicated that there was a significant difference between samples from Harvest 1 and samples from Harvest 2 and 3. For fungi there was no significant change in richness between harvests.  89  Fungi When comparing differences in fungal richness between harvests, we found that there was an overall significant difference between communities in plants of different ages (p=0.05 F=3.24) (Fig.5.4), which was mainly driven by samples from Harvest 2 hosting significantly richer fungal communities than samples from Harvest 1 (p=0.04). We also found that plants harvested at the same time hosted fungal communities compositionally more similar to each other, than plants harvested at a different age (p=0.0001 Pseudo-F=1.87). Pair-wise comparisons of the different harvests showed that they were all significantly different from each other, but that the plants harvested furthest apart in age (Harvest 1 and Harvest 3) were the most dissimilar (p=0.0001). Examining the compositional differences between harvests showed that the fungal community was heavily dominated by the genus Fusarium, which, on average, made up 75.6% of the community in samples across all treatments and harvests.  5.4 Discussion 5.4.1 Microbial community stability and plant age While numerous studies show that the composition of the root microbiota readily changes throughout the life of the plant (Yu et al. 2012; Chaparro et al. 2014) the stability of these communities has not been studied. We examined community stability in the root microbiota during early development, to see if plants were equally resistant to perturbation during different stages of early development. We found that only plants perturbed after one week of germination showed a significant divergence in bacterial community composition based on soil exposure. In comparison, plants perturbed as seeds, or after two weeks of germination were resistant to perturbation with an exogenous soil, and hosted bacterial and fungal communities indifferent from plants perturbed with the resident soil. We predicted that the root microbiota of young plants would be less stable than the root microbiota of older more established plants, but found that seeds, though the youngest age category, were just as resistant to soil perturbation as our oldest age category (two-week-old seedlings). Previous studies have found that root communities might not be equally stable across plant age classes when examining fungal pathogen establishment (Hart and Endo 90  1981; Raftoyannis and Dick 2002), and application of beneficial rhizo-bacteria (Bashan 1986), but their results showed that younger plants are more susceptible to colonisation by introduced microbes. One reason why we did not see a divergence in bacterial or fungal community development when plants were perturbed as seeds could be that no roots were developed at this stage. Without a developed root system, there would have been limited resources and surfaces for new microbes to colonise, which could have decreased the invasibility of the already existing community. The role of seed exposure and the persistence of seed-associated microbiota in root colonisation remains fairly undetermined (Green et al. 2007; Truyens et al. 2014), but is an interesting avenue for future research. We saw the strongest treatment effect in plants perturbed as one-week-old seedlings. The rapid development of new roots in one-week-old seedlings may have greatly increased the potential for exogenous microbes to colonise root surfaces. At this stage, seedlings would also be expected to provide a large supply of root exudates in the form of sugars (Chaparro et al. 2013), providing an energy source to sustain larger quantities of microbes.  We predicted that older plants would host richer microbial communities, but among the three age classes harvested in this experiment, we instead saw a clear trend of decreasing bacterial richness in older plants. If plants progress from a species-rich to species-poor root microbiota, then younger plants may host resource-rich and diverse communities. According to Wallenstein and Hall’s model of microbial dynamics in relation to environmental stability and resource availability, this would mean that the microbiota in younger plants would also rapidly respond to environmental changes through shifts in community structure (Wallenstein and Hall 2012).  Our results of decreasing diversity in older plants contradict findings from a recent publication by Edwards et al. (2015), who suggest that the bacterial root microbiota of Asian rice (Oryza sativa sub spp. japonica) enters a stable state after two weeks of establishment in soil; they reported no difference in richness in the rhizoplane or endorhizosphere between younger and older plants sampled within 13 days of when soil was introduced. Though these differences could be attributed to the fact that their time series began with germinated, sterile seedlings, the clear differences in results still implies that there might be additional variability among plant species, even though both our model plant (S. viridis) and theirs (O. sativa) belongs to the Poaceae family. This suggests a need for 91  more studies examining community dynamics in the root microbiota during plant development in order to establish what factors determine its stability.  5.4.2 Differences between responses in bacterial and fungal communities We predicted that bacterial communities would be more affected by soil perturbation than fungal communities. Overall, our results confirmed this prediction, as fungal communities were more resistant to perturbation than bacterial communities. The different response to perturbation in bacteria and fungi could be due to differences in dispersal and colonisation strategies between microbes. In our study the majority of bacteria would be easily dispersed into the system through the addition of the exogenous soil inoculum, whereas fungal dispersal may have been hindered because mycelial growth from already established hyphal networks was not facilitated. This may have made it easier for bacteria than fungi to disperse into the system and colonise roots. In addition, while bacteria have been shown to effectively colonise root systems within 24 hours of soil introduction (Edwards et al. 2015), fungi generally require more time to germinate from spores and extend hyphae in order to colonise roots (Ishida et al. 2008); therefore, it is possible that three weeks might not have been enough time to detect a measurable response to perturbation.  In our study, another difference we saw between bacteria and fungi was that fungal species richness varied little across all plants, compared to bacterial communities, which changed drastically. This may be because, at an early stage, fungal communities were dominated by a single fungal genus, Fusarium, which then retained dominance in the roots throughout all of the different ages and perturbation timings examined. It has previously been documented that root colonisation by mycorrhizal fungi (Kennedy et al. 2009; Werner and Kiers 2015) and dominance in fungal wood decomposer communities (Fukami et al. 2010) is affected by historical events, with early colonisers gaining an advantage in colonising root surfaces. In contrast, bacterial communities showed clear fluctuations in community composition, both as a response to perturbation, and in association with the age at which plants were harvested. These results reinforce the idea of bacterial taxa being more opportunistic and transient members of the root microbiota, as well as the importance of studying a broad set of microbial groups in order to get a holistic view of dynamics in the root microbiota, and the factors affect its composition. 92  5.4.3 Application and future directions This study shows that stability in the root microbiota varies over short time spans during early plant development, and supports the idea that there are stages in plant development when the plants are less resistant to perturbations and disturbance. Understanding how microbial communities are formed and the extent to which they are able to be shaped and manipulated could be of great importance in managed ecosystems. For example, because stability in the root microbiota fluctuated during early development, manipulation of root microbiota composition and structure might be easier to achieve if treatments are applied when plants are one-week-old (in the case of S. viridis). However, it also suggests that one-week-old plants might be extra sensitive to pathogen establishment and disturbances in the root microbiota.  Our findings contribute to the idea that events during a plant’s life, such as soil perturbation, has the potential to increase individual variation in root microbiota within a plant community (Aleklett and Hart 2013) by altering the direction of community development. However, in order to further explore stability in the root microbiota, we suggest experimental setups using mock communities, or traceable amounts of known root colonisers to disentangle colonisation dynamics in the root microbiota. It would also be useful to determine whether certain microbial taxa can help make the root microbiota more or less stable, as it has been suggested that microbial life history strategy could affect the stability of the community (De Vries and Shade 2013).   As our results differ from what has been shown in terms of root microbiota stability and age associated richness in crop plants such as lettuce and rice (Ibekwe and Grieve 2004; Edwards et al. 2015), it would also be useful to examine a larger variety of plant hosts in order to establish if there is variation between plant species in terms of root microbiota stability or if general trends can be established. For example, perennial root systems might function in a completely different manner, and be affected by seasonal fluctuations in microbial soil communities. In terms of land management and agricultural practices, it would also be interesting to investigate how the root microbiota responds to and recovers after more severe disturbances such as application of pesticides if introduced at different stages of plant development. Ultimately, understanding microbial community stability in plant hosts is crucial for establishing how to sensitive the root microbiota is to disturbances and in order to 93  determine how we best can manipulate its composition to benefit plant health and productivity.   94  Chapter 6: Conclusion  6.1 Main findings Microbes form some of the richest and most diverse communities on earth, and with high fine-scale diversity, variation in community composition can occur within centimetres in the environment (Grundmann 2004; Becker et al. 2006; Vos et al. 2013). Because microbial variation in the environment is so vast, chances of finding two individual plants hosting the exact same microbial community are little to none. In my work, I set out to determine just how much microbial community composition varies at the level of individual plants, and focused on identifying factors that could be driving this variation. In this thesis dissertation I present data both from the field - showing how much microbial community composition varies between and within plant species as well as across the plant body in a natural ecosystem, and manipulated growth chamber studies – illustrating that historical contingency (timing of a plant’s exposure to microbes during its life) has the possibility to significantly alter the development of its mature root microbiota. I further present evidence that plants of different ages are not equally sensitive to soil perturbations, and suggest that there might be times during early development when the plant is more likely to be affected by environmental perturbations. To conclude, I will revisit the objectives set up at the beginning of the thesis in order to draw broader conclusions around the findings of my experiments presented in this dissertation:  Objective 1: Determine how much individual variation in root microbiota there is within and between wild plant species. The findings of our field study show that individual plants host distinct microbiota in their root systems, and that plant species is a strong determinant of bacterial community composition. Among plants sampled in the field, individual plants growing within centimetres from each other, in the same soil, under the same environmental conditions showed stark differences in bacterial community composition. These differences were mainly based on species identity where plants of the same species hosted bacterial communities more similar than individuals belonging to different plant species. Comparing the amount of variation among individual plants within the species examined, we had 95  expected that P. aurantiaca, with its known low genetic variability, would show less individual variation in bacterial community composition. Instead, we found that the amount of individual variation was the highest among P. aurantiaca plants, comparing dispersion based on taxonomic differences. This shows that though plant genetics may structure the root microbiota at a plant species level, it does not necessarily explain individual variation among plants of the same species.  Objective 2: Determine how much microbial community composition varies within the body of an individual plant. By extracting DNA and sequencing 93 pieces of an entire plant, we were able to examine how bacterial communities were partitioned across the plant body. Our results reinforce the idea that bacterial communities are filtered by micro-habitats of different structures within the plant body, as we saw clear distinctions in bacterial community composition between samples taken from the roots, stem, leaves, inflorescence, bulb and runner of the plant. However, bacterial richness did not differ significantly between the majority of body parts, with the exception of the runner hosting significantly less diverse communities than the root or inflorescence samples. These results further suggest that morphological variation between individual plants could influence the composition of its microbiota, something that needs to be further examined.  Objective 3: Examine whether historical contingency of the root microbiota contributes towards creating individual variation among plants. By introducing plants to soil microbial communities at different developmental stages, we were able to examine how historical contingency (timing and order of arrival) affected the microbial community development in their root microbiota. Our findings show that plants are able to pick up new microbial associations across all stages of plant development. However, we also found that plants of the same age that had been exposed to the soil inoculums at different developmental stages formed distinct bacterial communities, indicating that timing of soil exposure in the plant life cycle matters for community development in the root microbiota.   Objective 4: Determine if microbial community stability is equal during all stages of early plant development. Our third study followed up on the results of our second study by 96  examining how plants already exposed to a resident soil inoculum from seeds responded to perturbation created by the introduction of an exogenous soil. Interestingly, we found that bacterial communities responded differently to soil perturbation based on the age at which plants were perturbed. Plants perturbed as seeds or after two weeks of germination were resistant to the introduction of an exogenous soil, whereas plants perturbed at one week showed a significant divergence in bacterial community composition based on whether the plants had been perturbed with the resident or exogenous soil. In fungal communities, however, we saw no effect of the introduced soil perturbation, reinforcing the findings in Chapter 4, which suggested that fungal communities were less affected by introduction of soil. These results indicate that plants may be sensitive to changes in bacterial exposure early in development, something that could be taken advantage of when attempting to manipulate the root microbiota. However, it also means that seedlings may be especially vulnerable to opportunistic pathogens.   6.2 Strengths and limitations of the dissertation work 6.2.1 Experimental design Studying microbial community assembly is challenging due to the amount of variation present in nature. One strength of the work presented in this thesis is the novelty of the experimental design presented in Chapter 3. Few have attempted to study timing of inoculation and priority effects of whole communities, likely because of the difficulty in isolating testable factors in such a complex system. Most manipulative work like mine involves single microbial inoculants, or very simple, artificial systems. By contrast, my approach provides a unique insight into how timing of microbial exposure could function in natural systems because I was able to control when plants became inoculated with soil microbial communities.  While there are many un-answered questions about how priority effects apply in more natural settings, the fact that we examined whole communities, and both bacteria and fungi, allowed us to get a broader understanding for what factors affect community formation for each of these groups. Our findings enhance the idea that not all microbes respond in the same way or at the same speed, which reinforces the need for more 97  studies examining a broad set of microbial groups in order to better understand dynamics in the root microbiota.  While providing a broader understanding for how whole microbial communities respond to historical events, the findings of this dissertation are at the same time limited in terms of being able to track specific microbial taxa, as the inoculation method used did not control for the exact quantities of microbes applied, nor the exact composition of each inoculation, as we did not use known strains as inoculants. In our opinion, our approach created a more natural inoculation scenario, since the soil inoculum provided the plants with a mixed microbial community collected from the rhizosphere of a wild plant specimen. However, future studies using mock communities of known quantities of specific microbes could be beneficial in order to better understand colonisation dynamics among specific microbial taxa, and to trace microbial colonisers from the source inoculum. Another aspect of the experimental design for Chapter 2, 4 and 5 worth discussing is the fact that we sampled and compared root tissues exclusively, rather than rhizosphere soil or a combination of roots and soil from the rhizosphere. This could be considered a strength as well as a limitation in terms of being able to properly depict the root microbiota and provide results comparable to other studies of root-associated bacterial and fungal communities. The practice of sampling microbial communities in plant root systems is not new, however, there has been little consensus as to what sampling methodology best captures the diversity of the root microbiota while selectively examining microbes closely associated with the plant (Hirsch and Mauchline 2012). It has been recognised that bacterial communities vary compositionally and in richness between the different compartments of the root system: the rhizosphere soil, the rhizoplane (surface of the roots) and the endorhizosphere (inside the root tissues) (Bulgarelli et al. 2012; Edwards et al. 2015), while many fungi are able to bridge all three compartments as well as provide additional habitat for bacterial colonisers within what is known as the mycorrhizosphere (De Boer et al. 2005; Hartmann et al. 2008).  Because of the variation present within the root system, some studies have chosen to sample all compartments separately (Bulgarelli et al. 2012; Lundberg et al. 2012; Edwards et al. 2015), while others have focused on sampling rhizosphere and sometimes bulk soil (Micallef et al. 2009; Peiffer et al. 2013; Tkacz et al. 2015). As our goal was to detect 98  changes in microbial communities associated with plant taxonomy, development and exposure, we opted to examine a combination of the rhizoplane and endorhizosphere, since microbial communities inside and at the surface of root tissues are more likely to be closely associated with, and regulated by the plant itself. For future studies however, it would be interesting to see if the same trends of priority effects documented in our studies also are visible in microbial communities in the rhizosphere soil of the root system.  6.2.2 Sequencing It has become obvious during the last few years that the methods for analysis of microbial communities through DNA extraction and sequencing are rapidly evolving, and new approaches for how to sequence DNA and analyse sequencing data are constantly being developed to maximise the accuracy with which we describe community composition and diversity. When the experiments presented in this thesis were designed, the sequencing method used (454 pyrosequencing) was still a very novel approach for studying microbial communities, and as part of my thesis work I even visited and apprenticed in the Fierer lab at the University of Colorado in order to learn how to process 454 pyrosequencing data through the QIIME pipeline.  However, within the last year, much of the work within the field of microbial ecology has switched over to using the Illumina platform. While Illumina is known to produce a larger amount of sequences for analysis, pyrosequencing data still provides a valid representation of microbial community composition. In fact, for bacteria, it has been shown that analysis of as few as 100 sequences of a bacterial community will still produce the same results in terms of detecting differences between samples and treatments (Kuczynski et al. 2010). For fungi, the suggested limit for how many sequences are needed to properly depict differences in community composition is thought to be a bit higher, around 400 sequences/sample, but tests analysing the same samples with 454 and Illumina techniques have still generated the same results in terms of differences in α- and β-diversity (Smith and Peay 2014).  Since sequencing methods are changing so rapidly, accommodating these shifts in approaches within a PhD thesis spanning over multiple years is a challenge, but as high throughput sequencing is a constantly evolving field of research, it is important to maintain 99  scepticism towards the methodology used and results presented, while integrating new knowledge about how to analyse and interpret our findings in the best way.  6.3 Future directions The field of plant microbiome research, studying microbial communities in plants as an extension of the plant body, has quickly developed over the last few years, and researchers as well as the general public are becoming increasingly interested in figuring out just how influential the microbes inhabiting the plant body are to the health and productivity of their hosts. In this thesis dissertation, I show just how intimately connected plants are to their microbial communities, addressing the idea of individual plant microbiota, and examining historical events that may influence the assembly and stability of the root microbiota.  For future studies, it would be fascinating to take the idea of examining early microbial exposure one step further by exploring the potential for microbial inheritance through vertical transmission of bacteria and fungi in seeds (Truyens et al. 2014). The idea of microbial inheritance from parent plant to seeds and seedlings is fascinating, as it would give plants a chance to provide their offspring with beneficial microbial associations and potentially increase their chance of survival. Microbial inheritance is also interesting to consider in the context of agriculture, where little is known about whether our current practices interfere with microbial inheritance, for example through sterilising of seeds. Another question that deserves further examination is what role seed dispersers play, not only in dispatching seeds from their parent plant, but also potentially in early microbial exposure, and inoculating the seeds with a microbial community with which to germinate (Theimer and Gehring 2007). Understanding these complex interactions in nature are important in order to successfully preserve and manage wild ecosystems, but could also be beneficial in agricultural practices as we are constantly looking for ways to enhance survival and productivity of crop plants. Therefore, resolving how to establish and maintain a ‘healthy microbiota’ in plants could be crucial in food production as well as forestry industry. An increasing amount of research should therefore be dedicated towards better establishing what a ‘healthy microbiota’ looks like for plants, as well as the mechanisms determining which microbes are allowed to, or choose to be a part of the plant microbiota. 100  Finally, the stark differences seen in root microbiota between plant species still need to be more explicitly explained. Even though it seems like plant genetics play a crucial role in structuring bacterial colonisation, it is still unclear which traits of the root system create this variation in bacterial community composition. Do differences root surface structure and architecture matter for microbial colonisation, or are differences in the microbial communities exclusively driven by exudation patterns? A lot still remains to be answered about the plant microbiota, and hopefully development of new methods and approaches will allow us to better understand the composition and dynamics of microbial communities in plants in the future.   101  Bibliography Abarenkov K, Nilsson RH, Larsson KH, et al (2010) The UNITE database for molecular identification of fungi - recent updates and future perspectives. New Phytol. 186:281–285. Agans R, Rigsbee L, Kenche H, et al (2011) Distal gut microbiota of adolescent children is different from that of adults. FEMS Microbiol Ecol 77:404–412. doi: 10.1111/j.1574-6941.2011.01120.x Aleklett K, Hart M (2013) The root microbiota-a fingerprint in the soil? Plant Soil 370:671–686. Aleklett K, Hart M, Shade A (2014) The microbial ecology of flowers: an emerging frontier in phyllosphere research. Botany 92:253–266. doi: 10.1139/cjb-2013-0166 Alford ER, Vivanco JM, Paschke MW (2009) The effects of flavonoid allelochemicals from knapweeds on legume - Rhizobia candidates for restoration. Restor Ecol 17:506–514. doi: 10.1111/j.1526-100X.2008.00405.x Allison SD, Martiny JBH (2008) Colloquium paper: resistance, resilience, and redundancy in microbial communities. Proc Natl Acad Sci U S A 105 Suppl :11512–9. doi: 10.1073/pnas.0801925105 Altschul SF, Gish W, Miller W, et al (1990) Basic local alignment search tool. J Mol Biol 215:403–410. doi: 10.1016/S0022-2836(05)80360-2 Anderson MJ (2005) PERMANOVA: a FORTRAN computer program for permutational multivariate analysis of variance. Department of Statistics, University of Auckland, Auckland, New Zealand Anderson MJ (2004) PERMDISP: a FORTRAN computer program for permutatinoal analysis of multivariate dispersions (for any two-factor ANOVA design) using permutation tests. Department of Statistics, University of Auckland, Auckland, New Zealand Anderson MJ, Walsh DCI (2013) PERMANOVA, ANOSIM, and the Mantel test in the face of heterogeneous dispersions: What null hypothesis are you testing? Ecol Monogr 83:557–574. doi: 10.1890/12-2010.1 Andrews J (1992) Fungal life-history strategies. In: Carroll C, Wicklow D (eds) The fungal community, 2nd edn. Marcel Dekker, New York, NY, USA, pp 119–145 Andrews JH, Harris RF (2000) The ecology and biogeography of microorganisms on plant surfaces. Annu Rev Phytopathol 38:145–180. 102  Aulakh MS, Wassmann R, Bueno C, et al (2001) Characterization of root exudates at different growth stages of ten rice (Oryza sativa L.) cultivars. Plant Biol 3:139–148. doi: 10.1055/s-2001-12905 Babich H, Stotzky G (1977) Sensitivity of Various Bacteria, Including Actinomycetes, and Fungi to Cadmium and the Influence of pH on Sensitivity. Appl Environ Microbiol 33:681–95. Badri D V., Vivanco JM (2009) Regulation and function of root exudates. Plant, Cell Environ 32:666–681. doi: 10.1111/j.1365-3040.2009.01926.x Bailey JK, Deckert R, Schweitzer JA, et al (2005) Host plant genetics affect hidden ecological players: links among Populus , condensed tannins, and fungal endophyte infection. Can J Bot 83:356–361. doi: 10.1139/b05-008 Bais HP, Park SW, Weir TL, et al (2004) How plants communicate using the underground information superhighway. Trends Plant Sci. 9:26–32. Bais HP, Weir TL, Perry LG, et al (2006) The role of root exudates in rhizosphere interactions with plants and other organisms. Annu Rev Plant Biol 57:233–266. doi: 10.1146/annurev.arplant.57.032905.105159 Bakker MG, Manter DK, Sheflin AM, et al (2012) Harnessing the rhizosphere microbiome through plant breeding and agricultural management. Plant Soil 360:1–13. Baldani J, Caruso L, Baldani V, et al (1997) Recent advances in BNF with non-legume plants. Soil Biol Biochem 29:911–922. Bardgett RD, Mommer L, De Vries FT (2014) Going underground: root traits as drivers of ecosystem processes. Trends Ecol Evol 29:692–699. doi: 10.1016/j.tree.2014.10.006 Barret M, Briand M, Bonneau S, et al (2014) Emergence shapes the structure of the seed-microbiota. Appl Environ Microbiolgy 81:1257–1266. doi: 10.1128/AEM.03722-14 Bashan Y (1986) Significance of timing and level of inoculation with rhizosphere bacteria on wheat plants. Soil Biol. Biochem. 18:297–301. Becker JM, Parkin T, Nakatsu CH, et al (2006) Bacterial activity, community structure, and centimeter-scale spatial heterogeneity in contaminated soil. Microb Ecol 51:220–231. doi: 10.1007/s00248-005-0002-9 Becklin KM, Hertweck KL, Jumpponen A (2012) Host Identity Impacts Rhizosphere Fungal Communities Associated with Three Alpine Plant Species. Microb Ecol 63:682–693. doi: 10.1007/s00248-011-9968-7 103  Belisle M, Peay KG, Fukami T (2012) Flowers as Islands: Spatial Distribution of Nectar-Inhabiting Microfungi among Plants of Mimulus aurantiacus, a Hummingbird-Pollinated Shrub. Microb Ecol 63:711–718. doi: 10.1007/s00248-011-9975-8 Bell C, Carrillo Y, Boot CM, et al (2014) Rhizosphere stoichiometry: Are C : N : P ratios of plants, soils, and enzymes conserved at the plant species-level? New Phytol 201:505–517. doi: 10.1111/nph.12531 Bennetzen JL, Schmutz J, Wang H, et al (2012) Reference genome sequence of the model plant Setaria. Nat. Biotechnol. 30:555–561. Benson AK, Kelly SA, Legge R, et al (2010) Individuality in gut microbiota composition is a complex polygenic trait shaped by multiple environmental and host genetic factors. Proc Natl Acad Sci U S A 107:18933–18938. doi: 10.1073/pnas.1007028107 Berendsen RL, Pieterse CMJ, Bakker P a HM (2012) The rhizosphere microbiome and plant health. Trends Plant Sci. 17:478–486. Berg G, Smalla K (2009) Plant species and soil type cooperatively shape the structure and function of microbial communities in the rhizosphere. FEMS Microbiol. Ecol. 68:1–13. Bodenhausen N, Horton MW, Bergelson J (2013) Bacterial Communities Associated with the Leaves and the Roots of Arabidopsis thaliana. PLoS One 8:e56329. doi: 10.1371/journal.pone.0056329 Bowers RM, Sullivan AP, Costello EK, et al (2011) Sources of bacteria in outdoor air across cities in the midwestern United States. Appl Environ Microbiol 77:6350–6. doi: 10.1128/AEM.05498-11 Bray J, Curtis J (1957) An ordination of the upland forest communities of southern Wisconsin. Ecol Monogr 27:325–349. Broeckling CD, Broz AK, Bergelson J, et al (2008) Root exudates regulate soil fungal community composition and diversity. Appl Environ Microbiol 74:738–744. doi: 10.1128/AEM.02188-07 Brutnell TP, Wang L, Swartwood K, et al (2010) Setaria viridis: a model for C4 photosynthesis. Plant Cell 22:2537–2544. doi: 10.1105/tpc.110.075309 Buée M, de Boer W, Martin F, et al (2009) The rhizosphere zoo: An overview of plant-associated communities of microorganisms, including phages, bacteria, archaea, and fungi, and of some of their structuring factors. Plant Soil 321:189–212. doi: 10.1007/s11104-009-9991-3 Bulgarelli D, Rott M, Schlaeppi K, et al (2012) Revealing structure and assembly cues for Arabidopsis root-inhabiting bacterial microbiota. Nature 488:91–95. 104  Bulgarelli D, Schlaeppi K, Spaepen S, et al (2013) Structure and functions of the bacterial microbiota of plants. Annu Rev Plant Biol 64:807–38. doi: 10.1146/annurev-arplant-050312-120106 Buyer JS, Roberts DP, Russek-Cohen E (1999) Microbial community structure and function in the spermosphere as affected by soil and seed type. Can. J. Microbiol. 45:138–144. Bååth E, Anderson T-H (2003) Comparison of soil fungal/bacterial ratios in a pH gradient using physiological and PLFA-based techniques. Soil Biol Biochem 35:955–963. doi: 10.1016/S0038-0717(03)00154-8 Caporaso JG, Bittinger K, Bushman FD, et al (2010a) PyNAST: A flexible tool for aligning sequences to a template alignment. Bioinformatics 26:266–267. doi: 10.1093/bioinformatics/btp636 Caporaso JG, Kuczynski J, Stombaugh J, et al (2010b) QIIME allows analysis of high-throughput community sequencing data. Nat. Methods 7:335–336. Chao A, Chazdon RL, Shen TJ (2005) A new statistical approach for assessing similarity of species composition with incidence and abundance data. Ecol Lett 8:148–159. doi: 10.1111/j.1461-0248.2004.00707.x Chaparro JM, Badri D V, Bakker MG, et al (2013) Root Exudation of Phytochemicals in Arabidopsis Follows Specific Patterns That Are Developmentally Programmed and Correlate with Soil Microbial Functions. PLoS One 8:e55731. doi: 10.1371/journal.pone.0055731 Chaparro JM, Badri D V, Vivanco JM (2014) Rhizosphere microbiome assemblage is affected by plant development. ISME J 8:790–803. doi: 10.1038/ismej.2013.196 Chase J (2003) Community assembly: when should history matter? Oecologia 136:489–498. Chelius MK, Triplett EW (2001) The Diversity of Archaea and Bacteria in Association with the Roots of Zea mays L. Microb Ecol 41:252–263. doi: 10.1007/s002480000087 Chesson P (2000) Mechanisms of maintenance of species diversity. Annu. Rev. Ecol. Syst. 31:343–366. Claridge AW, Tanton MT, Seebeck JH, et al (1992) Establishment of ectomycorrhizae on the roots of two species of Eucalyptus from fungal spores contained in the faeces of the long-nosed potoroo (Potorous tridactylus). Aust J Ecol 17:207–217. doi: 10.1111/j.1442-9993.1992.tb00799.x Clarke K, Gorley R (2006) PRIMER v6: User Manual/Tutorial.  Clay K (2001) Symbiosis and the Regulation of Communities1. Am. Zool. 41:810–824. 105  Compant S, Clément C, Sessitsch A (2010) Plant growth-promoting bacteria in the rhizo- and endosphere of plants: Their role, colonization, mechanisms involved and prospects for utilization. Soil Biol Biochem 42:669–678. doi: 10.1016/j.soilbio.2009.11.024 Compant S, Mitter B, Colli-Mull JG, et al (2011) Endophytes of Grapevine Flowers, Berries, and Seeds: Identification of Cultivable Bacteria, Comparison with Other Plant Parts, and Visualization of Niches of Colonization. Microb Ecol 62:188–197. doi: 10.1007/s00248-011-9883-y Conrath U, Beckers GJM, Flors V, et al (2006) Priming: getting ready for battle. Mol Plant Microbe Interact 19:1062–1071. doi: 10.1094/MPMI-19-1062 Costa R, Götz M, Mrotzek N, et al (2006) Effects of site and plant species on rhizosphere community structure as revealed by molecular analysis of microbial guilds. In: FEMS Microbiology Ecology. pp 236–249 Costello EK, Stagaman K, Dethlefsen L, et al (2012) The Application of Ecological Theory Toward an Understanding of the Human Microbiome. Science (80-. ). 336:1255–1262. Cueva C, Moreno-Arribas MV, Martín-Álvarez PJ, et al (2010) Antimicrobial activity of phenolic acids against commensal, probiotic and pathogenic bacteria. Res Microbiol 161:372–382. doi: 10.1016/j.resmic.2010.04.006 Da Silva LJ (2015) Package “laercio”: Duncan test, Tukey test and Scott-Knott test.  Dantas G, Sommer M (2012) Ecological and Clinical Consequences of Antibiotic Subsistence by Environmental Microbes. In: Montforts M, Keen P (eds) Antimicrobial Resistance in the Environment. John Wiley & Sons, Hoboken, NJ, USA,  Davis EC, Franklin JB, Shaw AJ, Vilgalys R (2003) Endophytic Xylaria (Xylariaceae) among liverworts and angiosperms: Phylogenetics, distribution, and symbiosis. Am J Bot 90:1661–1667. doi: 10.3732/ajb.90.11.1661 De Boer W, Folman LB, Summerbell RC, Boddy L (2005) Living in a fungal world: Impact of fungi on soil bacterial niche development. FEMS Microbiol. Rev. 29:795–811. De Ridder-Duine AS, Kowalchuk G a., Klein Gunnewiek PJA, et al (2005) Rhizosphere bacterial community composition in natural stands of Carex arenaria (sand sedge) is determined by bulk soil community composition. Soil Biol Biochem 37:349–357. doi: 10.1016/j.soilbio.2004.08.005 De Vries FT, Shade A (2013) Controls on soil microbial community stability under climate change. Front Microbiol 4:265. doi: 10.3389/fmicb.2013.00265 106  De-la-Peña C, Badri D V, Lei Z, et al (2010) Root secretion of defense-related proteins is development-dependent and correlated with flowering time. J Biol Chem 285:30654–65. doi: 10.1074/jbc.M110.119040 Denison RF, Kiers ET (2011) Life histories of symbiotic rhizobia and mycorrhizal fungi. Curr Biol 21:R775–R785. doi: 10.1016/j.cub.2011.06.018 DeSantis TZ, Hugenholtz P, Larsen N, et al (2006) Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl Environ Microbiol 72:5069–5072. doi: 10.1128/AEM.03006-05 Dodd J, Jeffries P (1986) Early development of vesicular-arbuscular mycorrhizas in autumn-sown cereals. Soil Biol. Biochem. 18:149–154. Doornbos RF, Geraats BPJ, Kuramae EE, et al (2011) Effects of jasmonic acid, ethylene, and salicylic acid signaling on the rhizosphere bacterial community of Arabidopsis thaliana. Mol Plant Microbe Interact 24:395–407. doi: 10.1094/MPMI-05-10-0115 Doornbos RF, Van Loon LC, Bakker PAHM (2012) Impact of root exudates and plant defense signaling on bacterial communities in the rhizosphere. A review. Agron. Sustain. Dev. 32:227–243. Drake JA (1991) Community-assemby mechanics and the structure of an experimental species ensemble. Am Nat 137:1–26. Dumbrell AJ, Ashton PD, Aziz N, et al (2011) Distinct seasonal assemblages of arbuscular mycorrhizal fungi revealed by massively parallel pyrosequencing. New Phytol 190:794–804. doi: 10.1111/j.1469-8137.2010.03636.x Dumbrell AJ, Nelson M, Helgason T, et al (2010) Relative roles of niche and neutral processes in structuring a soil microbial community. ISME J 4:337–345. doi: 10.1038/ismej.2009.122 Dunne MJ, Fitter AH (1989) The phosphorus budget of a field-grown strawberry (Fragaria x ananassa cv. Hapil) crop: evidence for a mycorrhizal contribution. Ann Appl Biol 114:185–193. doi: 10.1111/j.1744-7348.1989.tb06799.x Edwards J, Johnson C, Santos-Medellín C, et al (2015) Structure, variation, and assembly of the root-associated microbiomes of rice. Proc Natl Acad Sci 201414592. doi: 10.1073/pnas.1414592112 Eisenlord SD, Zak DR, Upchurch RA (2012) Dispersal limitation and the assembly of soil Actinobacteria communities in a long-term chronosequence. Ecol Evol 2:538–549. doi: 10.1002/ece3.210 107  Elton CS (1958) The Ecology of Invasions by animals and plants. Chapman and Hall Ltd, London, UK Farrar J, Hawes M, Jones D, Lindow S (2003) How roots control the flux of carbon to the rhizosphere. Ecology 84:827–837. Fierer N (2008) Microbial biogeography : patterns in microbial diversity across space and time. In: Zengler K (ed) Accessing Uncultivated Microorganisms: from the Environment to Organisms and Genomes and Back. ASM press, Washington DC, USA, pp 95–115 Fierer N, Bradford M a., Jackson RB (2007) Toward an ecological classification of soil bacteria. Ecology 88:1354–1364. doi: 10.1890/05-1839 Fierer N, Hamady M, Lauber CL, Knight R (2008) The influence of sex, handedness, and washing on the diversity of hand surface bacteria. Proc Natl Acad Sci U S A 105:17994–17999. doi: 10.1073/pnas.0807920105 Fierer N, Jackson RB (2006) The diversity and biogeography of soil bacterial communities. Proc Natl Acad Sci U S A 103:626–631. doi: 10.1073/pnas.0507535103 Fierer N, Nemergut D, Knight R, Craine JM (2010) Changes through time: Integrating microorganisms into the study of succession. Res Microbiol 161:635–642. doi: 10.1016/j.resmic.2010.06.002 Fierer N, Schimel JP, Holden PA (2003) Variations in microbial community composition through two soil depth profiles. Soil Biol Biochem 35:167–176. doi: 10.1016/S0038-0717(02)00251-1 Finlay BJ (2002) Global dispersal of free-living microbial eukaryote species. Science (80- ) 296:1061–1063. doi: 10.1126/science.1070710 Fitter A (2002) Characteristics and functions of root systems. In: Waisel Y, Eshel A, Kafafi U (eds) Plant roots: the hidden half, 3rd edn. Marcel Dekker Inc., New York, NY, USA,  Fourie JF, Holz G (1998) Effects of fruit and pollen exudates on growth of Botrytis cinerea and infection of plum and nectarine fruit. Plant Dis. 82:165–170. Friese CF, Allen MF (1991) The spread of VA mycorrhizal fungal hyphae in the soil: inoculum types and external hyphal architecture. Mycologia 83:409–418. doi: 10.2307/3760351 Fukami T, Bezemer TM, Mortimer SR, Van Der Putten WH (2005) Species divergence and trait convergence in experimental plant community assembly. Ecol Lett 8:1283–1290. doi: 10.1111/j.1461-0248.2005.00829.x 108  Fukami T, Dickie IA, Paula Wilkie J, et al (2010) Assembly history dictates ecosystem functioning: Evidence from wood decomposer communities. Ecol. Lett. 13:675–684. Fukami T, Nakajima M (2011) Community assembly: Alternative stable states or alternative transient states? Ecol Lett 14:973–984. doi: 10.1111/j.1461-0248.2011.01663.x Fulthorpe R (1998) High levels of endemicity of 3-chlorobenzoate-degrading soil bacteria. Appl Environ Microbiol 64:1620–1627. Fürnkranz M, Lukesch B, Müller H, et al (2012) Microbial diversity inside pumpkins: Microhabitat-specific communities display a high antagonistic potential against phytopathogens. Microb Ecol 63:418–428. doi: 10.1007/s00248-011-9942-4 Garbeva P, van Elsas JD, van Veen JA (2008) Rhizosphere microbial community and its response to plant species and soil history. Plant Soil 302:19–32. doi: 10.1007/s11104-007-9432-0 Garbeva P, van Veen J a, van Elsas JD (2004) Microbial diversity in soil: selection microbial populations by plant and soil type and implications for disease suppressiveness. Annu Rev Phytopathol 42:243–70. doi: 10.1146/annurev.phyto.42.012604.135455 Gilbert JA, Meyer F (2012) Modeling the Earth’s Microbiome A real world deliverable for microbial ecology.  Gilbert JA, Steele JA, Caporaso JG, et al (2012) Defining seasonal marine microbial community dynamics. ISME J 6:298–308. doi: 10.1038/ismej.2011.107 Giroday H, Baker V (2006) Invasive hawkweeds (Hieracium spp.) in Northeastern British Columbia. Prince George Gomes NCM, Heuer H, Schönfeld J, et al (2001) Edited by Foxit Reader Bacterial diversity of the rhizosphere of maize ( Zea mays ) grown in tropical soil studied by temperature gradient gel electrophoresis. Plant Soil 167–180. Gonzalez A, Clemente JC, Shade A, et al (2011) Our microbial selves: what ecology can teach us. EMBO Rep 12:775–784. doi: 10.1038/embor.2011.137 Gonzalez A, Stombaugh J, Lauber CL, et al (2012) SitePainter: A tool for exploring biogeographical patterns. Bioinformatics 28:436–438. doi: 10.1093/bioinformatics/btr685 Gottel NR, Castro HF, Kerley M, et al (2011) Distinct microbial communities within the endosphere and rhizosphere of Populus deltoides roots across contrasting soil types. Appl Environ Microbiol 77:5934–5944. doi: 10.1128/AEM.05255-11 109  Gransee A, Wittenmayer L (2000) Qualitative and quantitative analysis of water-soluble root exudates in relation to plant species and development. J Plant Nutr Soil Sci 163:381–385. doi: 10.1002/1522-2624(200008)163:4<381::AID-JPLN381>3.0.CO;2-7 Green J, Bohannan BJM (2006) Spatial scaling of microbial biodiversity. Trends Ecol. Evol. 21:501–507. Green SJ, Inbar E, Michel FC, et al (2006) Succession of bacterial communities during early plant development: Transition from seed to root and effect of compost amendment. Appl Environ Microbiol 72:3975–3983. doi: 10.1128/AEM.02771-05 Green SJ, Michel FC, Hadar Y, Minz D (2007) Contrasting patterns of seed and root colonization by bacteria from the genus Chryseobacterium and from the family Oxalobacteraceae. ISME J 1:291–299. doi: 10.1038/ismej.2007.33 Grimm V, Wissel C (1997) Babel, or the ecological stability discussions: An inventory and analysis of terminology and a guide for avoiding confusion. Oecologia 109:323–334. Grundmann GL (2004) Spatial scales of soil bacterial diversity - the size of a clone. FEMS Microbiol Ecol 48:119–127. doi: 10.1016/j.femsec.2004.01.010 Gutjahr C, Paszkowski U (2009) Weights in the balance: jasmonic acid and salicylic acid signaling in root-biotroph interactions. Mol Plant Microbe Interact 22:763–772. doi: 10.1094/MPMI-22-7-0763 Haas BJ, Gevers D, Earl AM, et al (2011) Chimeric 16S rRNA sequence formation and detection in Sanger and 454-pyrosequenced PCR amplicons. Genome Res 21:494–504. doi: 10.1101/gr.112730.110 Haichar FEZ, Marol C, Berge O, et al (2008) Plant host habitat and root exudates shape soil bacterial community structure. ISME J 2:1221–1230. doi: 10.1038/ismej.2008.80 Hamady M, Knight R (2009) Microbial community profiling for human microbiome projects: Tools, techniques, and challenges. Genome Res 19:1141–52. doi: 10.1101/gr.085464.108 Hamady M, Lozupone C, Knight R (2010) Fast UniFrac: facilitating high-throughput phylogenetic analyses of microbial communities including analysis of pyrosequencing and PhyloChip data. ISME J 4:17–27. doi: 10.1038/ismej.2009.97 Hansen CHF, Metzdorff SB, Hansen AK (2013) Customizing laboratory mice by modifying gut microbiota and host immunity in an early “window of opportunity.” Gut Microbes 4:241–5. doi: 10.4161/gmic.23999 110  Hardoim PR, Hardoim CCP, van Overbeek LS, van Elsas JD (2012) Dynamics of seed-borne rice endophytes on early plant growth stages. PLoS One 7:e30438. doi: 10.1371/journal.pone.0030438 Hardoim, PR, van Overbeek LS, Berg G et al. 2015. The hidden world within plants: ecological and evolutionary considerations for defining functioning of microbial endophytes. Microbiol Mol Biol Rev 79: 293–320. doi:10.1128/MMBR.00050-14. Hart LP, Endo RM (1981) The effect of time of exposure to inoculum, plant age, root development, and root wounding on Fusarium yellows of celery. Phytopathology 71:77. Hartmann A, Schmid M, Tuinen D van, Berg G (2008) Plant-driven selection of microbes. Plant Soil 321:235–257. doi: 10.1007/s11104-008-9814-y Hawkes C, DeAngelis K, Firestone M (2007) Root interactions with soil microbial communitites and processes. In: Cardon Z, Whitbeck J (eds) The Rhizosphere. Elsevier, New York, NY, USA,  Hazard C, Gosling P, van der Gast CJ, et al (2012) The role of local environment and geographical distance in determining community composition of arbuscular mycorrhizal fungi at the landscape scale. ISME J 1–11. doi: 10.1038/ismej.2012.127 van der Heijden MGA, Bardgett RD, Van Straalen NM (2008) The unseen majority: Soil microbes as drivers of plant diversity and productivity in terrestrial ecosystems. Ecol Lett 11:296–310. doi: 10.1111/j.1461-0248.2007.01139.x van der Heijden MGA, Streitwolf-Engel R, Riedl R, et al (2006) The mycorrhizal contribution to plant productivity, plant nutrition and soil structure in experimental grassland. New Phytol 172:739–52. doi: 10.1111/j.1469-8137.2006.01862.x Heslop-Harrison J, Heslop-Harrison Y (1985) Surfaces and secretions in the pollen-stigma interaction: a brief review. J Cell Sci 2:287–300. Heuer H, Kroppenstedt RM, Lottmann J, et al (2002) Effects of T4 lysozyme release from transgenic potato roots on bacterial rhizosphere communities are negligible relative to natural factors. Appl Environ Microbiol 68:1325–1335. doi: 10.1128/AEM.68.3.1325-1335.2002 Hibbing ME, Fuqua C, Parsek MR, Peterson SB (2010) Bacterial competition: surviving and thriving in the microbial jungle. Nat Rev Microbiol 8:15–25. doi: 10.1038/nrmicro2259 Hirsch PR, Mauchline TH (2012) Who’s who in the plant root microbiome? Nat. Biotechnol. 30:961–962. Hodge A, Berta G, Doussan C, et al (2009) Plant root growth, architecture and function. Plant Soil 321:153–187. 111  Hodge A, Helgason T, Fitter AH (2010) Nutritional ecology of arbuscular mycorrhizal fungi. Fungal Ecol. 3:267–273. Hoeksema JD, Chaudhary VB, Gehring C a, et al (2010) A meta-analysis of context-dependency in plant response to inoculation with mycorrhizal fungi. Ecol. Lett. 13:394–407. Horner-Devine MC, Carney KM, Bohannan BJM (2004) An ecological perspective on bacterial biodiversity. Proc R Soc B Biol Sci 271:113–122. doi: 10.1098/rspb.2003.2549 Houlden A, Timms-Wilson TM, Day MJ, Bailey MJ (2008) Influence of plant developmental stage on microbial community structure and activity in the rhizosphere of three field crops. FEMS Microbiol Ecol. pp 193–201 Hubbell SP (2001) The unified neutral theory of biodiversity and biogeography. Princeton University Press, Princeton, USA Huttenhower C, Gevers D, Knight R, et al (2012) Structure, function and diversity of the healthy human microbiome. Nature 486:207–214. Ibekwe AM, Grieve CM (2004) Changes in developing plant microbial community structure as affected by contaminated water. FEMS Microbiol Ecol 48:239–248. doi: 10.1016/j.femsec.2004.01.012 Ihrmark K, Bödeker ITM, Cruz-Martinez K, et al (2012) New primers to amplify the fungal ITS2 region - evaluation by 454-sequencing of artificial and natural communities. FEMS Microbiol Ecol 82:666–677. doi: 10.1111/j.1574-6941.2012.01437.x Inceoğlu O, Salles JF, van Overbeek L, van Elsas JD (2010) Effects of plant genotype and growth stage on the betaproteobacterial communities associated with different potato cultivars in two fields. Appl Environ Microbiol 76:3675–84. doi: 10.1128/AEM.00040-10 Ishida TA, Nara K, Tanaka M, et al (2008) Germination and infectivity of ectomycorrhizal fungal spores in relation to their ecological traits during primary succession. New Phytol 180:491–500. doi: 10.1111/j.1469-8137.2008.02572.x Jenkins DG, Brescacin CR, Duxbury C V., et al (2007) Does size matter for dispersal distance? Glob Ecol Biogeogr 16:415–425. doi: 10.1111/j.1466-8238.2007.00312.x Johnson D (2015) Priorities for research on priority effects. New Phytol 205:1375–1377. Johnson JH, Tymiak AA, Bolgar MS (1990) Janthinocins A, B and C, novel peptide lactone antibiotics produced by Janthinobacterium lividum. J Antibiot (Tokyo) 43:920–930. 112  Johnson NC (2010) Resource stoichiometry elucidates the structure and function of arbuscular mycorrhizas across scales. New Phytol. 185:631–647. Jones DL, Hodge A, Kuzyakov Y (2004) Plant and mycorrhizal regulation of rhizodeposition. New Phytol. 163:459–480. Jones JDG, Dangl JL (2006) The plant immune system. Nature 444:323–329. doi: 10.1038/nature05286 Junker RR, Keller A (2015) Microhabitat heterogeneity across leaves and flower organs promotes bacterial diversity. FEMS Microbiol Ecol. doi: 10.1093/femsec/fiv097 Junker RR, Loewel C, Gross R, et al (2011) Composition of epiphytic bacterial communities differs on petals and leaves. Plant Biol 13:918–924. doi: 10.1111/j.1438-8677.2011.00454.x Kane SR, Chakicherla AY, Chain PSG, et al (2007) Whole-genome analysis of the methyl tert-butyl ether-degrading beta-proteobacterium Methylibium petroleiphilum PM1. J Bacteriol 189:1931–1945. doi: 10.1128/JB.01259-06 Kardol P, Cornips NJ, Van Kempen MML, et al (2007) Microbe-mediated plant-soil feedback causes historical contingency effects in plant community assembly. Ecol Monogr 77:147–162. doi: 10.1890/06-0502 Kardol P, Souza L, Classen AT (2013) Resource availability mediates the importance of priority effects in plant community assembly and ecosystem function. Oikos 122:84–94. doi: 10.1111/j.1600-0706.2012.20546.x Kelly D, King T, Aminov R (2007) Importance of microbial colonization of the gut in early life to the development of immunity. Mutat Res - Fundam Mol Mech Mutagen 622:58–69. doi: 10.1016/j.mrfmmm.2007.03.011 Kennedy PG, Bruns TD (2005) Priority effects determine the outcome of ectomycorrhizal competition between two Rhizopogon species colonizing Pinus muricata seedlings. New Phytol 166:631–638. doi: 10.1111/j.1469-8137.2005.01355.x Kennedy PG, Peay KG, Bruns TD (2009) Root tip competition among ectomycorrhizal fungi: Are priority effects a rule or an exception? Ecology 90:2098–2107. doi: 10.1890/08-1291.1 Kirchman DL (2002) The ecology of Cytophaga-Flavobacteria in aquatic environments. FEMS Microbiol. Ecol. 39:91–100. Klein DA, McLendon T, Paschke MW, Redente EF (1996) Nitrogen availability and fungal-bacterial responses in successional semi-arid steppe soils. Arid Soil Restaur Rehabil 10:321–332. 113  Koenig JE, Spor A, Scalfone N, et al (2011) Succession of microbial consortia in the developing infant gut microbiome. Proc Natl Acad Sci U S A 108 Suppl :4578–4585. doi: 10.1073/pnas.1000081107 Kõljalg U, Nilsson RH, Abarenkov K, et al (2013) Towards a unified paradigm for sequence-based identification of fungi. Mol. Ecol. 22:5271–5277. Kotter MM, Farentinos RC (1984) Tassel-eared squirrels as spore dispersal agents of hypogeous mycorrhizal fungi. J Mammal 65:684. doi: 10.2307/1380853 Kuczynski J, Liu Z, Lozupone C, et al (2010) Microbial community resemblance methods differ in their ability to detect biologically relevant patterns. Nat Methods 7:813–819. doi: 10.1038/nmeth.1499 Kuske CR, Ticknor LO, Miller ME, et al (2002) Comparison of soil bacterial communities in rhizospheres of three plant species and the interspaces in an arid grassland. Appl Environ Microbiol 68:1854–1863. doi: 10.1128/AEM.68.4.1854 Larsen PE, Gibbons SM, Gilbert JA (2012) Modeling microbial community structure and functional diversity across time and space. FEMS Microbiol. Lett. 332:91–98. Latch GCM (1993) Physiological interactions of endophytic fungi and their hosts. Biotic stress tolerance imparted to grasses by endophytes. Agric. Ecosyst. Environ. 44:143–156. Lau JA, Lennon JT (2011) Evolutionary ecology of plant-microbe interactions: Soil microbial structure alters selection on plant traits. New Phytol 192:215–224. doi: 10.1111/j.1469-8137.2011.03790.x Lauber CL, Hamady M, Knight R, Fierer N (2009) Pyrosequencing-based assessment of soil pH as a predictor of soil bacterial community structure at the continental scale. Appl Environ Microbiol 75:5111–5120. doi: 10.1128/AEM.00335-09 Leff JW, Del Tredici P, Friedman WE, Fierer N (2015) Spatial structuring of bacterial communities within individual G inkgo biloba trees. Environ Microbiol 17:n/a–n/a. doi: 10.1111/1462-2920.12695 Leibold MA, McPeek MA (2006) Coexistence of the niche and neutral perspectives in community ecology. Ecology 87:1399–1410. Lekberg Y, Koide RT, Rohr JR, et al (2007) Role of niche restrictions and dispersal in the composition of arbuscular mycorrhizal fungal communities. J Ecol 95:95–105. doi: 10.1111/j.1365-2745.2006.01193.x 114  Leopold DR, Tanentzap AJ, Lee WG, et al (2015) Evolutionary priority effects in New Zealand alpine plants across environmental gradients. J Biogeogr 42:729–737. doi: 10.1111/jbi.12441 Ley RE, Peterson D a, Gordon JI (2006) Ecological and evolutionary forces shaping microbial diversity in the human intestine. Cell 124:837–848. Lindow SE, Brandl MT (2003) Microbiology of the phyllosphere. Appl Environ Microbiol 69:1875–1883. doi: 10.1128/AEM.69.4.1875 Liu Y, Zuo S, Zou Y, et al (2012) Investigation on diversity and population succession dynamics of indigenous bacteria of the maize spermosphere. World J Microbiol Biotechnol 28:391–396. doi: 10.1007/s11274-011-0822-3 Loomis ES, Fishman L (2009) A continent-wide clone: Populetion genetic variation of the invasive plant Hieracium arantiacum (Orange Hawkweed; Asteraceae) in North America. Int. J. Plant Sci. 170:759–765. Love MI, Huber W, Anders S (2014) Moderated estimation of fold change and dispersion for RNA-Seq data with DESeq2.  Lozupone C a, Hamady M, Kelley ST, Knight R (2007) Quantitative and qualitative β diversity measures lead to different insights into factors that structure microbial communities. Appl Environ Microbiol 73:1576–1585. doi: 10.1128/AEM.01996-06 Lozupone C a., Stombaugh JI, Gordon JI, et al (2012) Diversity, stability and resilience of the human gut microbiota. Nature 489:220–230. Lundberg DS, Lebeis SL, Paredes SH, et al (2012) Defining the core Arabidopsis thaliana root microbiome. Nature 488:86–90. Manter DK, Delgado JA, Holm DG, Stong RA (2010) Pyrosequencing reveals a highly diverse and cultivar-specific bacterial endophyte community in potato roots. Microb Ecol 60:157–166. doi: 10.1007/s00248-010-9658-x Marschner P, Crowley D, Rengel Z (2011) Rhizosphere interactions between microorganisms and plants govern iron and phosphorus acquisition along the root axis - model and research methods. Soil Biol. Biochem. 43:883–894. Marschner P, Grierson PF, Rengel Z (2005) Microbial community composition and functioning in the rhizosphere of three Banksia species in native woodland in Western Australia. Appl Soil Ecol 28:191–201. doi: 10.1016/j.apsoil.2004.09.001 Marschner P, Neumann G, Kania A (2002) Spatial and temporal dynamics of the microbial community structure in the rhizosphere of cluster roots of white lupin (Lupinus albus L.). Plant Soil 246:167–174. 115  Marschner P, Yang C, Lieberei R, Crowley DE (2001) Soil and plant specific effects on bacterial community composition in the rhizosphere. Soil Biol Biochem 33:1437–1445. Martiny JBH, Bohannan BJM, Brown JH, et al (2006) Microbial biogeography: putting microorganisms on the map. Nat Rev Microbiol 4:102–12. doi: 10.1038/nrmicro1341 Maser C, Trappe JM, Nussbaum RA (1978) Fungal-Small Mammal Interrelationships with Emphasis on Oregon Coniferous Forests. Ecology 59:799. doi: 10.2307/1938784 Masson-Boivin C, Giraud E, Perret X, Batut J (2009) Establishing nitrogen-fixing symbiosis with legumes: how many rhizobium recipes? Trends Microbiol 17:458–66. doi: 10.1016/j.tim.2009.07.004 McIlveen WD, Cole Jr. H (1976) Spore dispersal of Endogonaceae by worms, ants, wasps, and birds. Can. J. Bot. 54:1486–1489. McMurdie PJ, Holmes S (2014) Waste Not, Want Not: Why Rarefying Microbiome Data Is Inadmissible. PLoS Comput Biol 10:e1003531. doi: 10.1371/journal.pcbi.1003531 McMurdie PJ, Holmes S (2013) Phyloseq: An R Package for Reproducible Interactive Analysis and Graphics of Microbiome Census Data. PLoS One 8:e61217. doi: 10.1371/journal.pone.0061217 Meidinger D, Pojar J (1991) Ecosystems of British Columbia. British Columbia Ministry of Forests. 330. Mendes R, Garbeva P, Raaijmakers JM (2013) The rhizosphere microbiome: Significance of plant beneficial, plant pathogenic, and human pathogenic microorganisms. FEMS Microbiol. Rev. 37:634–663. Mendes R, Kruijt M, de Bruijn I, et al (2011) Deciphering the rhizosphere microbiome for disease-suppressive bacteria. Science 332:1097–100. doi: 10.1126/science.1203980 Merryweather J, Fitter A (1998) The arbuscular mycorrhizal fungi of Hyacinthoides non-scripta II. Seasonal and spatial patterns of fungal populations. New Phytol 138:131–142. doi: 10.1046/j.1469-8137.1998.00889.x Micallef SA, Shiaris MP, Colón-Carmona A (2009) Influence of Arabidopsis thaliana accessions on rhizobacterial communities and natural variation in root exudates. J Exp Bot 60:1729–1742. doi: 10.1093/jxb/erp053 Mougel C, Offre P, Ranjard L, et al (2006) Dynamic of the genetic structure of bacterial and fungal communities at different developmental stages of Medicago truncatula Gaertn. cv. Jemalong line J5. New Phytol 170:165–175. doi: 10.1111/j.1469-8137.2006.01650.x 116  Nelson EB (2004) Microbial dynamics and interactions in the spermosphere. Annu Rev Phytopathol 42:271–309. doi: 10.1146/annurev.phyto.42.121603.131041 Nemergut DR, Schmidt SK, Fukami T, et al (2013) Patterns and processes of microbial community assembly. Microbiol Mol Biol Rev 77:342–56. doi: 10.1128/MMBR.00051-12 Ngugi HK, Scherm H (2006) Biology of flower-infecting fungi. Annu Rev Phytopathol 44:261–82. doi: 10.1146/annurev.phyto.44.070505.143405 Normander B, Prosser JI (2000) Bacterial origin and community composition in the barley phytosphere as a function of habitat and presowing conditions. Appl Environ Microbiol 66:4372–4377. doi: 10.1128/AEM.66.10.4372-4377.2000 Norros V, Penttilä R, Suominen M, Ovaskainen O (2012) Dispersal may limit the occurrence of specialist wood decay fungi already at small spatial scales. Oikos 121:961–974. doi: 10.1111/j.1600-0706.2012.20052.x Oehl F, Laczko E, Bogenrieder A, et al (2010) Soil type and land use intensity determine the composition of arbuscular mycorrhizal fungal communities. Soil Biol Biochem 42:724–738. doi: 10.1016/j.soilbio.2010.01.006 Oksanen J, Blanchet F, Kindt R, et al (2014) Vegan: community ecology package.  Opelt K, Chobot V, Hadacek F, et al (2007) Investigations of the structure and function of bacterial communities associated with Sphagnum mosses. Environ Microbiol 9:2795–809. doi: 10.1111/j.1462-2920.2007.01391.x Osanai Y, Bougoure DS, Hayden HL, Hovenden MJ (2013) Co-occurring grass species differ in their associated microbial community composition in a temperate native grassland. Plant Soil 368:419–431. doi: 10.1007/s11104-012-1529-4 Ottesen AR, González Peña A, White JR, et al (2013) Baseline survey of the anatomical microbial ecology of an important food plant: Solanum lycopersicum (tomato). BMC Microbiol 13:114. doi: 10.1186/1471-2180-13-114 van Overbeek L, van Elsas JD (2008) Effects of plant genotype and growth stage on the structure of bacterial communities associated with potato (Solanum tuberosum L.). FEMS Microbiol Ecol 64:283–296. doi: 10.1111/j.1574-6941.2008.00469.x van Overbeek LS, Franke AC, Nijhuis EHM, et al (2011) Bacterial Communities Associated with Chenopodium album and Stellaria media Seeds from Arable Soils. Microb Ecol 62:257–264. doi: 10.1007/s00248-011-9845-4 Packer A, Clay K (2000) Soil pathogens and spatial patterns of seedling mortality in a temperate tree. Nature 404:278–281. doi: 10.1038/35005072 117  Peay KG, Belisle M, Fukami T (2012) Phylogenetic relatedness predicts priority effects in nectar yeast communities. Proc. R. Soc. B Biol. Sci. 279:749–758. Peiffer JA, Spor A, Koren O, et al (2013) Diversity and heritability of the maize rhizosphere microbiome under field conditions. Proc Natl Acad Sci U S A 110:6548–53. doi: 10.1073/pnas.1302837110 Pimm SL (1984) The complexity and stability of ecosystems. Nature 307:321–326. Pivato B, Mazurier S, Lemanceau P, et al (2007) Medicago species affect the community composition of arbuscular mycorrhizal fungi associated with roots. New Phytol 176:197–210. doi: 10.1111/j.1469-8137.2007.02151.x Pivato B, Offre P, Marchelli S, et al (2009) Bacterial effects on arbuscular mycorrhizal fungi and mycorrhiza development as influenced by the bacteria, fungi, and host plant. Mycorrhiza 19:81–90. doi: 10.1007/s00572-008-0205-2 Porras-Alfaro A, Bayman P (2011) Hidden fungi, emergent properties: endophytes and microbiomes. Annu Rev Phytopathol 49:291–315. doi: 10.1146/annurev-phyto-080508-081831 Raaijmakers JM, Paulitz TC, Steinberg C, et al (2009) The rhizosphere: A playground and battlefield for soilborne pathogens and beneficial microorganisms. Plant Soil 321:341–361. doi: 10.1007/s11104-008-9568-6 Raftoyannis Y, Dick MW (2002) Effects of inoculum density, plant age and temperature on disease severity caused by pythiaceous fungi on several plants. Phytoparasitica 30:67–76. Ramette A, Tiedje JM (2007) Multiscale responses of microbial life to spatial distance and environmental heterogeneity in a patchy ecosystem. Proc Natl Acad Sci U S A 104:2761–2766. doi: 10.1073/pnas.0610671104 Ramírez-Puebla ST, Servín-Garcidueñas LE, Jiménez-marín B, et al (2013) Gut and root microbiota commonalities. Appl. Environ. Microbiol. 79:2–9. Redford AJ, Bowers RM, Knight R, et al (2010) The ecology of the phyllosphere: Geographic and phylogenetic variability in the distribution of bacteria on tree leaves. Environ Microbiol 12:2885–2893. doi: 10.1111/j.1462-2920.2010.02258.x Reysenbach A, Pace N (1995) Reliable amplification of hyperthermophilic archaeal 16S rRNA genes by the polymerase chain reaction. In: Robb F (ed) Archaea: a laboratory manual. Cold Spring Harbor Laboratory Press, New York, NY, USA, pp 101–107 118  Rillig MC, Mummey DL, Ramsey PW, et al (2006) Phylogeny of arbuscular mycorrhizal fungi predicts community composition of symbiosis-associated bacteria. FEMS Microbiol Ecol 57:389–395. doi: 10.1111/j.1574-6941.2006.00129.x Robinson GR, Quinn JF, Stanton ML (1995) Invasibility of Experimental Habitat Islands in a California Winter Annual Grassland. Ecology 76:786–794. Rodriguez R, Redman R (2008) More than 400 million years of evolution and some plants still can’t make it on their own: Plant stress tolerance via fungal symbiosis. J Exp Bot 59:1109–1114. doi: 10.1093/jxb/erm342 Roumet C, Urcelay C, Díaz S (2006) Suites of root traits differ between annual and perennial species growing in the field. New Phytol 170:357–368. doi: 10.1111/j.1469-8137.2006.01667.x Rousk J, Brookes PC, Bååth E (2011) Fungal and bacterial growth responses to N fertilization and pH in the 150-year “Park Grass” UK grassland experiment. FEMS Microbiol Ecol 76:89–99. doi: 10.1111/j.1574-6941.2010.01032.x Rousk J, Bååth E, Brookes PC, et al (2010) Soil bacterial and fungal communities across a pH gradient in an arable soil. ISME J 4:1340–1351. doi: 10.1038/ismej.2010.58 Rousk J, Demoling LA, Bahr A, Bååth E (2008) Examining the fungal and bacterial niche overlap using selective inhibitors in soil. FEMS Microbiol Ecol 63:350–358. doi: 10.1111/j.1574-6941.2008.00440.x Schmid M, Baldani JIVO, Hartmann A (2006) The Prokaryotes. Springer New York, New York, NY Schnitzer SA, Klironomos JN, HilleRis Lambers J, et al (2011) Soil microbes drive the classic plant diversity-productivity pattern. Ecology 92:296–303. doi: 10.1890/10-0773.1 Schweitzer JA, Bailey JK, Fischer DG, et al (2008) Plant-soil-microorganism interactions: Heritable relationship between plant genotype and associated soil microorganisms. Ecology 89:773–781. doi: 10.1890/07-0337.1 Sekirov I, Russell SL, Antunes LCM (2010) Gut microbiota in health and disease. Physiol Rev 90:859–904. doi: 10.1152/physrev.00045.2009. Shade A, McManus PS, Handelsman J (2013) Unexpected diversity during community succession in the apple flower microbiome. MBio 4:e00602–12. doi: 10.1128/mBio.00602-12 119  Shivaji S, Ray M, Kumar G, et al (1991) Identification of Janthinobacterium lividum from the soils of the islands of Scotia Ridge and from Antarctic peninsula. Polar Biol 11:267–271. Shuster SM, Lonsdorf E V, Wimp GM, et al (2006) Community heritability measures the evolutionary consequences of indirect genetic effects on community structure. Evolution (N Y) 60:991–1003. doi: 10.1111/j.0014-3820.2006.tb01177.x Šimek K, Kasalický V, Jezbera J, et al (2010) Broad habitat range of the phylogenetically narrow R-BT065 cluster, representing a core group of the betaproteobacterial genus limnohabitans. Appl Environ Microbiol 76:631–639. doi: 10.1128/AEM.02203-09 Singh BK, Munro S, Potts JM, Millard P (2007) Influence of grass species and soil type on rhizosphere microbial community structure in grassland soils. Appl Soil Ecol 36:147–155. doi: 10.1016/j.apsoil.2007.01.004 Six J, Frey SD, Thiet RK, Batten KM (2006) Bacterial and Fungal Contributions to Carbon Sequestration in Agroecosystems. Soil Sci. Soc. Am. J. 70:555. Smalla K, Wieland G, Buchner A, et al (2001) Bulk and rhizosphere soil bacterial communities studied by Denaturing Gradient Gel Electrophoresis : Plant-dependent enrichment and seasonal shifts revealed. Appl Environ Microbiol 67:4742–4751. doi: 10.1128/AEM.67.10.4742 Smith DP, Peay KG (2014) Sequence depth, not PCR replication, improves ecological inference from next generation DNA sequencing. PLoS One 9:e90234. doi: 10.1371/journal.pone.0090234 Smith S (2001) Arbuscular Mycorrhizas: Physiology and Function. Soil Biol. Biochem. 33:1575–1576. Smith SE, Read DJ (2008) Mycorrhizal symbiosis. Academic Press, London, UK Spor A, Koren O, Ley R (2011) Unravelling the effects of the environment and host genotype on the gut microbiome. Nat Rev Microbiol 9:279–290. doi: 10.1038/nrmicro2540 Stockwell VO, McLaughlin RJ, Henkels MD, et al (1999) Epiphytic colonization of pear stigmas and hypanthia by bacteria during primary bloom. Phytopathology 89:1162–8. doi: 10.1094/PHYTO.1999.89.12.1162 Swinnen IAM, Bernaerts K, Dens EJJ, et al (2004) Predictive modelling of the microbial lag phase: A review. Int. J. Food Microbiol. 94:137–159. Tedersoo L, Nilsson RH, Abarenkov K, et al (2010) 454 Pyrosequencing and Sanger sequencing of tropical mycorrhizal fungi provide similar results but reveal substantial 120  methodological biases. New Phytol 188:291–301. doi: 10.1111/j.1469-8137.2010.03373.x Theimer T, Gehring C (2007) Mycorrhizal plants and vertebrate seed and spore dispersal: incorporating mycorrhizas into the seed dispersal paradigm. In: Dennis AJ, Green RJ, Schupp EW, Westcott DA (eds) Seed Dispersal: Theory and its Application in a Changing World. CAB International 2007, pp 463–478 Thomas H (2013) Senescence, ageing and death of the whole plant. New Phytol. 197:696–711. Tkacz A, Cheema J, Chandra G, et al (2015) Stability and succession of the rhizosphere microbiota depends upon plant type and soil composition. ISME J 1–11. doi: 10.1038/ismej.2015.41 Toljander JF, Santos-González JC, Tehler A, Finlay RD (2008) Community analysis of arbuscular mycorrhizal fungi and bacteria in the maize mycorrhizosphere in a long-term fertilization trial. FEMS Microbiol Ecol 65:323–338. doi: 10.1111/j.1574-6941.2008.00512.x Torsvik V, Øvreås L (2002) Microbial diversity and function in soil: from genes to ecosystems. Curr Opin Microbiol 5:240–5. Traveset A (1998) Effect of seed passage through vertebrate frugivores’ guts on germination: a review. Perspect Plant Ecol Evol Syst 1:151–190. doi: 10.1078/1433-8319-00057 Traveset A, Verdú M (2002) A Meta-analysis of the Seed Germination. In: Levey DJ, Silva WR, Galetti M (eds) Seed Dispersal and Frugivory: Ecology, Evolution and Conservation. CABI, pp 339–350 Treseder KK, Allen MF (2002) Direct nitrogen and phosphorus limitation of arbuscular mycorrhizal fungi: A model and field test. New Phytol 155:507–515. doi: 10.1046/j.1469-8137.2002.00470.x Truyens S, Weyens N, Cuypers A, Vangronsveld J (2014) Bacterial seed endophytes: genera, vertical transmission and interaction with plants. Environ Microbiol Rep n/a–n/a. Turnbaugh PJ, Hamady M, Yatsunenko T, et al (2009) A core gut microbiome in obese and lean twins. Nature 457:480–484. doi: 10.1038/nature07540  Verma V, Sudan P, Kour A (2008) Endophytes : A Novel Source for Bioactive Molecules. Proc Indian Natl Sci Acad 74:73–86.  121  van de Voorde TFJ, van der Putten WH, Martijn Bezemer T (2011) Intra- and interspecific plant-soil interactions, soil legacies and priority effects during old-field succession. J Ecol 99:945–953. doi: 10.1111/j.1365-2745.2011.01815.x Vorholt J a (2012) Microbial life in the phyllosphere. Nat Rev Microbiol 10:828–40. doi: 10.1038/nrmicro2910 Vos M, Wolf AB, Jennings SJ, Kowalchuk G a (2013) Micro-scale determinants of bacterial diversity in soil. FEMS Microbiol Rev 37:936–54. doi: 10.1111/1574-6976.12023 Wallenstein MD, Hall EK (2012) A trait-based framework for predicting when and where microbial adaptation to climate change will affect ecosystem functioning. Biogeochemistry 109:35–47. doi: 10.1007/s10533-011-9641-8 Wang B, Qiu YL (2006) Phylogenetic distribution and evolution of mycorrhizas in land plants. Mycorrhiza 16:299–363. Wang Q, Garrity GM, Tiedje JM, Cole JR (2007) Naïve Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol 73:5261–5267. doi: 10.1128/AEM.00062-07 Warner NJ, Allen MF, MacMahon JA (1987) Dispersal Agents of Vesicular-Arbuscular Mycorrhizal Fungi in a Disturbed Arid Ecosystem. Mycologia 79:721. doi: 10.2307/3807824 Watt M, Silk WK, Passioura JB (2006) Rates of root and organism growth, soil conditions, and temporal and spatial development of the rhizosphere. In: Annals of Botany. pp 839–855 Weinert N, Meincke R, Gottwald C, et al (2009) Rhizosphere communities of genetically modified zeaxanthin-accumulating potato plants and their parent cultivar differ less than those of different potato cultivars. Appl Environ Microbiol 75:3859–65. doi: 10.1128/AEM.00414-09 Werner GDA, Kiers ET (2015) Order of arrival structures arbuscular mycorrhizal colonization of plants. New Phytol 205:1515–1524. doi: 10.1111/nph.13092 Whitaker RJ, Grogan DW, Taylor JW (2003) Geographic barriers isolate endemic populations of hyperthermophilic archaea. Science (80- ) 301:976–978. doi: 10.1126/science.1086909 Whitham TG, Bailey JK, Schweitzer JA, et al (2006) A framework for community and ecosystem genetics: from genes to ecosystems. Nat Rev Genet 7:510–23. doi: 10.1038/nrg1877 122  Wieland G, Neumann R, Backhaus H (2001) Variation of microbial communities in soil, rhizosphere, and rhizoplane in response to crop species, soil type, and crop development. Appl Env Microbiol 67:5849–5854. doi: 10.1128/AEM.67.12.5849–5854.2001 Wilson GWT, Hartnett DC (1998) Interspecific variation in plant responses to mycorrhizal colonization in tallgrass prairie. Am J Bot 85:1732–1738. doi: 10.2307/2446507 Yu L, Nicolaisen M, Larsen J, Ravnskov S (2012) Succession of root-associated fungi in Pisum sativum during a plant growth cycle as examined by 454 pyrosequencing. Plant Soil 358:225–233. doi: 10.1007/s11104-012-1188-5 Zamioudis C, Pieterse CMJ (2012) Modulation of Host Immunity by Beneficial Microbes. Mol. Plant-Microbe Interact. MPMI 25:139–150. Zinger L, Shahnavaz B, Baptist F, et al (2009) Microbial diversity in alpine tundra soils correlates with snow cover dynamics. ISME J 3:850–9. doi: 10.1038/ismej.2009.20   123  Appendices Appendix A Supporting material for Chapter 4  Table A.1. Number of samples with successful sequence amplification available for analysis from each treatment and harvest, comparing the rarefied dataset (bacteria: 2000 sequences/sample, fungi: 1000 sequences/sample) with the non-rarefied dataset. Rarefied dataset        Harvest 1   Harvest 2   Harvest 3   Age at inoculation Bacteria Fungi Bacteria Fungi Bacteria Fungi 9 weeks 4 5 4 5 3 3 8 weeks 4 2 5 5 4 4 7 weeks 5 3 5 2 5 5 1 week 1 1 2 0 2 1 0 weeks Not available Not available 3 0 5 3        Non-rarefied dataset  Harvest 1   Harvest 2   Harvest 3   Age at inoculation Bacteria Fungi Bacteria Fungi Bacteria Fungi 9 weeks 5 5 5 5 5 4 8 weeks 5 3 5 5 4 5 7 weeks 5 3 5 4 5 5 1 week 4 1 4 0 2 3 0 weeks Not available Not available 4 1 5 5         Table A.2. Survival rates in plants inoculated at different ages out of 10 plants for each treatment that were dedicated for sampling at Harvest 3, when the plants had reached the age of 12 weeks  Age at inoculation Number of plants alive at Harvest 3  0 weeks 8 1 week 3 7 weeks 10 8 weeks 9 9 weeks 9 124  Table A.3. Results from running a 2-way PERMANOVA, comparing log transformed Bray Curtis distances between bacterial and fungal communities from Harvest 1 and Harvest 2 based on the factors Harvest and Age at harvest. Because of sample loss due to low amplification, we were not able to assess the interaction between harvest and age at harvest for fungal communities as several age classes were represented by too low number of samples (see table A.2. for numbers).    Bacteria Fungi  Pseudo-F p Pseudo-F p Harvest 1.61 0.04 1.16 0.24 Age at harvest 1.67 0.0001 1.17 0.09 HaXAg** 0.91 0.58 No test No test      ** Term has one or more empty cells 125    Fig. A.1. A comparison of the average community composition in bacterial (a) and fungal (b) communities of plants harvested at the same age (9 weeks old) prior to (Harvest 1) or two weeks after (Harvest 2) soil inoculation. Community composition is described as the proportion of sequences classified to different bacterial and fungal orders.  126  Appendix B Supporting material for Chapter 5  Table B.1. Additional information about the origin and properties of the resident and exogenous soils used in the experiment.  Resident soil Exogenous soil Location Okanagan campus grounds of University of British Columbia in Kelowna, BC, Canada Glacier national park, Montana, USA Coordinates 49.939975N, -119.399264W 48.288020N, -113.205170W Elevation 344m 2064m Ecosystem anthropogenic/disturbed/urban subalpine grassland Plant community  Setaria viridis, Tragopogon dubius, Achillea millefolium, Senecio vulgaris, Chenopodium album, Kochia scoparia, Rumex crispus,  Dasiflora fruticosa, Festuca scabrella, Pseudoroegneria spicata, Lupinus spp., Achillea millefolium, Townsendia parryi Soil type clay loam silt loam Soil organic matter content (LOI 360oC) 2%  6% Soil colour Gray black    127  Table B.2. Climate data for the two regions where the soils were collected. MAT = mean annual temperature (°C), MWMT = mean warmest month temperature (°C), MCMT = mean coldest month temperature (°C), TD = temperature difference between MWMT and MCMT or continentality (°C), MAP = mean annual precipitation (mm), MSP =May to September precipitation (mm), AHM = annual heat-moisture index (MAT+10)/(MAP/1000)), SHM = summer heat-moisture index ((MWMT)/(MSP/1000))  Resident soil  (UBC Okanagan) Exogenous soil (Glacier national park) MAT 9.1 1.8 MWMT 20.6 13.3 MCMT -1.6 -7.2 TD 22.3 20.5 MAP 362 1272 MSP 167 414 AHM 52.6 9.3 SHM 123.7 32.1   Table B.3. Results from running a 2-way PERMANOVA, comparing log transformed Bray Curtis distances between bacterial and fungal communities. Results show that for bacteria, there was a significant difference between root communities both based on when they were harvested and what soil they were exposed to, but that there was no significant interaction between harvest and exposure. For fungi, results showed that there was a significant effect of harvest time, but not soil exposure.  Bacteria Fungi  Pseudo-F p Pseudo-F p Harvest 4.38 0.0001 1.88 0.0002 Exposure 2.05 0.006 1.01 0.44 HaXEx 0.97 0.51 1.06 0.33  128   Figure B.1. Comparison of the average proportion of bacterial orders found in plants from different treatments. Orders that made up less than 1% of the total community were grouped as “Other”. The plants compared were perturbed either with a resident- (A1, B1, C1) or an exogenous soil (A2, B2, C2) as seeds (A), one-week-old seedlings (B) or 2-week-old seedlings (C).  42%40% 40%44%32%34%1%1%2%1%1%1%4%5%5%4%12%12%1%1%1%1%10%3% 13% 9%7%6%19%27%21%19%10%13%6%7% 7%13%27% 27%14%9% 11% 10%2% 3%1% 2% 1% 1%2% 2%1% 1% 1% 1%2%0%10%20%30%40%50%60%70%80%90%100%A1 A2 B1 B2 C1 C2OtherTM7-3_I025TM7-3_unknownXanthomonadalesEnterobacterialesBurkholderialesSphingomonadalesRickettsialesRhizobialesCaulobacteralesBacillalesSaprospiralesActinomycetales129   Figure B.2. Comparison of the average proportion of fungal orders found in plants from different treatments. Orders that made up less than 1% of the total community were grouped as “Other”. The plants compared were perturbed either with a resident- (A1, B1, C1) or an exogenous soil (A2, B2, C2) as seeds (A), one-week-old seedlings (B) or 2-week-old seedlings (C).    13%3% 3% 3% 1%1%3%1%6%1% 1%1%1% 1%1%1%1%1%68%75%85%83%89% 87%2%8%1%1% 5%1%1%10% 9%8% 5%6%3%1% 2% 1% 2% 1% 1%0%10%20%30%40%50%60%70%80%90%100%A1 A2 B1 B2 C1 C2OtherUnidentified FungiSordariomycetes_PhyllachoralesSordariomycetes_Incertae_sedisSordariomycetes_HypocrealesSordariomycetes;OtherLeotiomycetes_HelotialesEurotiomycetes_EurotialesDothideomycetes_PleosporalesUnassigned

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
http://iiif.library.ubc.ca/presentation/dsp.24.1-0223809/manifest

Comment

Related Items