- Open Access
Host species identity, site and time drive temperate tree phyllosphere bacterial community structure
© The Author(s). 2016
Received: 13 January 2016
Accepted: 24 May 2016
Published: 18 June 2016
The increasing awareness of the role of phyllosphere microbial communities in plant health calls for a greater understanding of their structure and dynamics in natural ecosystems. Since most knowledge of tree phyllosphere bacterial communities has been gathered in tropical forests, our goal was to characterize the community structure and assembly dynamics of phyllosphere epiphytic bacterial communities in temperate forests in Quebec, Canada. We targeted five dominant tree species: Acer saccharum, Acer rubrum, Betula papyrifera, Abies balsamea, and Picea glauca. We collected 180 samples of phyllosphere communities on these species at four natural forest sites, three times during the growing season.
Host functional traits (i.e., wood density, leaf nitrogen content) and climate variables (summer mean temperature and precipitation) were strongly correlated with community structure. We highlight three key findings: (1) temperate tree species share a “core microbiome”; (2) significant evolutionary associations exist between groups of bacteria and host species; and (3) a greater part of the variation in phyllosphere bacterial community assembly is explained by host species identity (27 %) and species-site interaction (14 %), than by site (11 %) or time (1 %).
We demonstrated that host species identity is a stronger driver of temperate tree phyllosphere bacterial communities than site or time. Our results suggest avenues for future studies on the influence of host functional traits on phyllosphere community functional biogeography across terrestrial biomes.
Microorganisms colonize aerial tree surfaces (i.e., bark, leaves), enabling interactions that are essential for plant growth and fitness [1–3]. Aerial plant surfaces (mostly leaves), a habitat known as the phyllosphere, are estimated to sum up to 4 × 108 km2 on Earth , which is almost equivalent to the total surface of the earth. The phyllosphere habitat is extremely poor in nutrients and exposed to a rapid and pronounced fluctuation of physical conditions . Tree phyllosphere microbial communities are mainly composed of bacteria and endophytic fungi [1, 5]. These communities are extremely diverse [6–9] and contribute to host protection and productivity [10, 11]. Although our knowledge of plant-microbe interactions on tree leaf surfaces is still limited (but see [11, 12] for reviews), most studies have focused on endophytic fungi [8, 13, 14] and pathogens [15, 16] limiting our knowledge of the complex dynamics at play for other organisms. Studies of the tree phyllosphere are more and more frequent, with most studies focusing on tropical forests [17–19].
Bacteria exhibit a wide range of metabolic diversity, which allows them to survive in stressful environments where sources of energy are limited . Although many aspects of phyllosphere bacterial metabolism and functional traits are poorly understood, the first censuses have revealed the presence of anoxygenic phototrophic bacteria . Many bacteria abundant in the phyllosphere, such as Methylobacterium, have been shown to positively influence plant health and development [22, 23] mainly through the production of secondary metabolites interacting with host hormone production and influencing plant growth . While high-throughput sequencing techniques provide more information on plant-bacteria interactions, there is still no clear understanding of host-bacteria association patterns across multiple host species. For example, individual trees have been shown to share part of their dominant bacterial community , yet little is known about this “core” microbiome, the group of bacterial taxa shared among multiple communities sampled from the same habitat . Understanding the drivers of phyllosphere bacterial diversity is the first step toward developing management strategies that encourage a healthy phyllosphere microbial community structure favoring tree health and function.
Phyllosphere bacterial community composition is the result of a combination of dispersal history, host selection [9, 17], growth, and survival in the face of environmental conditions and competition [11, 25]. Hypotheses for the ecological processes structuring phyllosphere communities have included lottery models of colonization , as well as filtering models whereby environmental attributes act as a filter restricting the bacterial taxa that are able to persist on the leaf . Although drivers of phyllosphere microbial assembly have been quantified in previous studies both for fungi [28, 29] and bacteria [9, 27, 30], most of these studies evaluated only a single potential driver of phyllosphere community structure.
In this study, we explore the ecological drivers of variation in leaf bacterial community composition of temperate trees, taking into account the influence of multiple drivers. Our objectives are (1) to identify the epiphytic bacteria present in the phyllosphere of temperate forest trees; (2) to detect the patterns of associations between host taxa and bacteria; and (3) to quantify the relative influence of three drivers on phyllosphere bacterial community composition: host species identity, site, and sampling time. We selected five common temperate tree species present at all sites to obtain a fair representation of Quebec’s temperate forests, including both angiosperms and gymnosperms: Abies balsamea (Balsam fir), Acer rubrum (Red maple), Acer saccharum (Sugar maple), Betula papyrifera (Paper birch), and Picea glauca (White spruce). We collected 180 samples of phyllosphere communities on these species at four natural forest sites (Additional file 1: Table S1, Figure S1), three times during the growing season. Bacterial community structure was determined through High-throughput Illumina sequencing of the bacterial 16S rRNA gene .
Sequences, OTUs, and taxonomy
Sequencing identified 15,873 bacterial operational taxonomic units (OTUs, sequences binned at 97 % similarity) in phyllosphere samples, an average of 517 ± 16 OTUs (mean ± standard error) per tree sampled. Most of these bacterial taxa were rare, with 52.6 % of bacterial OTUs occurring only on a single tree. Each tree sampled revealed additional bacterial taxa as shown by a collector’s curve of the number of OTUs per sample (Additional file 1: Figure S2). Four of the nine most abundant bacterial classes belonged to the phylum Proteobacteria: Alpha- (68 % of all sequences), Beta- (6 %), Gamma- (5 %), and Deltaproteobacteria (3 %); three belonged to the phylum Bacteroidetes: Cytophagia (4 %), Sphingobacteria (1 %), and Saprospirae (1 %); and finally the classes Acidobacteria (6 %) and Actinobacteria (5 %) were also abundant.
Linear models of the relationship between each of core microbiome OTU abundance and the drivers
Model total R2 (%)
Drivers of variation in phyllosphere bacterial community composition and diversity
Bacterial community structure variation explained by various factors (PERMANOVA on Bray-Curtis dissimilarities)
R 2 (%)
Second order interaction
Third order interaction
In terms of the taxonomic composition of phyllosphere communities, temperate leaf communities seem to differ slightly from past reports of tropical and temperate phyllosphere community structure. Natural temperate phyllosphere communities in Quebec forests were dominated by Alphaproteobacteria (68 % of all sequences), contrasting with 27 %  and 22.8 %  of sequences in tropical tree species and 24.5 % in suburban temperate stands . Due to the necessity of using chloroplast-excluding primers to prevent contamination of samples by plant DNA , we were unable to quantify the abundance of Cyanobacteria in the temperate forest phyllosphere. However, metagenomic studies have demonstrated that Cyanobacteria are typically rare in the vascular plant phyllosphere [11, 33], and by using the same chloroplast-excluding 16S primer employed by previous studies [9, 18, 25], we were able to eliminate primer taxonomic bias as an explanation of differences in clade abundances among studies.
In contrast with Redford et al., , we detected the presence of a core phyllosphere microbiome, a group of bacterial taxa shared among multiple communities sampled from the same habitat and thought to play key ecological roles . The core microbiome was composed of 19 OTUs representing 42.7 % of all sequences present in more than 99 % of samples, even when study sites were hundreds of kilometers apart. Assuming that bacterial OTUs represent ecologically or evolutionarily coherent units , this finding suggests that bacteria from a similar metacommunity colonize tree leaves across Quebec’s temperate forests by dispersal through a variety of vectors (i.e., air, rain, soil) , homogenizing the epiphytic phyllosphere community structure across broad geographic distances.
Despite the presence of a core microbiome of abundant taxa, individual trees also showed unique communities that varied predictably across species, sites and time, suggesting a role for selection- or niche-based mechanisms during community assembly. Linear models testing the association between core microbiome OTUs vs. host species identity, site, and time explained 18 to 60 % of the variation in phyllosphere bacterial community structure (Table 1), confirming these three drivers’ roles in shaping phyllosphere community structure. In addition, biomarker analyses confirmed the existence of host selective mechanisms on phyllosphere community structure as shown by associations between numerous bacterial taxa and different host species and sites (Fig. 1).
At the tree species level, Abies balsamea (balsam fir) tended to associate with the order Sphingomonadales, as with the families Acidobacteraceae, Solibacteraceae, and Frankiaceae. The three first groups mentioned above are common in soils [36, 37], and the Frankiaceae are nitrogen-fixing bacteria that colonize plant roots . This finding is in line with other studies showing that conifers select a different microbiome than other plant species: for example they harbor less ice nuclei active bacteria . In contrast, Betula papyrifera (paper birch) was associated with the family Rhodospirillaceae (Rhodospirillales: Alphaproteobacteria). This bacterial family is mostly composed of purple nonsulfur bacteria that produce energy through photosynthesis . Photosynthesis could be a key adaptation to the phyllosphere habitat, an environment where simple carbon sources are scarce and highly variable [1, 11]. Tree-bacteria associations were also observed at the angiosperm vs. gymnosperm level (Fig. 3), likely driven by the influence of the numerous plant functional trait differences between these clades (Lambais et al.  and Kembel et al. ).
Host species identity was the main driver of phyllosphere bacterial community structure among trees (R 2 = 27 %) when compared to site and time. As shown in other studies, each tree species harbors a distinctive phyllosphere bacterial community [9, 17, 41], but our results highlight for the first time the relative influence of site (R 2 = 11 % for site alone and R 2 = 14 % for site species interaction) and time (R 2 = 1 %) for multiple tree species. In accordance with the findings of Kembel et al.  in tropical forests, temperate phyllosphere epiphytic bacterial community structure was correlated with both traits linked to plant-resource uptake strategies such as leaf nitrogen content and leaf mass per area , and traits linked to the wood density/growth/mortality tradeoff such as wood density . This confirms that phyllosphere bacterial communities are shaped by the ecological strategies of their plant hosts. These similarities also suggest that the factors driving the functional biogeography of plant-microbe associations in the phyllosphere are similar across temperate and tropical biomes, as we found a similar set of traits influencing phyllosphere community structure in temperate forests vs. those described for tropical forests . Although many insights have been gained from individual tree microbiome studies in tropical and temperate biomes, meta-analyses controlling for methodological differences will be needed to better understand plant-microbe associations across terrestrial biomes and environmental gradients.
Consistent with the idea of environmental selective pressure on phyllosphere communities due to abiotic conditions such as temperature and precipitation, climate differences between sites (monthly precipitation and mean monthly temperature) were correlated with variation in phyllosphere bacterial community structure. In addition, the effect of sampling time and the interaction between sampling time and site on phyllosphere community structure suggests that phyllosphere communities undergo a succession during the growing season. As previously demonstrated for individual host tree species by Redford and Fierer  for bacterial communities and by Jumpponen and Jones  for fungal communities, leaf communities were temporally dynamic. However, the variance explained by sampling time was small relative to the importance of host species and site, suggesting that once a community of bacteria successfully colonizes a leaf, temporal changes are not enough to overcome the influence of host species identity and site on community assembly. In the temperate forest we studied, growing season had a significant impact on community structure at two sites at the beginning and end of the growing season: the months of June and August. To minimize phyllosphere community structure variation due to sampling time, leaf sampling in these forests should be completed in July once leaves are fully mature but before senescence begins in August.
We found consistent evidence that community composition and alpha diversity differed between coniferous (gymnosperm) vs. broadleaved (angiosperm) tree species. Our results show that several functional traits characteristic of tree ecological strategy explained differences in leaf community structure. However, additional leaf functional traits not measured here (i.e., increased leaf cuticle thickness and wax composition of gymnosperms) could also play a key role by limiting carbon compound availability and humidity at the leaf surface [9, 11]. Because our sampling did not exclusively target the new needles of conifers, a study of succession on conifer needles will be needed to determine if the diversity is due to the particular selective power of the host species or to the longer accumulation through leaf life span of the bacterial community on conifer leaves.
In this study, we describe for the first time natural temperate tree phyllosphere bacterial communities across multiple tree species while exploring the influence of host species identity, site, and time of sampling on phyllosphere community structure. In addition, we performed the first simultaneous evaluation of the importance of key dispersal-related and niche-based drivers such as host species identity (phylogeny, co-evolution, functional traits), geographical location (dispersal history and abiotic conditions) and time of sampling (abiotic conditions) on tree phyllosphere bacterial communities. Our key findings include: (1) that temperate host species share a “core microbiome”; (2) that there are significant associations between groups of bacteria and host species; and finally (3) that a greater part of the variation in phyllosphere bacterial community assembly is explained by host species identity rather than by site or time.
The study plots are located in four natural temperate forest stands in Quebec (Additional file 1: Table S1): Sutton (45° 6' 46" N; 72° 32' 28" W), Abitibi (48° 9' 45" N; 79° 24' 4" W), Gatineau (45° 44' 50" N; 75° 17' 57" W), and Bic (48° 20' 1" N; 68° 49' 3" W). Distances between sites range from 295 km (Sutton and Gatineau) to 765 km (Abitibi and Bic) (Additional file 1: Figure S1). This region is characterized by a cold and humid continental climate with temperate summer. We obtained monthly climate data from Canada’s public weather database  (Additional file 1: Table S1).
Bacterial community collection
We sampled at each site three times during the 2013 growing season (June, July, and August) from three individuals for each tree species, a total of 180 samples. For each randomly chosen tree, we clipped 50–100 g of shade leaves at mid-canopy height (1–2 m above the bottom of the tree’s canopy) into sterile roll bags with surface-sterilized shears. For bacterial community collection and amplification, we used the protocols described by Kembel et al. . We collected microbial communities from the leaf surface by agitating the samples in a diluted Redford buffer solution. We resuspended cells in 500 μL of PowerSoil bead solution (MoBio, Carlsbad, California). We extracted DNA from isolated cells using the PowerSoil kit according to the manufacturer’s instructions and stored at −80 °C.
DNA library preparation and sequencing
We used a two-stage PCR approach to prepare amplicon libraries for the high-throughput Illumina sequencing platform. The use of combinatorial primers for paired-end Illumina sequencing of amplicons reduced the number of primers while maintaining the diversity of unique identifiers . First, to avoid PCR contamination by chloroplast DNA amplification, we targeted the V5–V6 region of the bacterial 16S rRNA gene using cyanobacteria-excluding primers (16S primers 799 F-1115R [9, 25, 46]) following protocols described by Kembel et al. . These chloroplast-excluding primers have been widely used in studies of phyllosphere bacteria in order to avoid contamination by host plant DNA , and their use is justified since while they exclude both plant chloroplasts and cyanobacteria sequences, cyanobacteria are known to be rare in tree phyllosphere communities [11, 33]. Using cleaned PCR product as a template, a second PCR was performed with custom HPLC-cleaned primers to further amplify 16S products and complete the Illumina sequencing construct (PCRII_for: 5′-AAGCAGAAGACGGCATACGAGATCGGTCTCGGCATTCCTGC; PCRII_rev: 5′-ATGATACGGCGACCACCGAGATCTACACTCTTTCCCTACACGACG). We cleaned the resulting product using MoBio UltraClean PCR cleanup kit. We isolated a ~445-bp fragment by electrophoresis in a 2 % agarose gel, and recovered DNA with the MoBio GelSpin kit. We prepared multiplexed 16S libraries by mixing equimolar concentrations of DNA, and sequenced the DNA library using Illumina MiSeq 250-bp paired-end sequencing at Genome Quebec.
We processed the raw sequence data with PEAR  and QIIME  pipelines to merge paired-end sequences to a single sequence of length of approximately 350 bp, eliminate low quality sequences (mean quality score < 30 or with any series of 5 bases with a quality score < 30), and de-multiplex sequences into samples. We eliminated chimeric sequences using the Uclust and Usearch algorithms . Then, we binned the remaining sequences into operational taxonomic units (OTUs) at a 97 % sequence similarity cutoff. We determined the taxonomic identity of each OTU using the BLAST algorithm and Greengenes database  as implemented in QIIME .
Host plant trait data
We obtained data on host plant functional traits (Additional file 1: Table S5) including drought tolerance (Dtol), average maximum height (Hmax), leaf nitrogen mass (Nmass), seed mass (Smass), shade tolerance (Stol), specific leaf area (SLA), and wood density (WD) from a global database collected by Abrams and Kubiske , Burns and Honkala , Farrar , Shipley and Vu , Wright et al. , Niinemets and Valladares , Chave et al. , and USDA .
We tested for the significant associations between bacterial taxa and host species, host taxonomy (angiosperms vs. gymnosperms), and sites using the linear discriminant analysis effect size (LEfSe) algorithm . The LEfSe algorithm aims to discover biomarkers (genes, pathways, or taxa) of different sample groups employing the linear discriminant analysis to approximate the effect size of each biomarker identified. A significant association between bacterial clades and a specific group (i.e., a host tree species) will be detected when there is consistently higher relative abundance of the clade in the group’s samples. Among the bacterial clades detected as statistically and biologically relevant, the strongest scores identify which clades have the greatest explanatory power for differences between communities .
Because PCR and sequencing errors could lead to spurious OTU identification , we created a database excluding OTUs represented by less than 20 sequences to eliminate rare OTUs. Analyses were performed on both the full database and the database with rare OTUs excluded to assess the results’ sensibility to rarefaction. The number of sequences per sample ranged from 4574 to 86,280. From a database of 3,868,892 quality sequences, we rarefied each sample to 4000 sequences, with 38 samples excluded from subsequent analyses due to insufficient sequence reads as a result of extraction or sequencing errors, totalizing 668,000 sequences from 142 samples representing five tree species. Rarefaction and all subsequent statistical analyses were repeated 100 times. Results did not differ qualitatively across iterations of the rarefaction and we therefore present only the result of a single random rarefaction. We performed analyses with the ape , ggplot2 , picante , and vegan  packages in R .
We quantified the phylogenetic variation in bacterial community structure among samples with the weighted UniFrac index, an abundance-weighted measure of the phylogenetic differentiation among bacterial communities . To illustrate patterns of bacterial community structure, we performed a non-metric multidimensional scaling (NMDS) ordination of Bray-Curtis dissimilarities and weighted UniFrac distances among all samples. We identified relationships between bacterial community structure, host species identity, time, and site by conducting a permutational multivariate analysis of variance (PERMANOVA, ) on the community matrix. We identified functional traits and climate variables that are significant drivers of leaf community structure through a PERMANOVA. We employed a blocking randomization to account for the non-independence of observations across species and sites. The functional trait PERMANOVA was blocked by site and the climate variable PERMANOVA was blocked by species to correct for the absence of intra-site and intra-specific variation in our trait and climate data. To visualize the changes in bacterial communities with respect to different variables, we tested for correlations between these variables and community scores on the NMDS ordination axes while applying the Bonferroni correction for multiple comparisons to our significance threshold [67, 68]. The cutoffs for significant correlations (α = 0.05) were adjusted to p < 0.007 (functional traits) and p < 0.025 (climate data). To quantify the influence of host taxonomic levels on bacterial community structure, we performed a nested PERMANOVA (levels: angiosperm/gymnosperm, family, genus, species).
We estimated phyllosphere bacterial alpha diversity using the Shannon index calculated from OTU relative abundances for each community. We performed an analysis of variance (ANOVA) and subsequent post-hoc Tukey’s tests to test for differences in diversity across species, time, and site. To account for the repeated measures taken on individual trees in our data, we constructed a linear mixed model fitted by maximum likelihood. This model aimed to estimate the power of tree identity as a random factor in driving microbial community diversity in comparison with host species identity, site and, sampling time.
ANOVA analysis of variance, D tol drought tolerance, H max average maximum height, LEfSe linear discriminant analysis effect size, N mass , leaf nitrogen mass, NMDS non-metric multidimensional scaling, OTU operational taxonomic unit, PERMANOVA permutational analysis of variance, SLA specific leaf area, S mass seed mass, S tol shade tolerance, WD wood density
We thank Travis Dawson, Sophie Carpentier, and Gabriel Jacques for support in the field and laboratory.
Financial support was obtained from grants attributed by the Natural Sciences and Engineering Research Council of Canada (NSERC), the Fonds de Recherche du Québec - Nature et Technologies (FRQNT), and by the Canada Research Chairs Program.
Availability of the data and materials
The datasets supporting the conclusions of this article are available in the Natural Temperate Forest project repository on Figshare at: https://figshare.com/s/be5c5ed6b224667e8fd1 (barcodes), https://figshare.com/s/20517d2ebeb81de97dd5 (metadata) and https://figshare.com/s/b93099c2ea930c5d6553 (16S rRNA sequences).
ILL, CM, and SK designed the study. ILL and SK analyzed the data. ILL, CM, and SK wrote the manuscript. All authors read and approved the final manuscript.
The authors declare that the experiment complies with the current laws of the country in which the experiment was performed. The authors declare that they have no competing interests.
Ethics approval and consent to participate
No ethics committee required as there were no human or animal subjects involved.
Open AccessThis article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.
- Lindow SE, Brandl MT. Microbiology of the phyllosphere. Appl Environ Microbiol. 2003;69:1875–83.View ArticlePubMedPubMed CentralGoogle Scholar
- Herre EA, Mejía LC, Kyllo DA, Rojas E, Maynard Z, Butler A, et al. Ecological implications of anti-pathogen effects of tropical fungal endophytes and mycorrhizae. Ecology. 2007;88:550–8.View ArticlePubMedGoogle Scholar
- Fürnkranz M, Wanek W, Richter A, Abell G, Rasche F, Sessitsch A. Nitrogen fixation by phyllosphere bacteria associated with higher plants and their colonizing epiphytes of a tropical lowland rainforest of Costa Rica. Isme J. 2008;2:561–70.View ArticlePubMedGoogle Scholar
- Morris CE, Kinkel LL, Lindow SE, Hecht-Poinar EI, Elliott VJ. Fifty years of phyllosphere microbiology: significant contributions to research in related fields. In: Lindow SE, Hecht-Poinar EI, Elliott VJ, editors. Phyllosphere microbiology. 2002. p. 365–75.Google Scholar
- Andrews JH, Harris RF. The ecology and biogeography of microorganisms on plant surfaces. Annu Rev Phytopathol. 2000;38:145–80.View ArticlePubMedGoogle Scholar
- Lambais MR, Crowley DE, Cury JC, Büll RC, Rodrigues RR. Bacterial diversity in tree canopies of the Atlantic forest. Science. 2006;312:1917.View ArticlePubMedGoogle Scholar
- Jumpponen A, Jones KL. Massively parallel 454 sequencing indicates hyperdiverse fungal communities in temperate Quercus macrocarpa phyllosphere. New Phytol. 2009;184:438–48.View ArticlePubMedGoogle Scholar
- Rodriguez RJ, White Jr JF, Arnold AE, Redman RS. Fungal endophytes: diversity and functional roles. New Phytol. 2009;182:314–30.View ArticlePubMedGoogle Scholar
- Redford AJ, Bowers RM, Knight R, Linhart Y, Fierer N. The ecology of the phyllosphere: geographic and phylogenetic variability in the distribution of bacteria on tree leaves. Environ Microbiol. 2010;12:2885–93.View ArticlePubMedPubMed CentralGoogle Scholar
- Arnold AE, Mejía LC, Kyllo D, Rojas EI, Maynard Z, Robbins N, Herre EA. Fungal endophytes limit pathogen damage in a tropical tree. P Natl Acad Sci. 2003;100:15649–54.View ArticleGoogle Scholar
- Vorholt JA. Microbial life in the phyllosphere. Nat Rev Microbiol. 2012;10:828–40.View ArticlePubMedGoogle Scholar
- Müller T, Ruppel S. Progress in cultivation-independent phyllosphere microbiology. FEMS Microbiol Ecol. 2014;87:2–17.View ArticlePubMedGoogle Scholar
- Osono T. Role of phyllosphere fungi of forest trees in the development of decomposer fungal communities and decomposition processes of leaf litter. Can J Microbiol. 2006;52:701–16.View ArticlePubMedGoogle Scholar
- Suda W, Nagasaki A, Shishido M. Powdery mildew-infection changes bacterial community composition in the phyllosphere. Microb Environ. 2009;24:217–23.View ArticleGoogle Scholar
- Gilbert GS. Evolutionary ecology of plant diseases in natural ecosystems. Annu Rev Phytopathol. 2002;40:13–43.View ArticlePubMedGoogle Scholar
- Newton AC, Gravouil C, Fountaine JM. Managing the ecology of foliar pathogens: ecological tolerance in crops. Ann Appl Biol. 2010;157:343–59.View ArticleGoogle Scholar
- Kim M, Singh D, Lai-Hoe A, Go R, Rahim RA, Ainuddin AN, et al. Distinctive phyllosphere bacterial communities in tropical trees. Microb Ecol. 2012;63:674–81.View ArticlePubMedGoogle Scholar
- Kembel SW, O’Connor TK, Arnold HK, Hubbell SP, Wright SJ, Green JL. Relationships between phyllosphere bacterial communities and plant functional traits in a neotropical forest. P Natl Acad Sci. 2014;111:13715–20.View ArticleGoogle Scholar
- Kembel SW, Mueller RC. Plant traits and taxonomy drive host associations in tropical phyllosphere fungal communities. Botany. 2014;92:303–11.View ArticleGoogle Scholar
- Mercier J, Lindow SE. Role of leaf surface sugars in colonization of plants by bacterial epiphytes. Appl Environ Microbiol. 2000;66:369–74.View ArticlePubMedPubMed CentralGoogle Scholar
- Atamna-Ismaeel N, Finkel O, Glaser F, von Mering C, Vorholt JA, Koblížek M, et al. Bacterial anoxygenic photosynthesis on plant leaf surfaces. Environ Microbiol Rep. 2012;4:209–16.View ArticlePubMedGoogle Scholar
- Abanda-Nkpwatt D, Müsch M, Tschiersch J, Boettner M, Schwab W. Molecular interaction between Methylobacterium extorquens and seedlings: growth promotion, methanol consumption, and localization of the methanol emission site. J Exp Bot. 2006;57:4025–32.View ArticlePubMedGoogle Scholar
- Innerebner G, Knief C, Vorholt JA. Protection of Arabidopsis thaliana against leaf-pathogenic Pseudomonas syringae by Sphingomonas strains in a controlled model system. Appl Environ Microbiol. 2011;77:3202–10.View ArticlePubMedPubMed CentralGoogle Scholar
- Shade A, Handelsman J. Beyond the Venn diagram: the hunt for a core microbiome. Environ Microbiol. 2012;14:4–12.View ArticlePubMedGoogle Scholar
- Redford AJ, Fierer N. Bacterial succession on the leaf surface: a novel system for studying successional dynamics. Microb Ecol. 2009;58:189–98.View ArticlePubMedGoogle Scholar
- Burke C, Thomas T, Lewis M, Steinberg P, Kjelleberg S. Composition, uniqueness and variability of the epiphytic bacterial community of the green alga Ulva australis. Isme J. 2011;5:590–600.View ArticlePubMedGoogle Scholar
- Knief C, Ramette A, Frances L, Alonso-Blanco C, Vorholt JA. Site and plant species are important determinants of the Methylobacterium community composition in the plant phyllosphere. Isme J. 2010;4:719–28.View ArticlePubMedGoogle Scholar
- Osono T. Endophytic and epiphytic phyllosphere fungi of Camellia japonica: seasonal and leaf age-dependent variations. Mycologia. 2008;100:387–91.View ArticlePubMedGoogle Scholar
- Cordier T, Robin C, Capdevielle X, Desprez-Loustau ML, Vacher C. Spatial variability of phyllosphere fungal assemblages: genetic distance predominates over geographic distance in a European beech stand (Fagus sylvatica). Fungal Ecol. 2012;5:509–20.View ArticleGoogle Scholar
- Finkel OM, Burch AY, Lindow SE, Post AF, Belkin S. Geographical location determines the population structure in phyllosphere microbial communities of a salt-excreting desert tree. Appl Environ Microbiol. 2011;77:7647–55.View ArticlePubMedPubMed CentralGoogle Scholar
- Claesson MJ, Wang Q, O'Sullivan O, Greene-Diniz R, Cole JR, Ross RP, et al. Comparison of two next-generation sequencing technologies for resolving highly complex microbiota composition using tandem variable 16S rRNA gene regions. Nucleic Acids Res. 2010;38(22):e200-e200.Google Scholar
- Rastogi G, Tech JJ, Coaker GL, Leveau JH. A PCR-based toolbox for the culture-independent quantification of total bacterial abundances in plant environments. J Microbiol Methods. 2010;83(2):127–32.View ArticlePubMedGoogle Scholar
- Delmotte N, Knief C, Chaffron S, Innerebner G, Roschitzki B, Schlapbach R, et al. Community proteogenomics reveals insights into the physiology of phyllosphere bacteria. Proc Natl Acad Sci U S A. 2009;106:16428–33. doi:10.1073/pnas.0905240106.View ArticlePubMedPubMed CentralGoogle Scholar
- Schmidt TS, Rodrigues JFM, Von Mering C. Ecological consistency of SSU rRNA-based operational taxonomic units at a global scale. PLoS Comput Biol. 2014;10(4):e1003594.View ArticlePubMedPubMed CentralGoogle Scholar
- Bulgarelli D, Schlaeppi K, Spaepen S, van Themaat EVL, Schulze-Lefert P. Structure and functions of the bacterial microbiota of plants. Annu Rev Plant Biol. 2012;64:807–38.View ArticleGoogle Scholar
- Janssen PH. Identifying the dominant soil bacterial taxa in libraries of 16S rRNA and 16S rRNA genes. Appl Environ Microbiol. 2006;72:1719–28.View ArticlePubMedPubMed CentralGoogle Scholar
- Kim M, Heo E, Kang H, Adams J. Changes in soil bacterial community structure with increasing disturbance frequency. Microb Ecol. 2013;66:171–81.View ArticlePubMedGoogle Scholar
- Normand P. The families Frankiaceae, Geodermatophilaceae, Acidothermaceae and Sporichthyaceae. In: Starr MP, Stolp H, Trüper HG, Balows A, Schlegel HG, editors. The prokaryotes. New York: Springer; 2006. p. 669–81.View ArticleGoogle Scholar
- Lindow SE, Arny DC, Upper CD. Distribution of ice nucleation-active bacteria on plants in nature. Appl Environ Microbiol. 1978;36:831–8.PubMedPubMed CentralGoogle Scholar
- Biebl H, Pfennig N. Isolation of members of the family Rhodospirillaceae. In: Starr MP, Stolp H, Trüper HG, Balows A, Schlegel HG, editors. The prokaryotes. Berlin Heidelberg: Springer; 1981. p. 267–73.View ArticleGoogle Scholar
- Lambais MR, Lucheta AR, Crowley DE. Bacterial community assemblages associated with the phyllosphere, dermosphere, and rhizosphere of tree species of the atlantic forest are host taxon dependent. Microb Ecol. 2014;68:567–74.View ArticlePubMedGoogle Scholar
- Wright IJ, Reich PB, Westoby M, Ackerly DD, Baruch Z, Bongers F, et al. The worldwide leaf economics spectrum. Nature. 2004;428:821–7.View ArticlePubMedGoogle Scholar
- Wright SJ, Kitajima K, Kraft NJ, Reich PB, Wright IJ, Bunker DE, et al. Functional traits and the growth-mortality trade-off in tropical trees. Ecology. 2010;91:3664–74.View ArticlePubMedGoogle Scholar
- Canada Weather Database. climate.weatheroffice.gc.ca. Accessed 1 Aug 2014.
- Gloor GB, Hummelen R, Macklaim JM, Dickson RJ, Fernandes AD, MacPhee R, et al. Microbiome profiling by illumina sequencing of combinatorial sequence-tagged PCR products. PLoS One. 2010;5:e15406.View ArticlePubMedPubMed CentralGoogle Scholar
- Chelius MK, Triplett EW. The diversity of archaea and bacteria in association with the roots of Zea mays L. Microb Ecol. 2001;41:252–63.View ArticlePubMedGoogle Scholar
- Zhang J, Kobert K, Flouri T, Stamatakis A. PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics. 2014;30:614–20.View ArticlePubMedGoogle Scholar
- Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, et al. QIIME allows analysis of high-throughput community sequencing data. Nat Methods. 2010;7:335–6.View ArticlePubMedPubMed CentralGoogle Scholar
- Edgar RC. Search and clustering orders of magnitude faster than BLAST. Bioinformatics. 2010;26:2460–1.View ArticlePubMedGoogle Scholar
- DeSantis TZ, Hugenholtz P, Larsen N, Rojas M, Brodie EL, Keller K, Huber T, Dalevi D, Hu P, Andersen GL. Greengenes, a chimera-checked 16S rRNA gene database and workbench compatible with ARB. Appl Environ Microbiol. 2006;72:5069–72.View ArticlePubMedPubMed CentralGoogle Scholar
- Abrams MD, Kubiske ME. Leaf structural characteristics of 31 hardwood and conifer tree species in central Wisconsin: influence of light regime and shade-tolerance rank. Forest Ecol Manag. 1990;31:245–53.View ArticleGoogle Scholar
- Burns RM, Honkala BH. Silvics of North America: 1. Conifers; 2. Hardwoods. Washington, DC: Agriculture handbook 654, USDA For. Serv.; 1990.Google Scholar
- Farrar JL. Trees in Canada. Québec: Les Editions Fides; 1996.Google Scholar
- Shipley B, Vu TT. Dry matter content as a measure of dry matter concentration in plants and their parts. New Phytol. 2002;153:359–64.View ArticleGoogle Scholar
- Niinemets Ü, Valladares F. Tolerance to shade, drought, and waterlogging of temperate Northern Hemisphere trees and shrubs. Ecol Monogr. 2006;76:521–47.View ArticleGoogle Scholar
- Chave J, Coomes D, Jansen S, Lewis SL, Swenson NG, Zanne AE. Towards a worldwide wood economics spectrum. Ecol Lett. 2009;12:351–66.View ArticlePubMedGoogle Scholar
- USDA NRCS. The PLANTS Database. Baton Rouge: National Plant Data Center; 2009. http://plants.usda.gov.Google Scholar
- Segata N, Izard J, Waldron L, Gevers D, Miropolsky L, Garrett WS, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12:R60.View ArticlePubMedPubMed CentralGoogle Scholar
- Acinas SG, Sarma-Rupavtarm R, Klepac-Ceraj V, Polz MF. PCR-induced sequence artifacts and bias: insights from comparison of two 16S rRNA clone libraries constructed from the same sample. Appl Environ Microbio. 2005;71:8966–9.View ArticleGoogle Scholar
- Paradis E, Claude J, Strimmer K. APE: analyses of phylogenetics and evolution in R language. Bioinformatics. 2004;20:289–90.View ArticlePubMedGoogle Scholar
- Wickham H. ggplot2: elegant graphics for data analysis. New York: Springer; 2009.View ArticleGoogle Scholar
- Kembel SW, Cowan PD, Helmus MR, Cornwell WK, Morlon H, Ackerly DD, et al. Picante: R tools for integrating phylogenies and ecology. Bioinformatics. 2010;26:1463–4.View ArticlePubMedGoogle Scholar
- Oksanen J, Kindt R, Legendre P, O’Hara B, Stevens MHH, Oksanen MJ, et al. The vegan package. Community ecology package. 2007.Google Scholar
- R Development Core Team. R: A language and Environment for Statistical Computing. Vienna, Austria. 2013. URL http://www.r-project.org/.Google Scholar
- Lozupone C, Hamady M, Knight R. UniFrac—an online tool for comparing microbial community diversity in a phylogenetic context. BMC Bioinformatics. 2006;7:371.View ArticlePubMedPubMed CentralGoogle Scholar
- Anderson MJ. A new method for non-parametric multivariate analysis of variance. Aust Ecol. 2001;26:32–46.Google Scholar
- Hochberg Y. A sharper Bonferroni procedure for multiple tests of significance. Biometrika. 1988;75(4):800–2.View ArticleGoogle Scholar
- Bland JM, Altman DG. Multiple significance tests: the Bonferroni method. BMJ. 1995;310(6973):170.View ArticlePubMedPubMed CentralGoogle Scholar