Europe PMC

This website requires cookies, and the limited processing of your personal data in order to function. By using the site you are agreeing to this as outlined in our privacy notice and cookie policy.

Mainland and island populations of Mussaenda kwangtungensis differ in their phyllosphere fungal community composition and network structure.

Author information

Affiliations

  • 1. College of Life Sciences, Fujian Agriculture and Forestry University, Fuzhou, 350002, China.
      Authors
    • Qian X 1, 2
    (1 author)
  • 2. Key Laboratory of Plant Resources Conservation and Sustainable Utilization, South China Botanical Garden, Chinese Academy of Sciences, Guangzhou, 510650, China.
      Authors
    • Qian X 1, 2
    • Li S 2
    • Cai H 2
    • Shi M 2
    • Zhang D 2
    (5 authors)
  • 3. University of Chinese Academy of Sciences, Beijing, 100049, China.
      Authors
    • Wu B 3
    • Wang Y 3
    • Ji N 3
    • Yao H 3
    (4 authors)

Scientific Reports, 22 Jan 2020, 10(1):952
https://doi.org/10.1038/s41598-020-57622-6 PMID: 31969602 PMCID: PMC6976661

This article is in the Europe PMC Open access subset. Refer to the copyright information in the article for licensing details.
Free full text in Europe PMC

Abstract 

We compared community composition and co-occurrence patterns of phyllosphere fungi between island and mainland populations within a single plant species (Mussaenda kwangtungensis) using high-throughput sequencing technology. We then used 11 microsatellite loci for host genotyping. The island populations differed significantly from their mainland counterparts in phyllosphere fungal community structure. Topological features of co-occurrence network showed geographic patterns wherein fungal assemblages were less complex, but more modular in island regions than mainland ones. Moreover, fungal interactions and community composition were strongly influenced by the genetic differentiation of host plants. This study may advance our understanding of assembly principles and ecological interactions of phyllosphere fungal communities, as well as improve our ability to optimize fungal utilization for the benefit of people.

Free full text 

Logo of scirepAboutEditorial BoardFor AuthorsScientific Reports
Sci Rep. 2020; 10: 952.
Published online 2020 Jan 22. https://doi.org/10.1038/s41598-020-57622-6
PMCID: PMC6976661
PMID: 31969602

Mainland and island populations of Mussaenda kwangtungensis differ in their phyllosphere fungal community composition and network structure

Xin Qian,1,2 Shengchun Li,2 Binwei Wu,3 Yonglong Wang,3 Niuniu Ji,3 Hui Yao,3 Hongyue Cai,2 Miaomiao Shi,corresponding author2,4 and Dianxiang Zhangcorresponding author2
Author information Article notes Copyright and License information Disclaimer

Associated Data

Supplementary Materials

Abstract

We compared community composition and co-occurrence patterns of phyllosphere fungi between island and mainland populations within a single plant species (Mussaenda kwangtungensis) using high-throughput sequencing technology. We then used 11 microsatellite loci for host genotyping. The island populations differed significantly from their mainland counterparts in phyllosphere fungal community structure. Topological features of co-occurrence network showed geographic patterns wherein fungal assemblages were less complex, but more modular in island regions than mainland ones. Moreover, fungal interactions and community composition were strongly influenced by the genetic differentiation of host plants. This study may advance our understanding of assembly principles and ecological interactions of phyllosphere fungal communities, as well as improve our ability to optimize fungal utilization for the benefit of people.

Subject terms: Microbial ecology, Fungal ecology

Introduction

The phyllosphere provides a unique habitat for a diverse community of microorganisms that live within or on the surface of leaves1. Phyllosphere fungi have been considered pivotal determinants of host-plant fitness, productivity, and ecosystem functioning2,3. For example, phyllosphere fungi may confer salt, heat or herbivory tolerance to host plants4–6, they may potentially be a source of plant diseases due to the pathogenic phases in their life cycles7,8, or they may act as initial decomposers of leaf litter and play important roles in nutrient cycling9,10. With advances in high-throughput sequencing technologies where DNA samples can be extracted directly from leaves, phyllosphere fungal communities have been examined to determine the community composition and potential drivers of this high diversity11. Yang et al.12 found that leaf carbon was the main driver of changes for the foliar fungal community composition of Betula ermanii along a subalpine timberline. Zimmerman et al.13 showed that climate factors such as temperature and rainfall strongly structured fungal communities in the leaves of Metrosideros polymorpha across a Hawaiian landscape. A different study conducted in tropical lowland rainforests focused on the factors shaping the community structure of phyllosphere fungi and found that plant traits and taxonomy were critical factors14. Moreover, several studies have explored the relationship between the genetic identity of a conspecific host and the community composition of its phyllosphere fungi, but no consensus was established. Variation in phyllosphere fungal community composition in Fagus sylvatica and Mussaenda pubescens var. alba was mostly explained by host genotype15,16. In contrast, plant genotype had no significant effect on the phyllosphere fungal microbiome of Picea glauca17.

Microorganisms often vary across trophic modes and functionally distinct niches, which allows them to co-exist and to form complex ecological networks comprising microbial members that interact with each other18,19. A comprehensive exploration of co-occurrence networks is critical to understanding the assembly process and function of the microbial community, because it provides insights into potential relationships and reveals niche spaces. Further, exploration may uncover possible keystone taxa that exert a considerable impact on community structure and function, regardless of their abundance20,21. Co-occurrence network analysis based on high-throughput sequencing data has been increasingly applied to examine the ecological interactions among microorganisms in different habitats including water environments22–24, human and animal digestive tracts25,26, soil27,28, and tree trunks29. Nevertheless, compared to bacterial communities, far fewer studies have been conducted to investigate co-occurrence patterns of fungal microbiome especially in plant associated communities. Recently, Bouffaud et al.30 demonstrated that across the European regions, closely-related arbuscular mycorrhizal fungi (AMF) tended to co-occur with a significantly higher probability than distantly-related ones. Qian et al.31 found a trend of reduced connectivity in phyllosphere fungal co-occurrence networks with increasing elevation. However, we still lack fundamental knowledge concerning factors that are predictive of differences among network structures.

Allopatric isolation on islands has long been considered a driver of evolutionary diversification32,33. Moreover, groups of islands may act conveniently as replicates in which general evolutionary patterns can be distinguished from unique outcomes34. Many studies have presented the idea that island populations are inclined to display a significant genetic separation from corresponding species on the mainland due to their small population size, founder effects and limited immigration35. For instance, geographical isolation by sea may have effectively led to genetic differentiation in Weigela coraeensis between the Honshu mainland and the Izu islands, although gene flow was still occurring between the mainland and the islands33. Remarkable genetic differentiation was also observed in Periploca laevigata, albeit the overall genetic diversity in plant taxa did not differ significantly between mainland and island populations36. In addition, the degree of genetic divergence also depends on when plants emigrated to the islands, on the distance from island to mainland, and on biological traits, including seed dispersal ability and reproductive systems33. Islands can also be natural laboratories for studying community assembly and the processes that shape species distributions and interactions37. However, most island biogeography studies have focused on macro-organisms, so information about the distribution pattern of plant associated fungal microbiomes is quite limited. Recently, Davison et al.38 found that island AMF communities consist of few endemic taxa, but are as diverse as mainland ones. This indicates that efficient dispersal outweighs potential negative effects from taxogenesis and extinction on islands. Whereas da Silva et al.39 showed that AMF community assemblages differed significantly between island and mainland environments due to climatic differences, as well as edaphic and spatial factors. Nonetheless, the island and mainland patterns of phyllosphere fungal distribution have seldom been studied, and thus the relationships between their community structure and host plant genetic differentiation are poorly characterized.

Mussaenda kwangtungensis H. L. Li (Rubiaceae) is a drought-resistant shrub, 1–2.5 m high, distributed across mainland and islands in Guangdong Province, China; This plant and this location present an ideal study system to compare distribution patterns of phyllosphere fungal communities and host genetic differentiation between island and mainland populations. M. kwangtungensis has historically been used in Chinese traditional herbal medicine as an antichloristic and antipyretic agent against laryngopharyngitis, acute gastroenteritis and dysentery40. Our prior efforts have investigated its reproductive characteristics as well as its population genetics40,41. However, little is known about the distribution pattern of phyllosphere fungal microbiome associated with this plant species. In this study, we used high-throughput sequencing of fungal ribosomal internal transcribed spacer 2 (ITS2) on the Illumina MiSeq platform to analyze the phyllosphere fungal community structure and the co-occurrence network characteristics of island and mainland M. kwangtungensis populations. Then, we also examined the host population genetic structure and plant genotyping using 11 nuclear DNA microsatellite loci. The main aims of this study were to: (1) investigate whether the community composition of the island phyllosphere fungi dffered significantly from their mainland counterparts; (2) explore the co-occurrence patterns of phyllosphere fungal microbiomes in island and mainland populations, respectively; (3) evaluate the importance of host genetic structure on variation in the fungal community composition and co-occurrence network structure of a single plant species with an island-mainland distribution.

Material and Methods

Study site and sampling

This study was conducted at three island and three mainland sites having Mussaenda kwangtungensis populations in Guangdong Province, South China (Fig. 1, Table S1). At each sampling site, we recorded the geographic coordinates (latitude and longitude) and elevation with a handheld GPS unit. Mean annual temperature (MAT) and mean annual precipitation (MAP) of each population were obtained from the WorldClim global climate data set with a high spatial resolution of 30 seconds (http://worldclim.org/version2).

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig1_HTML.jpg

Map of the study region and the results of genetic clustering based on DNA microsatellite data.

Ten mature and healthy individuals (>20 m away from each other) were selected randomly for sampling at each site during anthesis, which brought the total of 60 individuals across all sites. With scissors and while wearing sterilized gloves, we collected nine asymptomatic sun leaves from three current-year shoots growing in different directions from each plant. Samples were then packed into sterile polyethylene bags containing sufficient silica gel (>300 g) for quick drying.

DNA library preparation and MiSeq sequencing

Leaves were pooled at the individual plant level and were then pulverized mixed in liquid nitrogen using sterilized pestles and mortars under aseptic conditions in a laminar airflow to avoid external contamination. We extracted DNA from approximately 500 mg of leaf powder using the PowerSoil DNA Isolation kit (MoBio Laboratories, CA, USA) according to the manufacturer’s instructions. DNA quality and quantity were measured on a NanoDrop 1000 spectrophotometer (Thermo Scientific, Wilmington, USA).

A two-step PCR procedure was used to prepare amplicon libraries for the Illumina Miseq sequencing platform. First, the entire fungal internal transcribed spacer (ITS) region was amplified using the conventional primers ITS1F42 and ITS443. PCR amplification was performed in a 25 μL reaction with 0.5 μM of each primer, 2.5 µL Phusion HF Buffer, c. 10 ng DNA template, 250 μM of dNTP Mix, and 1 Unit Phusion DNA polymerase (New England Biolabs, Hitchin, UK). The PCR thermal profile was as follows: initial denaturation at 94 °C for 5 min, 20 cycles of 94 °C for 1 min, 56 °C for 50 s and 68 °C for 30 s, and a final extension at 68 °C for 10 min. The PCR products were then diluted 50 times with sterile deionized water, and 1.0 μL of the resulting solution was then used as a template for the second PCR round to ITS2 region, using the same conditions except for primers fITS744 and ITS4 linked with 12 base pair (bp) barcode sequences. Thermocycling profile was also identical to those used in the first PCR round. To minimize PCR biases, each sample was amplified in triplicate and the three replicate products were grouped together into a general sample. The entire PCR product was separated on a 1.5% agarose gel and the target fragment was excised and purified using a MinElute PCR Purification Kit (Qiagen, Venlo, the Netherlands). Then, the DNA samples were pooled and concentrated using the DNA Clean and Concentrator kit (Zymo Research, Orange, CA, USA) and adjusted to 10 ng μL−1. The sequencing library was constructed with the Illumina sequencing adaptor using the Illumina TruSeq DNA PCR-Free LT Library Prep Kit (Illumina, CA, USA) according to the manufacturer’s instructions. The library was applied to an Illumina MiSeq PE250 platform for sequencing using the paired end (2 × 250 bp) option, and was conducted at the Environmental Genomic Platform of Chengdu Institute of Biology, Chinese Academy of Sciences, China. The sequences have been submitted to the NCBI Sequence Read Archive (SRA) with accession number SRP226848 and PRJNA579231.

Microsatellite genotyping

The genomic DNA was also used for host plant genetic analysis. Microsatellite loci have been developed by Duan et al.45. Eleven loci showed polymorphisms in M. kwangtungensis populations (Table S2). Each forward primer was fluorescently labelled with TAMRA, FAM, HEX, or ROX (Invitrogen, CA, USA) for further screening. The PCR was carried out in a 10 μl reaction solution including 35 ng of genomic DNA, 3.5 μL deionized water, 0.2 μM of each primer pair and 5 μL PCR reaction Mix (Tiangen, Guangzhou, China). The cycling conditions for the amplification were set at 94 °C for 5 min, then 35 cycles of denaturation at 94 °C for 30 s, 30 s at annealing temperature (the value of each primer is shown in Table S2), and then extension at 72 °C for 1 min, with a final extension of 8 min at 72 °C. In addition, a touchdown PCR program was used for four primers (AC30, CT113, CT135, and CT142) with initial denaturation for 5 min at 94 °C, 7 touchdown cycles of 94 °C for 30 s, 30 s at 60 °C with decreasing annealing temperatures in decrements of 1 °C per cycle and 72 °C for 60 s, and then 28 cycles of 94 °C for 30 s, 53 °C for 30 s, and 72 °C for 60 s. The final extension was done at 72 °C for 8 min. PCR products were separated by an ABI PRISM 3100 Genetic Analyser (Applied Biosystems, CA, USA). Genotyping data was obtained from allele binning and calling using GeneMarker ver. 2.4.0 (SoftGenetics LLC, Pennsylvania, USA).

Host population genetic structure

Observed heterozygosity (Ho) and hierarchical analysis of molecular variance (AMOVA) were estimated to quantify the genetic diversity and assess genetic differentiation among populations from the two geographical regions (the mainland and the islands) using Arlequin ver. 3.546. Nei genetic distance47,48, and principal coordinates analysis (PCoA) were determined with the GenAlEx 6.50249. Nei genetic distance was then employed to construct a UPGMA dendrogram to visualize the genetic relationships among the populations. Isolation-by-distance (IBD) effects were tested based on pairwise genetic and geographic distances at the population level by using the “ecodist” R package50. Population structure was explored using the program STRUCTURE 2.3.451. Ten replicates of an admixture model assuming genetic groups ranging from 1 to 6 with admixture and dependent allele frequencies were performed with 100 000 Markov Chain Monte Carlo (MCMC) burn-in steps followed by 50 000 iterations. Program STRUCTURE HARVESTER52 was then used to determine the optimal number of clusters (K). The results of the replicates at the best number of clusters were combined and analyzed with the Full Search algorithm using CLUMPP53. Finally, the corresponding Q matrices were visualized by DISTRUCT program54.

Bioinformatics

The raw sequence data was filtered using PEAR55 and QIIME56 programs to merge paired-end sequences, to remove low quality sequences, and to demultiplex all sequences into each sample. The ITS2 region was detected and extracted using the fungal ITSX software package57, and de novo chimera detection and deletion were performed in the UCHIME algorithm58. The remaining sequences were then binned into operational taxonomic unites (OTUs) at a 97% sequence similarity level by employing the UPARSE pipeline59 after dereplication and discarding all singletons. A representative sequence (the most abundant) of each OTU was selected and searched against the UNITE database using a BLAST algorithm to determine the taxonomic identity of fungal OTUs60. Ten best-matching references were considered to taxonomically annotate OTUs as accurately as possible. The results of the BLASTn search were considered reliable for robustly assigning sequences to fungal Kingdom if e-values < e-50. Therefore those with > e-50 were eliminated. Sequence identities of 90%, 85%, 80% and 75% were used as criteria for assigning OTUs to the taxonomic levels of genus, family, order and class, respectively60. The OTUs belonging to mycorrhizal fungi or animal pathogens were excluded because they might be contaminated. The OTUs with less than 10 reads were then excluded in all the samples because their sequences might have contained PCR or sequencing errors61. To eliminate the effects of sequence number variation from the different samples prior to downstream analysis, the number of sequences was rarifed to the minimum sequencing depth of each sample using MOTHUR62 through a subset of randomly selected reads.

Statistical analysis of fungal community composition

Most statistical analyses were conducted in R v.3.3.363. Rarefaction curves for all phyllosphere fungi at island and mainland regions were computed to evaluate the comprehensiveness of the sampling strategy using the “vegan” R package64. After rarefying based on the smallest read number from any individual sample (5415 reads per sample), 324900 reads remained, representing 772 fungal OTUs in the phyllosphere of M. kwangtungensis populations. Because of the rejection of homoscedasticity by the Levene’s test for equal variance, the non-parametric two-sample Wilcoxon test65 was used to compare the OTU richness between mainland and island populations. Simple linear regression analysis was conducted to test the relationship between fungal OTU richness and plant genetic diversity. The distance matrix for the phyllosphere fungal community composition (Hellinger-transformation of the OTU read data) was constructed by calculating dissimilarity with the Bray-Curtis method66. The Bray-Curtis matrix was used to perform hierarchical clustering analysis of fungal communities across different plant populations. To investigate patterns of phyllosphere fungal community structure, nonmetric multidimensional scaling (NMDS) ordination of analysis was performed based on Bray-Curtis dissimilarity. Then, differences between the mainland and island fungal community compositions were tested by conducting analysis of similarity (ANOSIM) and a permutational multivariate analysis of variance (PERMANOVA). Furthermore, Linear Discriminant Analysis (LDA) effect size (LEfSe, http://huttenhower.sph.harvard.edu/galaxy/)67 was employed to determine indicator OTUs for each of the two regions. A logarithmic LDA score of 3.0 was set as the threshold for discriminative features.

Principal coordinate analysis was used to convert the pairwise Nei genetic distances to plant genetic eigenvectors (PGE) by using the “vegan” R package. The spatial principal coordinate of neighbor matrices (PCNM) vectors with positive eigenvalues were obtained based on the transformation of geographic distance (latitude and longitude) via the “PCNM” R package68. Significant spatial PCNM and host PGE vectors were forward-selected (α = 0.05) using the “Packfor” R package69 prior to subsequent statistical analyses. A PERMANOVA70 performed on the community matrix (Bray-Curtis dissimilarities) was used to evaluate the effect of host genetic eigenvectors, geographic distance, and climate factors on phyllosphere fungal community structure, which is also implemented through the “vegan” package with population site as a random effect. Moreover, multispecies generalized linear models (GLM) fitted with the “mvabund” R package were also used to identify the main drivers of phyllosphere fungal community structure, as Bálint et al.71 did. Congruence among distance matrices (CADM)72 and the Mantel test were then used to identify congruent patterns between host genetic structure and fungal community structure in mainland and island populations.

Fungal co-occurrence networks

Co-occurrence analyses were implemented to obtain a better understanding of fungal interactions in the phyllosphere of M. kwangtungensis. The fungal networks were built up with the “WGCNA” R package73 based on the Spearman correlation index. The nodes and the edges in the network represent fungal OTUs and the significant interactions between pairs of OTUs, respectively. The OTUs with relative abundances less than 0.01% were filtered because they were poorly represented. The P-values for multiple testing were calculated using the Benjamini and Hochberg discovery rate (FDR) test controlling procedure74. Only the rank correlation coefficient with values above 0.6 or below −0.6 and a statistically significant P value lower than 0.05 were considered as a valid correlation in the network. Sub-networks for mainland and island habitats, and then for each individual sample from the meta-community network, were then identified by preserving fungal OTUs present in each plant using the “igraph” package75. The networks of the two habitats were graphically displayed in Gephi (http://gephi.github.io/). Erdös-Réyni model random networks76 with the same number of nodes and edges as the observed networks were also constructed for each habitat. Network topological parameters (number of nodes and edges, average degree, degree centralization, centralization closeness, and modularity) provided in the “igraph” R package were calculated for each sample sub-network. The Wilcoxon test was then employed to assess significant difference in measured topological parameters between island and mainland networks77. The importance of biotic and abiotic factors (host plant genetic eigenvectors, geographic distance, and climate factors) for network-level topological features was evaluated with a random forest analysis using the “randomForest”78 and “rfPermute”79 R packages. The connectivity of network node was determined by its value of within-module connectivity (Zi) and among-module connectivity (Pi), which were calculated based on the methods of metabolic networks80. Node topologies were classified into four categories, according to the definition by Poudel et al.81: peripherals (Zi < 2.5 and Pi < 0.62, nodes with few links to other species), connectors (Pi > 0.62, nodes that connect modules), module hubs (Zi > 2.5, highly connected nodes within modules,), and network hubs (Zi > 2.5 and Pi > 0.62, highly connected nodes both in general and within a module). Connectors, module hubs, and network hubs were called keystone OTUs, which were considered to play unique and crucial roles in the stability of the fungal co-occurrence network structure and functioning20,82.

Compliance with ethical standards

This article does not contain any studies with human participants or animals performed by any of the authors.

Results

Fungal taxon identification

The rarefaction curves of the observed fungal OTUs reached a saturation plateau in both island and mainland populations, indicating that the majority of the phyllosphere fungal diversity was sampled (Fig. 2A). An average of 94 ± 4 OTUs (mean ± SE) per individual plant was found with 101 ± 6 OTUs per mainland sample and 86 ± 6 OTUs per island sample. The result of the Two-Sample Wilcoxon test showed that fungal richness was significantly lower on islands than the mainland (Fig. 2B). Fungal richness demonstrated a significantly positive correlation with plant genetic diversity (R2 = 0.645, P = 0.033). The observed OTUs were mainly found to belong to two fungal phyla: Ascomycota (529 OTUs; 61.3% of the reads) and Basidiomycota (199 OTUs; 21.1%). At the class level, the Ascomycota fungal reads recovered were dominated by the Dothideomycete (279 OTUs; 21.1%) and Eurotiomycetes (61 OTUs; 16.8%), while the most common Basidiomycota class was Tremellomycetes (60 OTUs; 11.6%) (Fig. 2C). In total, 260 genera were identified in phyllosphere fungal community of M. kwangtungensis. The most abundant genera (Top 5) were Aureobasidium (5 OTUs; 5.6%), Phaeococcus (1 OTU; 4.4%), Cladosporium (2 OTUs; 2.9%), Phaeophleospora (5 OTUs; 2.7%)and Phyllosticta (6 OTUs; 2.4%).

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig2_HTML.jpg

(A) Rarefaction curves and (B) box plots of phyllosphere fungal OTU richness in mainland and island Mussaenda kwangtungensis populations. Each point represents an individual plant. Different letters indicate statistical significance at P < 0.05. (C) Taxonomic composition of the fungal communities at the class level. The taxa with relative abundances of <1% were grouped into “others”.

Population structure of host plant

The eleven microsatellite loci showed an overall Ho of 0.609 for mainland populations of M. kwangtungensis, and 0.516 for island region. The AMOVA results demonstrated significant genetic differentiation between mainland and island populations (FCT = 0.099, P < 0.001). The IBD test revealed no significant correlation between population genetic and geographic distances (r = 0.529, P = 0.100). Delta K (ΔK) values from STRUCTURE analysis peaked sharply at K = 2, which indicates that there were two distinct genetic clusters, corresponding closely to the mainland and island regions (Fig. S1; Fig. 3). The UPGMA tree based on Nei genetic distance clustered all populations into two major groups (Fig. 3) with the mainland populations (BZS, ZHS and MSS) forming one group, and the remaining island populations (DAD, XWD and GSD) forming the other group. In accordance with the STRUCTURE and UPGMA cluster analysis, principal coordinate analysis also separated the populations into two main groups. The group coincided with the continental and insular habitats along Axis 1, and accounted for 73.2% of the total genetic variability (Fig. S2).

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig3_HTML.jpg

Topological congruencies between host genetic cluster and fungal dendrogram relationships. Left panel: a UPGMA dendrogram of host populations based on Nei genetic distance. Bar plot: population genetic structure of host plants using STRUCTURE analysis. Each thin vertical bar represents an individual plant, showing the proportion of its genetic pool assigned to each cluster. Right panel: a hierarchical dendrogram tree of the phyllosphere fungal communities based on Bray–Curtis dissimilarity.

Fungal community composition

The NMDS plot showed an observed structuring of phyllosphere fungal community between mainland and island populations based on Bray-Curtis dissimilarity (Fig. 4a). The significant differentiation was statistically confirmed by the results of ANOSIM (R = 0.100, P = 0.002) and PERMANOVA (R2 = 0.045, P = 0.002) analyses (Fig. 4a). The variation in fungal community structure was mainly explained by host plant genetic eigenvectors (14.2%) and the climate factor MAP (2.2%) (Table 1). These results closely corresponded to the multispecies GLM, where host genetic structure and MAP exerted the largest and second largest impacts on fungal community composition (Table 1). The global CADM test rejected the null model of incongruence between the matrices (P < 0.05), suggesting significant congruence between host population genetic structure and fungal dendrogram relationships (Fig. 3). The Mantel test results revealed a strong correlation between host genetic distance and phyllosphere fungal community dissimilarity (R = 0.797, P = 0.011; Fig. 3).

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig4_HTML.jpg

(a) Non-metric multidimensional scaling (NMDS) ordination of the fungal community structure. (b) Histogram of the linear discriminant analysis (LDA) scores computed for features differentially abundant between mainland and island populations with effect size measurements (LEfSe). Only OTUs with a LDA score greater than 3.0 are displayed.

Table 1

Effects of host and abiotic variables on the community composition of phyllosphere fungi revealed by permutational multivariate analysis of variance (PERMANOVA) and multispecies generalized linear models (GLM).

VariablePERMANOVAMultispecies GLM
SSR2P valueTSP value
Host genetic eigenvectors2.3830.142<0.001***47810.010**
Geographic distance0.3770.0230.05518510.535
MAP0.3660.0220.006**16550.040*
MAT0.2650.0160.59924680.188
Elevation0.2910.0170.36413350.980

SS: sum of squares; TS: test statistics; MAT: mean annual temperature; MAP: mean annual precipitation; *: 0.05 < P < 0.01; **: 0.01 < P < 0.001; ***: P < 0.001.

Seventeen fungal OTUs were found to respond significantly to the two habitat categories, with eight indicator OTUs (Cladosporium sp., Ramularia sp., Exobasidiales sp., Phyllachora sp., Pestalotiopsis sp., Phialophora sp., Paraconiothyrium sp., and Trichoderma sp.) for mainland and nine (Trimorphomyces sp., Saitozyma sp., Farysizyma sp., Ascomycota sp. (OTU 75), Ascomycota sp. (OTU 30), Vermiconia sp., Thelephoraceae sp., Erythrobasidiaceae sp., and Psathyrella sp.) for island (Fig. 4b).

Fungal co-occurrence network

Both mainland and island networks displayed a scale-free degree distribution (Fig. S2), suggesting that most phyllosphere fungal OTUs had low-degree values, and only a few hub nodes had a large number of connections to other members. Each network showed overwhelmingly more positive associations (99.8% and 98.8%, in mainland and island network, respectively) than negative ones (Fig. 5a,b). Main network-level topological features of each individual subnetwork were calculated to compare the co-occurrence patterns between mainland and island regions. There were significantly more nodes and edges, and a greater average degree for mainland networks than island ones (Fig. 5c–e). This indicates that the island habitat has a less complex structure. Degree centralization, and centralization closeness of individual sub-networks were higher in the mainland than the island regions (Fig. 5f,g), suggesting that the phyllosphere fungal networks were more connected in the mainland than the island. In contrast, modularity values were higher in island networks than mainland ones (Fig. 5h). This indicates that the island networks had more fragmented and modular structures. The random forest modelling outcomes demonstrated that plant genetic eigenvectors (PEGs) were the most important contributors to all the involved network topological features (Table 2). Additionally, MAP was the second most important factor explaining the variation in number of nodes. Elevation and MAP had a significant impact on the number of edges and degree centralization.

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig5_HTML.jpg

Co-occurrence networks of phyllosphere fungal communities in mainland (a) and island (b) populations. Each node represents a fungal OTU and is colored by module. The size of the node is proportional to the number of reads. The color of each link reflects positive (red) or negative (blue) associations. (c–h) Comparison of main network properties between the two different regions. Different letters indicate statistical significance at P < 0.05.

Table 2

Mean predictor importance of host and abiotic variables on network structure of phyllosphere fungal community revealed by random forest analyses.

Network propertyVariableImportanceP value
Number of nodesPGE217.0390.029
PGE284.6480.049
PGE14.0770.029
MAP2.5920.030
Number of edgesPGE285.8340.004
Elevation4.1970.002
PGE313.7310.042
MAP2.9490.010
Average degreePGE216.1610.011
PGE114.1900.044
PGE304.1040.042
Centralization closenessPGE253.2770.039
PGE223.0680.049
Degree centralizationPGE28.9680.002
PGE335.7480.012
PGE314.9940.026
Elevation4.7900.006
MAP2.9450.032
ModularityPGE345.4410.024
PGE24.0540.028

PGE: plant genetic eigenvector; MAP: mean annual precipitation. Only significant variables were shown.

The majority of the nodes in both mainland (95.5%) and island (96.9%) networks were peripherals with most of their links inside their modules (Fig. 6). Neither mainland nor island networks possessed network hubs highly connected with other nodes within entire network (Fig. 6). Three module hubs (Nigrospora sp., Phaeophleospora sp., and Acremonium sp.) and 13 connectors (Stagonospora sp., Capnodiales sp., Phomopsis sp., Phoma sp., Fusidium sp., Hyalocladosporiella sp., Septobasidiaceae sp., Hortaea sp., Paraconiothyrium sp., Preussia sp., Sphaceloma sp., Trichomerium sp., and Rhytidhysteron sp.) were identified in the mainland network. Connectors (Myrothecium sp., Eriosporella sp., Trimorphomyces sp., Capnodiales sp., Fungi sp., Curvularia sp., Corticiaceae sp., and Glomerella sp.) were also found in the island network, but no module hubs were detected. Members of Dothideomycetes were the most prominent keystone taxa (50%), followed by Sordariomycetes (25%). A large proportion of keystone OTUs (62.5%) had low relative abundance (<0.05%).

An external file that holds a picture, illustration, etc. Object name is 41598_2020_57622_Fig6_HTML.jpg

Zi-Pi plots to identify putative keystone OTUs within the phyllosphere networks in mainland (A) and island (B) plant populations. Each point represents a fungal OTU. Threshold values of Zi and Pi for classification are 2.5 and 0.62, respectively. The modules hubs and connectors are listed on the plots.

Discussion

Taxon identification

The phyllosphere fungal microbiome of M. kwangtungensis was largely comprised of Ascomycota taxa and dominated by members of Dothideomycetes and Eurotiomycetes at the class level (Fig. 2C). This outcome concurs with previous researches on the phyllosphere fungi in Sequoia sempervirens83, Picea glauca17, and Mussaenda pubescens var. alba16 using high-throughput sequencing methods. Kembel and Mueller14 found that the phyllosphere Dothideomycetes and Eurotiomycetes members across 51 tree species in a Panamanian rainforest contained a large number of melanized black yeasts, which were usually isolated from stressful environments and were remarkable for the wide variety of extremes they can tolerate. The most abundant genus in our study (Aureobasidium) is also known as a group of cosmopolitan black yeasts due to its melanin production84, which can help it withstand high ultraviolet (UV) radiation and moderately osmotic condition85. This stress-tolerance may be an important adaptation for the survival of fungi in a phyllosphere environment with low nutrient availability, strong UV exposure, and high temperature fluctuations throughout the diurnal rhythm14,86. Additionally, Aureobasidium spp. exhibit antagonistic activity against a number of phytopathogens, thus protecting host plants from disease87. Other high frequent Ascomycota species, such as Cladosporium spp. and Phyllosticta spp., may also offer enhanced stress tolerance for the host plant and have been frequently isolated from the leaf endosphere in previous studies88–91. Basidiomycota was the second most abundant phylum with Tremellomycetes as its dominant class. This class has previously been reported in foliar fungal communities associated with vascular plants in Svalbard, Norway92. A high abundance of Basidiomycota yeasts (Papiliotrema, Symmetrospora, and Cryptococcus) were detected in the phyllosphere fungal community of M. kwangtungensis. A similar scenario was also observed in the cereal phyllosphere mycobiome93. Andrews and Harris94 suggested that yeasts appear to be more active and permanent colonizers of leaf surfaces, whereas the filamentous fungi were mostly transients and often existed on the leaves as dormant spores.

Community differentiation

The ANOSIM results suggested that phyllosphere fungal community structure differed significantly between mainland and insular habitats, even resembling the patterns in AMF community assemblages found by da Silva et al.39. This structure pattern was mainly influenced by host plant genetic structure (Table 1, Fig. 3). Plant genetic structure results from the joint action of migration, mutation, selection and drift95, thus reflecting the historical and biological context of plant species and determining their evolutionary potential96. One of the reasons for this significant population structure for M. kwangtungensis may be the relatively limited gene exchange due to the oceanic barrier. In addition, small and spatially isolated populations will lead to allelic fixation at some neutral loci within populations and subsequently promote population genetic divergence, mainly as a result of stochastic events or genetic drift95,96. We found lower genetic diversity displayed in island populations, which corroborate the traditional idea35. Founder effects, small population size, genetic bottlenecks, and geographical isolation from the mainland were commonly considered as major factors responsible for such a pattern arising36.

Host genetic divergence may result in variation in phenotypic traits including leaf chemistry, morphology or physiology, even within a single plant species97–99. The assembly and development of phyllosphere fungal communitis must involve those plant phenotypic traits controlled primarily by their genetic make-up6. For example, Saunders and Kohn100 demonstrated that benzoxazinoid production by host plant improved the ecological success of Fusarium species in the host. In addition, non-Fusarium species with intermediate 2-benzoxazolinone tolerance levels showed a colonization advantage over sensitive fungi, which indicates that phyllosphere fungal assemblage are at least partially filtered by host defense compounds. This conclusion was further confirmed through leaf-extract assays by Lau et al.101, who showed a strong effect of host versus non-host leaf chemistry on fungal endophyte growth in vitro. Additionally, leaf topographic features, such as trichomes, wax crystals and cuticle thickness, with variation occurring among intraspecies and cultivars, will also affect which and how many fungal species can persist in a location. For instance, trichomes or hairy extensions may ensnare water and fungal spores, or they may keep spores from adhering to the leaf surface102. Bailey et al.103 found that the establishment of some endophytic Trichoderma species were closely associated with glandular trichomes of Theobroma cacao. Moreover, leaf surface secretions (e.g., organic acids, simple sugars, and other easily utilized carbon source) have remarkably influenced the community structure of many yeast epiphytes104. Plant intraspecific phenotypic plasticity is now generally considered to be heritable and of potential importance for species’ evolution97. Therefore, adaptive plant genetic divergence among populations, which can result in local adaptation, may be the starting point for coevolution between a plant and its associated microbiome105. Balint et al.106 demonstrated that host genotype-specific phyllosphere fungal communities could live in the plant systemically, existing in the host tree even after two round of reproduction. We found strong congruence between host genetic divergence and phyllosphere fungal compositional divergence (Fig. 3), indicating an eco-evolutionary pattern in which evolutionary changes in the M. kwangtungensis associate with ecological changes in the fungal microbiome107. However, the functional basis and filtering mechanism for this obvious plant-fungi interaction still require further investigation.

MAP has a small but significant influence on structuring phyllosphere fungal communities (Table 1), which is in line with the patterns observed in foliar fungal communities of Metrosideros polymorpha across a Hawaiian landscape13, in AMF community assembly associated with four plant species in a grassland ecosystem108, and in endophytic fungal community of Panicum hallii across the Edwards Plateau109. Precipitation can directly influence the fungal species pool, and can also exert important influence on ecosystem processes including respiration, decomposition and plant productivity, as well as may indirectly affect plant-associated fungal communities via changes in the local host community properties110. Moreover, variation in environmental factors could also augment, weaken or veil the effects of host genes, which would lead to context-dependent expressions of genetic variation for plant phenotypic traits that may finally affect the assembly of phyllosphere fungal communities111.

Co-occurrence networks

Investigating co-occurrence patterns among fungal community members can aid in detecting potential biological interactions, habitat affinities, or shared physiologies, and may provide new insight into the structure of complex fungal communities112. The potential biological interactions revealed by the ecological microbial network often range from positive (e.g. commensalism and mutualisms) to negative (e.g. predation and competition) relationships113. For example, fungal propagules may stimulate or antagonize growth of other phyllosphere members by producing metabolites, offering structural complexity, competing for resources, and changing foliar trophic webs114. In this study, the correlations among fungal OTUs in each network were revealed to be predominantly positive (Fig. 5), thus indicating the potential for widespread cooperative and syntrophic relationships between most fungal members, or multiple trophic levels, as well as the potential for sharing niches based on nutritional preference and functional distinctiveness in the phyllosphere micro-environment. This overwhelmingly positive association has also been reported in microbial networks associated with mosses, lichens and the bark of maple trees29, and in the rhizosphere zone of wild oat115.

Comparison of the phyllosphere fungal co-occurrence networks revealed significant structural differences between island and mainland populations (Fig. 5). The island fungal subnetworks were significantly less complex, less densely interconnected and more compartmentalized and modular than their mainland counterparts. The most important explanation for the topological differentiation is the genetic background of the host plant (Table 2). Host genotype could act as an important plant-imposed filter by preventing the establishment of species that lack the phenotype required to survive, and thus further influencing the fungal distribution and interactions. In addition to direct interspecific microbial interactions, interactions could also be indirect, via their effects on plant defenses and the chemistry of leaves116,117. For example, a colonizing fungal species may induce a host immune defense reaction, triggered by well-conserved traits (e.g., chitin) that are produced by current fungal inhabitants98,118,119. We also identified a significant effect of MAP on topological variations across the regions. This indicates that water-energy dynamics may play a critical role in determining co-occurrence patterns at a regional scale. Precipitation variability could rapidly alter key carbon cycling processes of plants (such as net photosynthesis), which would result in long-term consequences for carbon storage and biotic-atmospheric feedbacks120. Further, it has been reported that topological features of fungal co-occurrence networks are closely related to efficient carbon utilization121. Therefore, more connected networks are expected to have higher carbon uptake, as in the mainland.

Keystone OTUs are considered to have potentially crucial roles in network structure maintenance and may exert a large influence on community assembly77,122. The disappearance of these keystone nodes may segregate networks into more modules. The island network has fewer keystone taxa, which could partly contribute to its more modular and compartmentalized network structure. Keystone nodes may also possess vital ecological functions in the phyllopshere fungal community. Some connectors and module hubs detected in our study, such as Acremonium sp., Preussia sp., Curvularia sp., Phomopsis sp., Rhytidhysteron sp., Fusidium sp., Glomerella sp. and Phoma sp., have been frequently identified in previous studies as foliar fungal endophytes123–129. They are widely acknowledged as important determinants of host-plant fitness, productivity, and function. We found that more than half of the keystone nodes had relatively low abundances, as has been previously reported in soil microbial networks77. This suggests that rare microorganisms can also play functionally important roles in maintaining microbial networks. Future research focusing on uncultured keystone fungal species will be critical to understanding the mechanisms in network assembly better.

In conclusion, the phyllosphere fungal community structure and co-occurrence patterns of M. kwangtungensis showed significant dissimilarity between mainland and island populations. This differentiation was determined first by divergence in host plant genetics and secondarily by environmental variation. This study contributes to a richer understanding of the interplay between host plants and phyllosphere fungal microbiomes. However, much more research is needed to further explore how host plants influence niche differentiation and interspecific interactions in phyllosphere fungal communities, as well as how these influences may shape community structural variability, co-occurrence patterns and also long-term eco-evolutionary processes of plant-fungi interactions.

Supplementary information

Acknowledgements

This study was financially supported by the Strategic Priority Research Program of the Chinese Academy of Sciences (XDA13020504), Science & Technology Basic Works Program of the Ministry of Science and Technology of China (Grant No. 2013FY111200), and National Natural Science Foundation of China (Grant No. 31400209).

Author contributions

X.Q. conceived and performed the experiments, analyzed the data, and drafted the manuscript. S.L., B.W., Y.W., N.J., H.Y. and H.C. performed the experiments and processed Illumina sequencing data. M.S. and D.Z. designed the study and approved the final draft.

Competing interests

The authors declare no competing interests.

Footnotes

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Miaomiao Shi, nc.ca.gbcs@ihsmm.

Dianxiang Zhang, nc.ca.gbcs@gnahz-xd.

Supplementary information

is available for this paper at 10.1038/s41598-020-57622-6.

References

1. Jumpponen A, Jones KL. Massively parallel 454 sequencing indicates hyperdiverse fungal communities in temperate Quercus macrocarpa phyllosphere. New Phytol. 2009;184:438–448. 10.1111/j.1469-8137.2009.02990.x. [Abstract] [CrossRef] [Google Scholar]
2. Gomes T, Pereira JA, Benhadi J, Lino-Neto T, Baptista P. Endophytic and epiphytic phyllosphere fungal communities are shaped by different environmental factors in a Mediterranean ecosystem. Microb. Ecol. 2018;76:668–679. 10.1007/s00248-018-1161-9. [Abstract] [CrossRef] [Google Scholar]
3. Nutaratat P, Srisuk N, Arunrattiyakorn P, Limtong S. Plant growth-promoting traits of epiphytic and endophytic yeasts isolated from rice and sugar cane leaves in Thailand. Fungal. Biol. 2014;118:683–694. [Abstract] [Google Scholar]
4. Rodriguez RJ, et al. Stress tolerance in plants via habitat-adapted symbiosis. ISME J. 2008;2:404–416. 10.1038/ismej.2007.106. [Abstract] [CrossRef] [Google Scholar]
5. Rodriguez R, White JJ, Arnold A, Redman R. Fungal endophytes: diversity and functional roles. New Phytol. 2009;182:314–330. 10.1111/j.1469-8137.2009.02773.x. [Abstract] [CrossRef] [Google Scholar]
6. Whipps JM, Hand P, Pink D, Bending GD. Phyllosphere microbiology with special reference to diversity and plant genotype. J. Appl. Microbiol. 2008;105:1744–1755. 10.1111/j.1365-2672.2008.03906.x. [Abstract] [CrossRef] [Google Scholar]
7. Newton AC, Fitt BDL, Atkins SD, Walters DR, Daniell T. J. Pathogenesis, parasitism and mutualism in the trophic space of microbe–plant interactions. Trends Microbiol. 2010;18:365–373. 10.1016/j.tim.2010.06.002. [Abstract] [CrossRef] [Google Scholar]
8. Fisher PJ, Petrini O. Fungal saprobes and pathogens as endophytes of rice (Oryza sativa L.) New Phytol. 1992;120:137–143. 10.1111/j.1469-8137.1992.tb01066.x. [CrossRef] [Google Scholar]
9. Voříšková J, Baldrian P. Fungal community on decomposing leaf litter undergoes rapid successional changes. ISME J. 2012;7:477–486. 10.1038/ismej.2012.116. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
10. Purahong W, et al. Life in leaf litter: novel insights into community dynamics of bacteria and fungi during litter decomposition. Mol. Ecol. 2016;25:4059–4074. 10.1111/mec.13739. [Abstract] [CrossRef] [Google Scholar]
11. Izuno A, et al. Structure of phyllosphere fungal communities in a tropical dipterocarp plantation: A massively parallel next-generation sequencing analysis. Mycoscience. 2016;7:171–180. 10.1016/j.myc.2015.12.005. [CrossRef] [Google Scholar]
12. Yang T, et al. Carbon constrains fungal endophyte assemblages along the timberline. Environ. Microbiol. 2016;18:2455–2469. 10.1111/1462-2920.13153. [Abstract] [CrossRef] [Google Scholar]
13. Zimmerman NB, Vitousek PM. Fungal endophyte communities reflect environmental structuring across a Hawaiian landscape. Proc. Natl. Acad. Sci. USA. 2012;109:13022–13027. 10.1073/pnas.1209872109. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
14. Kembel SW, Mueller RC. Plant traits and taxonomy drive host associations in tropical phyllosphere fungal communities. Botany. 2014;92:303–311. 10.1139/cjb-2013-0194. [CrossRef] [Google Scholar]
15. Cordier T, et al. The composition of phyllosphere fungal assemblages of European beech (Fagus sylvatica) varies significantly along an elevation gradient. New Phytol. 2012;196:510–519. 10.1111/j.1469-8137.2012.04284.x. [Abstract] [CrossRef] [Google Scholar]
16. Qian X, et al. Host genotype strongly influences phyllosphere fungal communities associated with Mussaenda pubescens var. alba (Rubiaceae) Fungal Ecol. 2018;36:141–151. 10.1016/j.funeco.2018.10.001. [CrossRef] [Google Scholar]
17. Eusemann P, et al. Habitat conditions and phenological tree traits overrule the influence of tree genotype in the needle mycobiome-Picea glauca system at an arctic treeline ecotone. New Phytol. 2016;211:1221–1231. 10.1111/nph.13988. [Abstract] [CrossRef] [Google Scholar]
18. Berry D, Widder S. Deciphering microbial interactions and detecting keystone species with co-occurrence networks. Front. Microbiol. 2014;5:219. 10.3389/fmicb.2014.00219. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
19. Banerjee S, et al. Network analysis reveals functional redundancy and keystone taxa amongst bacterial and fungal communities during organic matter decomposition in an arable soil. Soil Biol. Biochem. 2016;97:188–198. 10.1016/j.soilbio.2016.03.017. [CrossRef] [Google Scholar]
20. Banerjee S, Schlaeppi K, van der Heijden MGA. Keystone taxa as drivers of microbiome structure and functioning. Nat. Rev. Microbiol. 2018;16:567–576. 10.1038/s41579-018-0024-1. [Abstract] [CrossRef] [Google Scholar]
21. Ma B, Lv X, Cai Y, Chang SX, Dyck MF. Liming does not counteract the influence of long-term fertilization on soil bacterial community structure and its co-occurrence pattern. Soil Biol. Biochem. 2018;123:45–53. 10.1016/j.soilbio.2018.05.003. [CrossRef] [Google Scholar]
22. Mikhailov IS, et al. Co-occurrence networks among bacteria and microbial eukaryotes of lake baikal during a spring phytoplankton bloom. Microb. Ecol. 2018;96:96–109. [Abstract] [Google Scholar]
23. Milici M, et al. Co-occurrence analysis of microbial taxa in the Atlantic Ocean reveals high connectivity in the free-Living bacterioplankton. Front. Microbiol. 2016;7:649. 10.3389/fmicb.2016.00649. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
24. Hu A, et al. Strong impact of anthropogenic contamination on the co-occurrence patterns of a riverine microbial community. Environ. Microbiol. 2017;19:4993–5009. 10.1111/1462-2920.13942. [Abstract] [CrossRef] [Google Scholar]
25. Jackson MA, et al. Detection of stable community structures within gut microbiota co-occurrence networks from different human populations. PeerJ. 2018;6:e4303. 10.7717/peerj.4303. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
26. Ramayo-Caldas Y, et al. Phylogenetic network analysis applied to pig gut microbiota identifies an ecosystem structure linked with growth traits. ISME J. 2016;10:2973–2977. 10.1038/ismej.2016.77. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
27. Zheng W, Zhao Z, Gong Q, Zhai B, Li Z. Responses of fungal–bacterial community and network to organic inputs vary among different spatial habitats in soil. Soil Biol. Biochem. 2018;125:54–63. 10.1016/j.soilbio.2018.06.029. [CrossRef] [Google Scholar]
28. Li S, Wu F. Diversity and co-occurrence patterns of soil bacterial and fungal communities in seven intercropping systems. Front. Microbiol. 2018;9:1521. 10.3389/fmicb.2018.01521. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
29. Aschenbrenner IA, Cernava T, Erlacher A, Berg G, Grube M. Differential sharing and distinct co-occurrence networks among spatially close bacterial microbiota of bark, mosses and lichens. Mol. Ecol. 2017;26:2826–2838. 10.1111/mec.14070. [Abstract] [CrossRef] [Google Scholar]
30. Bouffaud ML, et al. Indicator species and co-occurrence in communities of arbuscular mycorrhizal fungi at the European scale. Soil Biol. Biochem. 2016;103:464–470. 10.1016/j.soilbio.2016.09.022. [CrossRef] [Google Scholar]
31. Qian X, et al. Shifts in community composition and co-occurrence patterns of phyllosphere fungi inhabiting Mussaenda shikokiana along an elevation gradient. PeerJ. 2018;6:e5767. 10.7717/peerj.5767. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
32. Ashley MV, Backs JR, Kindsvater L, Abraham ST. Genetic variation and structure in an endemic island oak, Quercus tomentella, and mainland canyon oak, Quercus chrysolepis. Int. J. Plant. Sci. 2018;179:151–161. 10.1086/696023. [CrossRef] [Google Scholar]
33. Yamada T, Maki M. Impact of geographical isolation on genetic differentiation in insular and mainland populations of Weigela coraeensis (Caprifoliaceae) on Honshu and the Izu Islands. J. Biogeogr. 2012;39:901–917. 10.1111/j.1365-2699.2011.02634.x. [CrossRef] [Google Scholar]
34. Losos JB, Ricklefs RE. Adaptation and diversification on islands. Nature. 2009;457:830–836. 10.1038/nature07893. [Abstract] [CrossRef] [Google Scholar]
35. Frankham R. Do island populations have less genetic variation than mainland populations? Heredity. 1997;78:311–327. 10.1038/hdy.1997.46. [Abstract] [CrossRef] [Google Scholar]
36. García-Verdugo C, et al. Do island plant populations really have lower genetic variation than mainland populations? Effects of selection and distribution range on genetic diversity estimates. Mol. Ecol. 2015;24:726–741. 10.1111/mec.13060. [Abstract] [CrossRef] [Google Scholar]
37. Santos AMC, Field R, Ricklefs RE. New directions in island biogeography. Global. Ecol. Biogeogr. 2016;25:751–768. 10.1111/geb.12477. [CrossRef] [Google Scholar]
38. Davison J, et al. Microbial island biogeography: isolation shapes the life history characteristics but not diversity of root-symbiotic fungal communities. ISME J. 2018;12:2211–2224. 10.1038/s41396-018-0196-8. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
39. Da Silva IR, De Souza FA, Da Silva DKA, Oehl F, Maia LC. Patterns of arbuscular mycorrhizal fungal distribution on mainland and island sandy coastal plain ecosystems in Brazil. Microb. Ecol. 2017;74:654–669. 10.1007/s00248-017-0979-x. [Abstract] [CrossRef] [Google Scholar]
40. Luo Z, et al. Reproductive isolation between sympatric sister species, Mussaenda kwangtungensis and M. pubescens var. alba. J. Integr. Plant Biol. 2015;57:859–870. 10.1111/jipb.12325. [Abstract] [CrossRef] [Google Scholar]
41. Duan TT, Gong W, Zhang DX. Development of microsatellite markers from Mussaenda pubescens (Rubiaceae) Am. J. Bot. 2012;99:437–439. 10.3732/ajb.1200188. [Abstract] [CrossRef] [Google Scholar]
42. Gardes M, Bruns TD. ITS primers with enhanced specificity for basidiomycetes-application to the identification of mycorrhizae and rusts. Mol. Ecol. 1993;2:113–118. 10.1111/j.1365-294X.1993.tb00005.x. [Abstract] [CrossRef] [Google Scholar]
43. White, T. J, Bruns, T., Lee, S. & Taylor, J. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics in PCR Protocols: A Guide to Methods and Applications. (eds. Michael, A. I., David, H. G., John, J. S. & Thomas, J. W.) 315–322 (Academic Press, 1990).
44. Ihrmark K, et al. New primers to amplify the fungal ITS2 region-evaluation by 454-sequencing of artificial and natural communities. FEMS. Microbiol. Ecol. 2012;82:666–677. 10.1111/j.1574-6941.2012.01437.x. [Abstract] [CrossRef] [Google Scholar]
45. Duan T, Zhang D. Fourteen additional microsatellite markers for Mussaenda pubescens and cross-species amplification. J. Genet. 2014;93:44–47. 10.1007/s12041-012-0142-x. [Abstract] [CrossRef] [Google Scholar]
46. Excoffier L, Lischer HE. Arlequin suite ver 3.5: a new series of programs to perform population genetics analyses under Linux and Windows. Mol. Ecol. Resour. 2010;10:564–567. 10.1111/j.1755-0998.2010.02847.x. [Abstract] [CrossRef] [Google Scholar]
47. Nei M. Genetic Distance between Populations. Am. Nat. 1972;106:283–292. 10.1086/282771. [CrossRef] [Google Scholar]
48. Wright S. Evolution in mendelian populations. Genetics. 1931;16:97–159. [Europe PMC free article] [Abstract] [Google Scholar]
49. Peakall R, Smouse PE. GenAlEx 6.5: genetic analysis in Excel. Population genetic software for teaching and research-an update. Bioinformatics. 2012;28:2537–2539. 10.1093/bioinformatics/bts460. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
50. Goslee SC, Urban DL. The ecodist package for dissimilarity-based analysis of ecological data. J. Stat. Softw. 2007;22:1–19. 10.18637/jss.v022.i07. [CrossRef] [Google Scholar]
51. Pritchard JK, Stephens M, Donnelly P. Inference of population structure using multilocus genotype data. Genetics. 2000;155:945–959. [Europe PMC free article] [Abstract] [Google Scholar]
52. Earl DA, Vonholdt BM. STRUCTURE HARVESTER: a website and program for visualizing STRUCTURE output and implementing the Evanno method. Conserv. Genet. Resour. 2012;4:359–361. 10.1007/s12686-011-9548-7. [CrossRef] [Google Scholar]
53. Jakobsson M, Rosenberg NA. CLUMPP: a cluster matching and permutation program for dealing with label switching and multimodality in analysis of population structure. Bioinformatics. 2007;23:1801–1806. 10.1093/bioinformatics/btm233. [Abstract] [CrossRef] [Google Scholar]
54. Rosenberg NA. DISTRUCT: a program for the graphical display of population structure. Mol. Ecol. Notes. 2004;4:137–138. 10.1046/j.1471-8286.2003.00566.x. [CrossRef] [Google Scholar]
55. Zhang J, Kobert K, Flouri T, Stamatakis A. PEAR: a fast and accurate Illumina Paired-End reAd mergeR. Bioinformatics. 2014;30:614–620. 10.1093/bioinformatics/btt593. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
56. Caporaso JG, et al. QIIME allows analysis of high-throughput community sequencing data. Nat. Methods. 2010;7:335–336. 10.1038/nmeth.f.303. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
57. Bengtsson-Palme J, et al. Improved software detection and extraction of ITS1 and ITS2 from ribosomal ITS sequences of fungi and other eukaryotes for analysis of environmental sequencing data. Methods Ecol. Evol. 2013;4:914–919. [Google Scholar]
58. Edgar RC, Haas BJ, Clemente JC, Quince C, Knight R. UCHIME improves sensitivity and speed of chimera detection. Bioinformatics. 2011;27:2194–2200. 10.1093/bioinformatics/btr381. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
59. Edgar RC. UPARSE: highly accurate OTU sequences from microbial amplicon reads. Nat. Methods. 2013;10:996–998. 10.1038/nmeth.2604. [Abstract] [CrossRef] [Google Scholar]
60. Tedersoo L, et al. Fungal biogeography. Global diversity and geography of soil fungi. Science. 2014;346:1256688. [Abstract] [Google Scholar]
61. Toju H, Tanabe AS, Ishii HS. Ericaceous plant-fungus network in a harsh alpine-subalpine environment. Mol. Ecol. 2016;25:3242–3257. 10.1111/mec.13680. [Abstract] [CrossRef] [Google Scholar]
62. Schloss PD, et al. Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl. Environ. Microbiol. 2009;75:7537–7541. 10.1128/AEM.01541-09. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
63. Ihaka R, Gentleman R. R: A language for data analysis and graphics. J. Comput. Graph. Stat. 1996;5:299–314. [Google Scholar]
64. Oksanen J, et al. Vegan: Community Ecology Package. R package version. 2016;2:4–1. [Google Scholar]
65. Wilcoxon F. Individual comparisons by ranking methods. Biometrics Bull. 1945;1:80–83. 10.2307/3001968. [CrossRef] [Google Scholar]
66. Clarke KR, Somerfield PJ, Chapman MG. On resemblance measures for ecological studies, including taxonomic dissimilarities and a zero-adjusted Bray–Curtis coefficient for denuded assemblages. J. Exp. Mar. Biol. Ecol. 2006;330:55–80. 10.1016/j.jembe.2005.12.017. [CrossRef] [Google Scholar]
67. Segata N, et al. Metagenomic biomarker discovery and explanation. Genome Biol. 2011;12:R60. 10.1186/gb-2011-12-6-r60. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
68. Legendre, P., Borcard, D., Blanchet, G. & Dray, S. PCNM: PCNM spatial eigenfunction and principal coordinate analyses. R package version 2.1/r82 (2010).
69. Dray S, Legendre P, Blanchet F. Packfor: forward selection with permutation. R package version. 2007;0:0–7. [Google Scholar]
70. Anderson MJ. A new method for non-parametric multivariate analysis of variance. Austral. Ecol. 2001;26:32–46. [Google Scholar]
71. Bálint M, et al. Relocation, high-latitude warming and host genetic identity shape the foliar fungal microbiome of poplars. Mol. Ecol. 2015;24:235–248. 10.1111/mec.13018. [Abstract] [CrossRef] [Google Scholar]
72. Legendre P, Lapointe FJ. Assessing congruenceamong distance matrices: single-malt scotch whiskies revisited. Aust. NZ J. Stat. 2004;46:615–629. 10.1111/j.1467-842X.2004.00357.x. [CrossRef] [Google Scholar]
73. Langfelder P, Horvath S. WGCNA: an R package for weighted correlation network analysis. BMC Bioinformatics. 2008;9:559. 10.1186/1471-2105-9-559. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
74. Benjamini Y, Hochberg Y. Controlling the false discovery rate: a practical and powerful approach to multiple testing. J. R. Statist. Soc. B. 1995;57:289–300. [Google Scholar]
75. Csardi G, Nepusz T. The igraph software package for complex network research. Interj. Complex Syst. 2006;1695:1–9. [Google Scholar]
76. Erdös P, Rényi A. On the evolution of random graphs. Publ. Math. Inst. Hung. Acad. Sci. 1960;5:17–61. [Google Scholar]
77. Ma B, et al. Geographic patterns of co-occurrence network topological features for soil microbiota at continental scale in eastern China. ISME J. 2016;10:1891–1901. 10.1038/ismej.2015.261. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
78. Liaw A, Wiener M. Classification and regression by RandomForest. R news. 2002;2:18–22. [Google Scholar]
79. Archer, E. Estimate permutation p-values for random forest importance metrics. R package (Zenodo), Version (2002).
80. Guimerà R, Nunes Amaral LA. Functional cartography of complex metabolic networks. Nature. 2005;433:895–900. 10.1038/nature03288. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
81. Poudel R, et al. Microbiome networks: A systems framework for identifying candidate microbial assemblages for disease management. Phytopathology. 2016;106:1083–1096. 10.1094/PHYTO-02-16-0058-FI. [Abstract] [CrossRef] [Google Scholar]
82. Tylianakis JM, Morris RJ. Ecological networks across environmental gradients. Annu. Rev. Ecol. Evol. 2017;48:25–48. 10.1146/annurev-ecolsys-110316-022821. [CrossRef] [Google Scholar]
83. Harrison JG, Forister ML, Parchman TL, Koch GW. Vertical stratification of the foliar fungal community in the world’s tallest trees. Am. J. Bot. 2016;103:2087–2095. 10.3732/ajb.1600277. [Abstract] [CrossRef] [Google Scholar]
84. Li Y, et al. Taxonomy of Aureobasidium spp. and biosynthesis and regulation of their extracellular polymers. Crit. Rev. Microbiol. 2015;41:228–237. 10.3109/1040841X.2013.826176. [Abstract] [CrossRef] [Google Scholar]
85. Turk M, Gostinčar C. Glycerol metabolism genes in Aureobasidium pullulans and Aureobasidium subglaciale. Fungal Biol. 2018;122:63–73. 10.1016/j.funbio.2017.10.005. [Abstract] [CrossRef] [Google Scholar]
86. Remus-Emsermann MNP, Schlechter RO. Phyllosphere microbiology: at the interface between microbial individuals and the plant host. New Phytol. 2018;218:1327–1333. 10.1111/nph.15054. [Abstract] [CrossRef] [Google Scholar]
87. Pinto C, et al. Understand the potential role of Aureobasidium pullulans, a resident microorganism from grapevine, to prevent the infection caused by diplodia seriata. Front. Microbiol. 2018;9:3047. 10.3389/fmicb.2018.03047. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
88. Guarnaccia V, et al. First report of Phyllosticta citricarpa and description of two new species, P. paracapitalensis and P. paracitricarpa, from citrus in Europe. Stud. Mycol. 2017;87:161–185. 10.1016/j.simyco.2017.05.003. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
89. Rajamani T, Suryanarayanan TS, Murali TS, Thirunavukkarasu N. Distribution and diversity of foliar endophytic fungi in the mangroves of Andaman Islands, India. Fungal Ecol. 2018;36:109–116. 10.1016/j.funeco.2018.09.007. [CrossRef] [Google Scholar]
90. Venkateswarulu N, et al. Isolation and characterization of plumbagin (5-hydroxyl-2-methylnaptalene-1,4-dione) producing endophytic fungi Cladosporium delicatulum from endemic medicinal plants. Biotechnol. Rep. 2018;20:e00282. 10.1016/j.btre.2018.e00282. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
91. Dastogeer KMG, Li H, Sivasithamparam K, Jones MGK, Wylie SJ. Host specificity of endophytic mycobiota of wild nicotiana plants from arid regions of northern Australia. Microb. Ecol. 2018;75:74–87. 10.1007/s00248-017-1020-0. [Abstract] [CrossRef] [Google Scholar]
92. Zhang T, Yao YF. Endophytic fungal communities associated with vascular plants in the high arctic zone are highly diverse and host-plant specific. PLoS One. 2015;10:e0130051. 10.1371/journal.pone.0130051. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
93. Sapkota R, Knorr K, Jorgensen LN, O’Hanlon KA, Nicolaisen M. Host genotype is an important determinant of the cereal phyllosphere mycobiome. New Phytol. 2015;207:1134–1144. 10.1111/nph.13418. [Abstract] [CrossRef] [Google Scholar]
94. Andrews JH, Harris RF. The Ecology and biogeography of microorganisms on plant surfaces. Annu. Rev. Phytopathol. 2000;38:145–180. 10.1146/annurev.phyto.38.1.145. [Abstract] [CrossRef] [Google Scholar]
95. Loveless MD, Hamrick JL. Ecological determinants of genetic structure in plant populations. Annu. Rev. Ecol. Syst. 1984;15:65–95. 10.1146/annurev.es.15.110184.000433. [CrossRef] [Google Scholar]
96. Alors D, et al. Panmixia and dispersal from the Mediterranean Basin to Macaronesian Islands of a macrolichen species. Sci. Rep. 2017;7:40879. 10.1038/srep40879. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
97. Nicotra AB, et al. Plant phenotypic plasticity in a changing climate. Trends Plant Sci. 2010;15:684–692. 10.1016/j.tplants.2010.09.008. [Abstract] [CrossRef] [Google Scholar]
98. Saunders M, Glenn AE, Kohn LM. Exploring the evolutionary ecology of fungal endophytes in agricultural systems: using functional traits to reveal mechanisms in community processes. Evol. Appl. 2010;3:525–537. 10.1111/j.1752-4571.2010.00141.x. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
99. Wink M. Evolution of secondary metabolites from an ecological and molecular phylogenetic perspective. Phytochemistry. 2003;64:3–19. 10.1016/S0031-9422(03)00300-5. [Abstract] [CrossRef] [Google Scholar]
100. Saunders M, Kohn LM. Evidence for alteration of fungal endophyte community assembly by host defense compounds. New Phytol. 2003;182:229–238. 10.1111/j.1469-8137.2008.02746.x. [Abstract] [CrossRef] [Google Scholar]
101. Lau MK, Arnold AE, Johnson NC. Factors influencing communities of foliar fungal endophytes in riparian woody plants. Fungal Ecol. 2013;6:365–378. 10.1016/j.funeco.2013.06.003. [CrossRef] [Google Scholar]
102. Allen, E. A., Hoch, H. C., Steadman, J. R. & Stavely, R. J. Influence of leaf surface features on spore deposition and the epiphytic growth of phytopathogenic fungi in Microbial Ecology of Leaves (eds. Andrews, J. H. & Hirano, S. S.) 87–110 (Springer, 1991).
103. Bailey BA, Strem MD, Wood D. Trichoderma species form endophytic associations within Theobroma cacao trichomes. Mycol. Res. 2009;113:1365–1376. 10.1016/j.mycres.2009.09.004. [Abstract] [CrossRef] [Google Scholar]
104. Glushakova AM, Yurkov AM, Chernov IY. Massive isolation of anamorphous ascomycete yeasts Candida oleophila from plant phyllosphere. Microbiology. 2007;76:799–803. 10.1134/S0026261707060215. [Abstract] [CrossRef] [Google Scholar]
105. Hoeksema JD, Forde SE. A Meta-analysis of factors affecting local adaptation between interacting species. Am. Nat. 2008;171:275–290. 10.1086/527496. [Abstract] [CrossRef] [Google Scholar]
106. Bálint M, et al. Host genotype shapes the foliar fungal microbiome of balsam poplar (Populus balsamifera) PLoS One. 2013;8:e53987. 10.1371/journal.pone.0053987. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
107. Brooks AW, Kohl KD, Brucker RM, van Opstal EJ, Bordenstein SR. Phylosymbiosis: relationships and functional effects of microbial communities across host evolutionary history. PLoS Biol. 2016;14:e2000225. 10.1371/journal.pbio.2000225. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
108. Deveautour C, Donn S, Power SA, Bennett AE, Powell JR. Experimentally altered rainfall regimes and host root traits affect grassland arbuscular mycorrhizal fungal communities. Mol. Ecol. 2018;27:2152–2163. 10.1111/mec.14536. [Abstract] [CrossRef] [Google Scholar]
109. Giauque H, Hawkes CV. Historical and current climate drive spatial and temporal patterns in fungal endophyte diversity. Fungal Ecol. 2016;20:108–114. 10.1016/j.funeco.2015.12.005. [CrossRef] [Google Scholar]
110. Hawkes CV, et al. Fungal community responses to precipitation. Global Change Biol. 2011;17:1637–1645. 10.1111/j.1365-2486.2010.02327.x. [CrossRef] [Google Scholar]
111. Wagner MR, et al. Host genotype and age shape the leaf and root microbiomes of a wild perennial plant. Nat. Commun. 2016;7:12151. 10.1038/ncomms12151. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
112. Barberan A, Bates ST, Casamayor EO, Fierer N. Using network analysis to explore co-occurrence patterns in soil microbial communities. ISME J. 2012;6:343–351. 10.1038/ismej.2011.119. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
113. Jiang Y, et al. Plant cultivars imprint the rhizosphere bacterial community composition and association networks. Soil Biol. Biochem. 2017;109:145–155. 10.1016/j.soilbio.2017.02.010. [CrossRef] [Google Scholar]
114. Blakeman JP, Fokkema NJ. Potential for biological control of plant diseases on the phylloplane. Annu. Rev. Phytopathol. 1982;20:167–190. 10.1146/annurev.py.20.090182.001123. [CrossRef] [Google Scholar]
115. Shi S, et al. The interconnected rhizosphere: high network complexity dominates rhizosphere assemblages. Ecol. Lett. 2016;19:926–936. 10.1111/ele.12630. [Abstract] [CrossRef] [Google Scholar]
116. Adame-Álvarez RM, Mendiola-Soto J, Heil M. Order of arrival shifts endophyte–pathogen interactions in bean from resistance induction to disease facilitation. FEMS Microbiol. Lett. 2014;355:100–107. 10.1111/1574-6968.12454. [Abstract] [CrossRef] [Google Scholar]
117. May G, Nelson P. Defensive mutualisms: do microbial interactions within hosts drive the evolution of defensive traits? Funct Ecol. 2014;28:356–363. 10.1111/1365-2435.12166. [CrossRef] [Google Scholar]
118. Trillas MI, Segarra G. Interactions between nonpathogenic fungi and plants. Adv. Bot. Res. 2009;51:321–359. 10.1016/S0065-2296(09)51008-7. [CrossRef] [Google Scholar]
119. Conrath U, Pieterse CMJ, Mauch-Mani B. Priming in plant-pathogen interactions. Trends Plant Sci. 2002;7:210–216. 10.1016/S1360-1385(02)02244-6. [Abstract] [CrossRef] [Google Scholar]
120. Knapp AK, et al. Rainfall variability, carbon cycling, and plant species diversity in a mesic grassland. Science. 2002;298:2202–2205. 10.1126/science.1076347. [Abstract] [CrossRef] [Google Scholar]
121. Morriën E, et al. Soil networks become more connected and take up more carbon as nature restoration progresses. Nat. Commun. 2017;8:14349. 10.1038/ncomms14349. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
122. Deng Y, et al. Molecular ecological network analyses. BMC Bioinformatics. 2012;13:113. 10.1186/1471-2105-13-113. [Europe PMC free article] [Abstract] [CrossRef] [Google Scholar]
123. Huang WY, Cai YZ, Hyde KD, Corke H, Sun M. Biodiversity of endophytic fungi associated with 29 traditional Chinese medicinal plants. Fungal Divers. 2008;33:61–75. [Google Scholar]
124. Mapperson RR, Kotiw M, Davis RA, Dearnaley JDW. The diversity and antimicrobial activity of Preussia sp. endophytes isolated from Australian dry rainforests. Curr. Microbiol. 2014;68:30–37. 10.1007/s00284-013-0415-5. [Abstract] [CrossRef] [Google Scholar]
125. Teles HL, et al. Benzopyrans from Curvularia sp., an endophytic fungus associated with Ocotea corymbosa (Lauraceae) Phytochemistry. 2005;66:2363–2367. 10.1016/j.phytochem.2005.04.043. [Abstract] [CrossRef] [Google Scholar]
126. Murali TS, Suryanarayanan TS, Geeta R. Endophytic Phomopsis species: host range and implications for diversity estimates. Can. J. Microbiol. 2006;52:673–680. 10.1139/w06-020. [Abstract] [CrossRef] [Google Scholar]
127. Chokpaiboon S, Choodej S, Boonyuen N, Teerawatananond T, Pudhom K. Highly oxygenated chromones from mangrove-derived endophytic fungus Rhytidhysteron rufulum. Phytochemistry. 2016;122:172–177. 10.1016/j.phytochem.2015.12.010. [Abstract] [CrossRef] [Google Scholar]
128. Krohn K, et al. Fusidilactones, a new group of polycyclic lactones from an endophyte, Fusidium sp. Eur. J. Org. Chem. 2002;2002:2331–2336. 10.1002/1099-0690(200207)2002:14<2331::AID-EJOC2331>3.0.CO;2-P. [CrossRef] [Google Scholar]
129. Kumar DSS, Hyde KD. Biodiversity and tissue-recurrence of endophytic fungi in Tripterygium wilfordii. Fungal Divers. 2004;17:69–90. [Google Scholar]
Articles from Scientific Reports are provided here courtesy of Nature Publishing Group

Full text links 

Read article at publisher's site: https://doi.org/10.1038/s41598-020-57622-6

Read article for free, from open access legal sources, via Unpaywall: https://www.nature.com/articles/s41598-020-57622-6.pdfUnpaywall

Citations & impact 

Impact metrics

7
Citations
Jump to Citations

Citations of article over time

Smart citations by scite.ai
Smart citations by scite.ai include citation statements extracted from the full text of the citing article. The number of the statements may be higher than the number of citations provided by EuropePMC if one paper cites another multiple times or lower if scite has not yet processed some of the citing articles.
Explore citation contexts and check if this article has been supported or disputed.
https://scite.ai/reports/10.1038/s41598-020-57622-6

Supporting
Mentioning
Contrasting
0
15
0

Article citations

Go to all (7) article citations

Data 

Data behind the article

This data has been text mined from the article, or deposited into data resources.

BioStudies: supplemental material and supporting data

BioProject

Nucleotide Sequences

Similar Articles 

To arrive at the top five similar articles we use a word-weighted algorithm to compare words from the Title and Abstract of each citation.