Abstract
Within their tissues, plants host diverse communities of fungi, termed fungal endophytes. These fungi can affect plant growth, competitiveness, and resistance to stressors, thereby influencing plant community structure. Research characterizing fungal endophyte communities has so far mostly focused on seed plants, but information on the endophytes of other plant lineages is needed to understand how plant microbiomes impact whole ecosystems and how major changes through land plant evolution have affected plant-microbe relationships. In this study, we assess the fungal endophyte community of the model liverwort Marchantia polymorpha L. by both culturing and Illumina amplicon sequencing methods. We detect a very diverse fungal community that is distinct between M. polymorpha patches and only shares a few core fungi between populations across the United States. We also show low overlap in taxa detected by the different methods. This study helps build a foundation for using M. polymorpha and other Marchantia species as models for the ecology and dynamics of bryophyte microbiomes.
Similar content being viewed by others
Avoid common mistakes on your manuscript.
1 Introduction
Plants harbor diverse microbiomes of bacteria, archaea, fungi, and other eukaryotic microbes (Berg et al. 2015). The microbes living inside healthy plant tissues are termed endophytes (Stone et al. 2000). In the case of fungi, this broad definition includes mycorrhizal fungi, latent pathogens and saprobes, and a diverse array of other fungi with functions across the spectrum from mutualist to pathogen (Porras-Alfaro and Bayman 2011; Schulz and Boyle 2005). Fungal endophytes have a wide range of effects on host success including changes in growth, competitive ability, and resistance to pathogens or stressors (Porras-Alfaro and Bayman 2011). The many effects of fungal symbionts make them important, if inconspicuous drivers of plant community structure and therefore broader ecosystem function (Afkhami and Strauss 2016; Aguilar-Trigueros and Rillig 2016; Rudgers et al. 2004).
Fungal endophytes have been surveyed across a growing number of host plants and habitats, revealing diverse and complex communities that assemble based on host identity, geographic location, elevation, season, and plant tissue (Arnold 2007; Kivlin et al. 2017; Peršoh 2015). The focus, however, has been on seed plants, with much less work on ferns, lycophytes, and the non-vascular bryophytes (mosses, liverworts, and hornworts). The differences in evolutionary history, life strategies, life cycles, and ecology between seed plants and the rest of the land plant lineages may lead to distinct microbiome dynamics. Indeed, sampling that has assessed bryophyte and vascular plant culturable endophytes from the same sites has hinted at separation between their communities (U’Ren et al. 2012).
Associations with fungi are also thought to have been important in allowing plants to initially colonize land (Delaux et al. 2012; Heckman et al. 2001; Taylor and Krings 2005), making bryophyte fungal symbionts evolutionarily important to study because of the bryophytes’ early-branching positions in the embryophyte clade (Shaw and Renzaglia 2004). Mycorrhiza-like fungal symbioses have been found in hornworts and liverworts, but mosses appear to lack them (Desirò et al. 2013; Pressel et al. 2010). Experiments in liverworts have demonstrated symbiotic nutrient exchange with mycorrhizal fungi belonging to Glomeromycotina, Mucoromycotina, and Ascomycota (Field et al. 2012; Field et al. 2016; Kowal et al. 2018). Research on microbes associated with mosses has instead focused mostly on parasites and pathogens (Döbbeler 1997; Davey and Currah 2006; Akita et al. 2011). In contrast to such work on individual symbionts, only a few researchers have attempted to characterize the diversity of the whole fungal communities found in liverworts (Davis et al. 2003; Davis and Shaw 2008; Knack et al. 2015) or mosses (Davey et al. 2013; Davey et al. 2012; Davey et al. 2014; Davey et al. 2017).
In this study, we explored the diversity of fungal endophytes in wild populations of the liverwort Marchantia polymorpha L. This widely distributed liverwort has been used as a model for botanical research for well over a century and is now emerging as a molecular model system with a recently published genome sequence (Bowman et al. 2016; Bowman et al. 2017). This species appears to contain three genetic groups having distinct ecologies and, based on morphology and isozyme assessment, these have been designated as M. polymorpha subspp. polymorpha, ruderalis, and montivagans (Bischler-Causse and Boisselier-Dubayle 1991; Boisselier-Dubayle et al. 1995). The subspecies used for the recent genome sequence and molecular work is ruderalis, but resequencing to investigate the diversity of the broader species is underway. The broad range and variable ecology of M. polymorpha make it an attractive candidate for studies on how microbiome assembly depends on geography, disturbance, and other ecological factors. The availability of genetic tools and the ease of growing M. polymorpha and other Marchantia species axenically in the lab also make the genus a promising model for experimental investigations of microbial interactions (Bowman et al. 2016).
Much of the mycorrhizal symbiosis work on liverworts thus far has not focused on M. polymorpha itself since it is not often found to harbor arbuscular mycorrhizal fungi (Ligrone et al. 2007; Pressel et al. 2010). Differences in fungal colonization between the three subspecies of M. polymorpha have been reported, with Glomeromycotina associations only detected in subsp. montivagans (Ligrone et al. 2007). On the other hand, M. paleacea, the sister species of M. polymorpha, does form mycorrhizal symbioses and is being used as a model for these interactions (Field et al. 2012; Humphreys et al. 2010; Radhakrishnan 2017). The comparison of the unusual lack of mycorrhizal association in some M. polymorpha to other Marchantia species promises to provide interesting insights into the effects of this dominant symbiosis on microbiome dynamics in bryophytes.
The microbiomes of Marchantia liverworts beyond mycorrhizal fungi are only just beginning to be investigated. Hipol et al. (2015) isolated a few culturable fungal endophytes from M. polymorpha in Baguio City, Philippines. Bacterial communities have been assessed by amplicon sequencing for M. polymorpha and M. paleacea near Oxtlapa, Mexico (Alcaraz et al. 2018) and Marchantia inflexa from Trinidad (Marks et al. 2018).
In this study, we collected M. polymorpha plants of all three subspecies as defined by Bischler-Causse and Boisselier-Dubayle (1991), and surveyed their endophytes both by culturing fungi and using direct Illumina amplicon sequencing across a broad geographic area in the United States. This dataset provides a broader sampling of the microbiome of this model species and will serve as a starting point for developing experiments on microbial interactions using M. polymorpha, as well as allowing future assessments of patterns of microbiome composition between major plant lineages.
The specific goals of this study were (1) to characterize the fungal endophyte communities of M. polymorpha, (2) to determine how its fungal endophyte communities vary between geographic locations, and (3) to directly compare the results of culturing and amplicon sequencing methods.
2 Materials and methods
2.1 Collections
Twenty-four patches of M. polymorpha were collected from 16 sites in eight US states and one Canadian territory (Fig. 1) from October 2013 to August 2015. At each site, one to four patches, each approximately 8 cm in diameter, were collected (Table 1). One of the sites (AUX) was sampled in both 2014 and 2015; all other sites were only sampled once. All three M. polymorpha subspecies were collected: subsp. polymorpha was found at the two Vermont sites (BFP, WRG), subsp. montivagans was found at three Oregon sites and one Washington site (BCB, LCB, PXF, MSH), and all other collections were subsp. ruderalis (Table 1).
For testing hypotheses about broad geographic patterns, sampling sites were grouped into three major regions: southeast (North Carolina, Tennessee, Kentucky) with four sites, northeast (Pennsylvania, New York, Vermont) with five, and northwest (Oregon, Washington, the Yukon) with seven (Fig. 1). All plant voucher specimens were deposited into the L.E. Anderson Bryophyte Herbarium at Duke University (DUKE).
After collection, plants with adhering soil were refrigerated in plastic bags until being cleaned and processed, no longer than 2 days (except the Canadian sample which was shipped live first). Plants were washed in sterile deionized water then surface sterilized. Surface sterilization was based on the methods of Arnold (2002). Plant fragments were immersed in 95% ethanol for 30 s, 10% bleach for 2 min, and 70% ethanol for 2 min, then dried on an ethanol treated Kimwipe.
2.2 Endophyte isolation and identification
Briefly, fungi were isolated on 1.5% malt extract agar (MEA) from plant fragments approximately 2 mm2 in area. Fungi were isolated from three tissue types: thallus, rhizoids (unicellular holdfasts), and gametangiophores (gametophytic receptacles for reproductive structures). A few additional cultures were isolated from tiny fungal fruiting bodies found on M. polymorpha thalli. Isolation and culturing of endophytes was conducted as previously described (Nelson et al. 2018).
DNA was extracted from pure fungal cultures with a cetyl trimethylammonium bromide (CTAB) protocol as previously described (Nelson et al. 2018). A segment of the rRNA gene including the internal transcribed spacer region (ITS) and part of the large subunit (LSU) was then amplified by PCR using the primers ITS1f (Gardes and Bruns 1993) and LR3 (Vilgalys and Hester 1990) and the amplicons were Sanger sequenced, as previously described (Nelson et al. 2018). Fungal isolates were identified by comparing their ITS1f-LR3 sequences to the National Center for Biotechnology Information (NCBI) GenBank database with BLAST searches using a 97% similarity threshold. When multiple isolates were at least 97% similar to each other, they were considered the same species. If multiple isolates formed a grade such that all samples were 97% similar to some but not all isolates in the grade with no independent groupings that could be split out, the whole grade was treated as one species.
2.3 DNA extraction from M. polymorpha
Whole M. polymorpha thalli were surface sterilized using the same protocol described above for plant fragments. Sterilized plants were stored in sterile 2 mL tubes at −80 °C. Frozen whole liverwort thalli were ground in 1.2 mL tubes using two 3.5 mm stainless steel balls and a mix of 2 mm and 0.7 mm zirconia beads on a Geno/Grinder. 500 μL of 2% CTAB buffer with 2% β-mercaptoethanol was added to each sample. Thereafter, the extraction followed the same protocol as used for fungal culture extractions. DNA extract concentrations were determined using a Qubit 2.0 fluorimeter (Thermo Fisher) with the High Sensitivity DNA assay. All extracts with a concentration higher than 8 ng/μL were diluted to 5 ng/μL before PCR amplification.
2.4 Amplicon sequencing
Mock community positive controls were created to run with the samples for assessing quality and biases in the library preparation and sequencing process (Nguyen et al. 2015). Mock communities included DNA from a set of 19 taxonomically diverse endophytes isolated from M. polymorpha (https://doi.org/10.6084/m9.figshare.7560434). Two mock communities were constructed: the first with equal ng amounts of raw fungal DNA extract and the second with equal ng amounts of the amplicon from ITS1f to LR3 in order to remove possible discrepancies from differences in copy number of the rRNA genes between fungi.
Mock community DNA mixes, extracts from whole M. polymorpha plants, and negative controls (water) were used for PCR targeting two fungal rRNA gene regions: ITS (ITS1, 5.8S, ITS2) using the primers ITS1f (Gardes and Bruns 1993) and ITS4 (Innis et al. 2012) and part of the LSU using primers LROR (Bunyard et al. 1994) and LR3 (Vilgalys and Hester 1990). Two replicate libraries were prepared in parallel for each sample.
Libraries of amplicons were prepared in three separate PCR stages using frame-shifted primers to improve Illumina data quality by increasing sequence diversity, based on the methods of Lundberg et al. (2013). In the first stage, all samples were amplified using standard primer pairs for 15 cycles to enrich for the target region. In the second stage, Illumina sequencing adaptors were added in 10 cycles of PCR using 2 μL of enriched sample. Adaptor primers in this step were mixes of six frameshifted primers to increase sequence diversity for Illumina sequencing. In the third stage, unique barcodes and Illumina adaptors were added to each sample in a final 10 cycle PCR. This final barcode step was set up in a sterile flow hood to prevent cross-contamination. For details of PCR recipes and programs, see Nelson (2017).
After library preparation, concentrations of all samples were determined with a Qubit fluorimeter. Samples were used if they had at least 5 ng/μL concentration. This resulted in 94 usable amplicon samples from the original 140 plants, of which all samples had at least one replicate for LSU and 73 samples had at least one for ITS. Amounts of DNA from all samples and controls were normalized and pooled to 50 ng per sample based on Qubit readings. The pooled amplicons were cleaned using Agencourt AmpPure XP beads at a 1.2:1 ratio of bead solution to sample to remove sequences shorter than the target amplicons. The prepared library was sequenced on an Illumina MiSeq at the Duke Center for Genomic and Computational Biology.
2.5 Amplicon data analysis
Reads obtained from the MiSeq library were demultiplexed using QIIME (Caporaso et al. 2010), and split by primer and cleaned using cutadapt (Martin 2011). Adaptors were removed and reads were quality filtered using Trimmomatic (Bolger et al. 2014). The remaining processing steps were conducted with VSEARCH (Rognes et al. 2016): reads were filtered by length and quality and dereplicated, singletons and chimeras were removed; operational taxonomic units (OTUs) were clustered at 95% similarity for ITS and 96% for LSU; OTUs with fewer than 10 reads were removed; and reads were assigned to OTUs using a 95% threshold for ITS reads and 96% for LSU reads. Clustering thresholds were selected by running the data processing with multiple thresholds and comparing the accuracy of the mock community recovery. Only forward reads were used because reverse reads were low quality and poorly recovered the mock community. Code for data processing pipeline is provided at https://doi.org/10.6084/m9.figshare.7544852.
For ITS reads, taxonomy was assigned with mothur implemented in QIIME (Schloss et al. 2009) using the UNITE database (Abarenkov et al. 2010). For LSU reads, taxonomy was assigned with the RDP classifier web portal (Wang et al. 2007). Identifications for taxa of interest were refined using individual BLAST searches of the NCBI database.
To filter out unreliable data, the number of reads appearing in sequenced negative controls was checked. The maximum number of reads for an OTU in any negative control was four, so all cells in the OTU table with four or fewer reads were changed to zero. Any OTUs with no reads remaining after this filtering step were removed. Based on automated taxonomy assignments and BLAST searches, non-fungal OTUs were removed. The LSU amplicon detected many non-fungal organisms and these sequences were separated out and their identifications refined with BLAST searches. Any samples with no reads remaining after restricting the data to fungi were removed. Replicate PCR samples for the same plants were pooled by adding read counts for each OTU. After pooling, samples with too few reads were removed using a threshold of 500 for LSU samples and 1000 for ITS samples because the LSU samples resulted in fewer fungal sequences on average. The OTU tables were normalized by converting all read counts into percent of total reads for the sample. OTUs with mean abundances under one per million after normalization were removed.
To determine similarities between the sequenced fungal communities, Pearson correlation coefficients were calculated in R 3.4.4 (Team RC 2018) and Bray-Curtis and Jaccard indices were calculated with the R vegan package (Oksanen et al. 2017). Downstream results from both Bray-Curtis and Jaccard indices were similar and we chose to present the Jaccard index which does not use OTU abundances. This was based on the observation that read counts did not accurately reflect known starting amounts of fungi in the mock community data. Principal coordinate analysis (PCoA) and non-metric multidimensional scaling (NMDS) using the MASS package (Venables and Ripley 2002) were conducted in R and visualized with the ggplot2 (Wickham 2009) and plyr (Wickham 2011) packages. Significance of separation between communities from different sites and from the three North American regions sampled (northwest, northeast, and southeast) was determined with a permutational analysis of variance (PERMANOVA) using the vegan package in R. The R code used for all data analyses and visualization is provided at https://doi.org/10.6084/m9.figshare.7542095.
2.6 Fungal phylogeny
A phylogenetic analysis of the sequences from the fungal cultures and LSU amplicons was conducted using the online T-BAS portal applying maximum likelihood with RAxML and default settings to place query sequences into a reference fungal phylogeny using the nucLSU dataset (Carbone et al. 2016).
2.7 Estimating minimum total species richness
To estimate how many species or OTUs would be likely to be found with sufficient sampling in each method, the chao1 index was calculated using the vegan command specpool() in R. For amplicon datasets, mock community reads were removed before calculating chao1. For culture data, cultures from the Yukon samples were removed since their initial handling differed from the other samples.
2.8 Identifying core endophyte community members
We identified endophytes that were common across our sampling to determine core members of the M. polymorpha fungal community both at local (site) and broad (full study) scales. For individual sites in the amplicon datasets with data for at least five plants, core members of the local community were defined as fungi found in all plants from that site with an average abundance over 1%. For the culture data set, because so few fungi were isolated multiple times, we defined core members of the M. polymorpha fungal community as ones isolated from more than one site. For the full amplicon datasets, fungi were considered core members if they were detected in at least 50% of sites and had an average read abundance over 1%.
3 Results
3.1 Diversity of fungi isolated in culture
A total of 93 fungal isolates were obtained from M. polymorpha plants (Genbank accessions MG004722-MG004798 and MK359672-MK359687), seven from surficial fruiting bodies and the remainder from surface-sterilized tissues. Fungi were obtained from 29 of 43 plants sampled. Pieces from a single plant yielded up to six fungal species. Fungi were isolated from 14.5% of the 592 surface-sterilized plant fragments. This isolation rate varied among sites, from 1.9% recovery from one patch in Vermont (WRG2 with only one isolate), to 68.4% isolation rate from one Kentucky collection. Plant tissues also had differing isolation rates with gametangiophores yielding the highest (27.9%), thalli intermediate (14.1%), and rhizoids the lowest (7.6%). This is reflected in the observation that fertile material often yielded higher total isolate numbers. Isolation rates and identities show no clear correlation with time of sampling or sex of host plant.
The isolated fungi represented at least 50 distinct species and only 15 species were isolated more than once. For the endophytes, calculation of chao1 indicated 35.3% completeness of species sampling. Six of the isolated species could not be identified past class. The vast majority of the fungal cultures belonged to Ascomycota; only two Basidiomycota cultures were obtained. Six classes of Ascomycota were represented: Eurotiomycetes, Pezizomycetes, Saccharomycetes, Leotiomycetes, Dothideomycetes, and, most abundantly, Sordariomycetes. Sordariomycetes were found in all M. polymorpha patches that yielded more than a single isolate, and at two sites they were the only isolated class (NCB, UTN) (Fig. 2). At finer taxonomic scales, no individual taxon predominated across all or most sites. As indicated by the small number of species found at more than a single site (nine), there was little species overlap between sampling locations. Only Phoma herbarum and Candida sake were found in both western and eastern North America (although the western isolate of P. herbarum came from the Canadian sample which had longer before refrigeration); all other isolates found at multiple sites were restricted to one sampling region (northeast, southeast, or northwest).
Some repeatedly isolated fungi were only found in a particular tissue. Phoma herbarum and Colletotrichum spp. only came from gametangiophores. Species isolated from different tissues were mostly non-overlapping and only one fungus was isolated from rhizoids, gametangiophores, and thalli: Xylaria cubensis.
3.2 Amplicon sequencing diversity
The Illumina MiSeq run yielded 22,837,264 raw paired-end reads (SRA BioProject PRJNA514318). After filtering, the LSU and ITS datasets contained 463,809 and 1,636,556 reads, respectively. After all filtering, the LSU data yielded 367 OTUs and the ITS data 891 (OTU tables available at https://doi.org/10.6084/m9.figshare.c.4352507). The final sampling included data from 69 plants, seven of which only had LSU data and eight only had ITS, with each site having data for one to nine plants. The OTU accumulation curves for both amplicon data sets show only slight leveling (data not shown), and calculation of chao1 indicated that the sequencing recovered 72.4% (ITS) and 78.2% (LSU) of predicted OTUs detectable for the sampled area. Pearson correlations indicated in almost all cases that PCR replicates of the same plant samples had communities most similar to their paired replicate (data not shown).
Both ITS and LSU datasets were dominated by Ascomycota fungi, followed by Basidiomycota; some earlier-diverging fungi also were detected (Table 2, Fig. 3). Among the Ascomycota, the most abundant classes were Dothideomycetes, Leotiomycetes, and Sordariomycetes. Agaricomycetes were the most abundant Basidiomycota (Table 2).
PERMANOVA analyses comparing individual sites and regions (northwest, northeast, and southeast) indicate significant separation in fungal communities between both geographic units (p < 0.01), but Pearson correlations between all samples and PCoA and NMDS analyses show that differences between sites did not strongly correlate with geographic distance between them (Fig. 4). For example, both ordination methods place the Kentucky site (AUX) as most similar to sites in Washington (MSH, UPS) rather than the much closer sites in Pennsylvania or New York. Differences between sites did not clearly correlate with bioclimatic variables, subspecies of host plant populations, time of sampling, nor level of human influence on the habitat.
Core endophyte community members of individual sites varied, but all sites with enough plants sampled showed at least one core OTU found in all plants at that site. Core fungi for individual sites are listed in Table 3.
3.3 Comparison of fungal communities inferred from culturing and sequencing datasets
The fungi detected in M. polymorpha come from across the fungal tree of life with few major groups lacking representatives (Fig. 3). Fungal orders frequently recovered by both culturing and amplicon methods include Xylariales, Helotiales, and Pleosporales (all Ascomycota). The culture collection was greatly enriched in Sordariomycetes (Table 2), especially Xylariaceae which have been abundantly isolated from other liverworts (Davis et al. 2003), but Xylariaceae were rare in our amplicon data. There was little overlap in the core fungi identified from the full culturing and sequencing datasets, with the exception of Phoma (Table 4). This fungus was isolated multiple times and from the most widespread set of sites of any species in the culture dataset (Kentucky, Vermont, and the Yukon). It was also persistent over time, as demonstrated by its isolation from the Kentucky population in two subsequent years. Comparing amplicon OTUs to the culture sequences indicates that some of the OTUs consistently detected at high abundances are likely this same species of Phoma. One OTU of Phoma from the ITS dataset was found in every site with an average of over 20% of each site’s reads.
Sites for which both culturing and sequencing were conducted on plants collected at the same time provide a direct comparison of taxa recovered by the two methods. Collection of both data types was attempted at five sites: LCC and PXF in Oregon, AUX in Kentucky, LIB in New York, and WRG in Vermont. LCC yielded only one culture and too few reads to analyze. PXF only yielded one culture, but sequencing detected mostly sebacenales and chytrids in these plants, neither of which would have been likely to grow in the culture conditions used. The other three sites had enough data to compare, but showed little overlap between cultured and sequenced fungi, with the exception of Phoma herbarum at AUX. Of the OTUs that contributed on average more than 2% of reads at AUX, only two were putatively detected by both ITS and LSU: Agaricomycetes sp. and Sebacinales sp. Similarly, ITS and LSU at LIB shared Plectosphaerella sp., Tetracladium sp., and Pleosporales sp. WRG showed the greatest correspondence between abundant OTUs from ITS and LSU with half (five) of the OTUs overlapping: Agaricales sp., Botryosphaeriaceae sp., Helotiales sp., Tetracladium sp., Sebacinales sp., and Lachancea kluyveri (data not shown).
3.4 Non-fungal eukaryotes
Non-fungal taxa detected by LSU sequencing included oomycetes, chlorophyte algae, animals such as mites and nematodes, alveolates including Paramecium and Stentor, and other microbial eukaryote groups (Fig. 5). Two OTUs likely representing mites even appeared in the more fungal-specific ITS data set. LSU reads from trebouxiophycean algae, alveolates of the class Colpodea, and nematodes were detected across all sites. None of the non-fungal OTUs appeared at every site, but five were found in at least three quarters of the sites: one nematode, two chlorophyte algae, and two ciliophorans.
4 Discussion
Our sampling of M. polymorpha endophytes indicates a diverse community of fungi with high variation across space and even between habitats close to one another. Some of the taxa repeatedly cultured or frequently detected by sequencing have no close relatives in sequence databases, making M. polymorpha a potential source for discovering new fungi.
4.1 Dominant and notable members of the M. polymorpha mycobiome
One fungus stood out as common between culturing and amplicon data sets: Phoma herbarum. This fungus is widespread in habitats ranging from terrestrial plants to marine fish (Ross et al. 1975). A strain of this species isolated as a soybean endophyte promoted growth of rice and soybean (Hamayun et al. 2009). A close relative of P. herbarum, Phoma muscivora, is a pathogen identified in boreal mosses (Davey and Currah 2009; Davey et al. 2009). Phoma muscivora produces enzymes capable of degrading bryophyte cell walls and may function as an opportunistic saprobe as well as a pathogen of live mosses (Davey and Currah 2009). Tests of the effects of P. herbarum isolates on M. polymorpha under controlled laboratory conditions do not indicate strong benefits or detriments (Nelson et al. 2018), but it may be that this widespread associate of M. polymorpha has effects only induced under natural conditions or certain stressors.
All datasets indicated that fungi from the order Helotiales were common (Table 4). Many of these could not be further identified, but one that could be identified to species stood out for its previously known association with M. polymorpha: Pezoloma marchantiae (Egertová et al. 2015; Garcia and Van Vooren 2005). Fruiting bodies with the macroscopic morphology of P. marchantiae were observed on plants collected from the Kentucky site in 2015 (Nelson 2017), which was reflected in the high recovery of P. marchantiae OTUs in the LSU dataset. This fungus is a close relative of the ericoid mycorrhizal fungus Pezoloma ericae and various Pezoloma species are also known to associate with bryophytes (Stenroos et al. 2010). These bryophyte-associated Pezolomas have been hypothesized to be saprobes specializing on bryophytes (Stenroos et al. 2010), but the recent demonstration of growth enhancement and nutrient transfer by Pezoloma ericae in two leafy liverwort hosts indicates that some fungi in the group could also function as mutualists (Kowal et al. 2018). Formation of fruiting bodies on plant hosts is sometimes interpreted as evidence of a parasitic function (Döbbeler 1997), but experiments with our P. marchantiae isolates under controlled conditions indicate no detrimental effects and even some growth enhancement of M. polymorpha (Nelson et al. 2018). If P. marchantiae has a mycorrhizal mutualism with M. polymorpha, this would likely only manifest in nutrient limiting conditions unlike those used in these growth experiments.
Another fungus previously known to associate with M. polymorpha is Loreleia marchantiae (Bresinsky and Schotz 2006). Mushrooms matching the macroscopic characteristics of L. marchantiae were found in the MSH2 site in Washington at the time of collection (Nelson 2017), and Mt. St. Helens was one of the locations where it was abundant in the ITS amplicon dataset. It can be found fruiting on a number of complex thalloid liverworts and has been observed to infect both types of M. polymorpha rhizoids: smooth (live at maturity) and pegged (dead at maturity) (Bresinsky and Schotz 2006; Guminska and Mierzenska 1992). Marchantia polymorpha plants infected with L. marchantiae have been observed to form more oil cells and more smooth rhizoids and to attract colonies of cyanobacteria to gather among their rhizoids (Guminska and Mierzenska 1992).
Consistent with previous reports that M. polymorpha does not form arbuscular mycorrhizal associations (Ligrone et al. 2007), the few Glomeromycetes we detected by amplicon sequencing were in very low abundance. But contrary to previous findings (Ligrone et al. 2007), we detected these Glomeromycete OTUs in all three subspecies of M. polymorpha. Methods specifically targeting arbuscular mycorrhizal fungi would be needed to confirm this result. Our sequencing did, however, detect OTUs of other mycorrhiza-forming fungi in high abundance at certain sites, notably fungi belonging to Sebacinales, Russulaceae, and Thelephoraceae. Kowal et al. (2016, 2018) demonstrated that the ericoid mycorrhizal fungus Pezoloma ericae forms mutualisms with leafy liverworts and can provide an effective source inoculum for ericaceous plants; it is thus possible that associations of various mycorrhizal fungi with M. polymorpha could provide similar services.
A number of common endophyte genera frequently cultured from diverse vascular hosts were conspicuously absent from our culture collections. These included Alternaria, Fusarium, Penicillium, and Tricoderma. Alternaria and Penicillium were detected in relatively low abundance by the amplicon sequencing and OTUs of Fusarium were abundantly detected in the LSU data set at the Pennsylvania site which had no culture sampling. In contrast, no Trichoderma were identified in the amplicon data. Trichoderma cultures might have been discarded as contamination due to their fast-growing “green mold” morphology, but this would not have changed the amplicon sampling and the absence of this genus that is common in soils and plants is notable (Harman et al. 2004).
4.2 Factors organizing fungal communities in M. polymorpha
The amplicon sequencing data indicate that M. polymorpha plants collected from a single patch usually had fungal communities most similar to other plants from the same patch. However, there was still significant variation within a patch, with few fungal OTUs being found in every plant from one patch. This could indicate stochasticity in the assembly of the fungal community for each separate M. polymorpha thallus, adjacent plants at different developmental stages having distinct communities, or inflated differences from incomplete sampling. Although there was some signal of differentiation by geographic region, distance between sites did not correlate strongly with differentiation in fungal communities. For example, NMDS analyses show two of the Mt. St. Helens (MSH) sites having fungi more similar to those found in Kentucky (AUX) and elsewhere in Washington (UPS) than the third MSH site. This spatial heterogeneity is consistent with previous studies of endophyte communities of a single host at multiple sites (Arnold et al. 2001; Glynou et al. 2016), though caution should be used in interpreting this variability since accumulation curves for the fungal diversity collected from M. polymorpha were not saturated. If the variability does reflect reality rather than incomplete sampling, it suggests that local environmental factors on a relatively small scale or stochastic processes are determining the pool of fungi available to colonize M. polymorpha. Local environmental controls may include soil characteristics, moisture, and the community of surrounding plants (Peay et al. 2016; Peršoh 2015).
Although we saw no clear correlation of sampling time with fungal community, it is likely that differences in seasonal conditions contributed to variability in the fungi detected, especially in the culture dataset which spanned multiple seasons and years. The site that we isolated cultures from in two consecutive years showed some continuity over time with three of the same taxa detected the first year also found the second year. The impacts of time and seasonal change on endophyte communities are so far underexplored, but the studies that have taken these variables into account show sometimes dramatic differences between years and even at scales as short as a month (Cordier et al. 2012; Davey et al. 2012; Junker et al. 2012; Peršoh 2015). Future repeated sampling of the same M. polymorpha populations over time would be needed to define the patterns and scale of its fungal endophyte community turnover.
Another factor that can structure endophyte communities is tissue specificity. Previous studies have observed tissue specificity in culturable fungi from conifer and angiosperm hosts, documenting differences between fungal assemblages of stems, bark, leaves, fruits, roots, and flowers (Arnold 2007; Junker et al. 2012; Wu et al. 2013; Yuan et al. 2009). None of these sporophytic structures are present in haploid-dominant bryophytes and tissue specificity of microbes has not been assessed for bryophytes. In our culturing, we separated M. polymorpha thalli, rhizoids, and gametangiophores and found some indication of tissue-specificity for certain fungi. Even without the complex, highly differentiated tissues found in tracheophytes, bryophytes may present multiple niches to their fungal inhabitants. Our observations suggest that differences even in the simpler tissues of a liverwort may be significant for structuring fungal communities at the small scale of bryophytes.
4.3 Comparison of culture and amplicon sampling of M. polymorpha endophytes
The enrichment of Sordariomycetes we observed in cultures relative to sequencing data has been observed previously using similar methods (Chen et al. 2018) and suggests either that the sequencing library preparation is biased against Sordariomyctes or that MEA culturing selects for them (or both). Overall, the extreme differences in abundance and limited overlap between culturing and sequencing data, and even between sequencing of two different gene regions, indicates that multiple methods must be combined to detect anything approaching the full fungal communities of plants. In addition, despite the improvements in estimated sampling completeness possible with amplicon sequencing over culturing, more intensive sampling is needed for all methods to sufficiently circumscribe the communities. Even with the combination of multiple sampling methods, many questions about the community composition remain. Our mock community sequencing control indicated that read abundances correlate poorly to initial DNA content from different taxa, making interpretation of dominance of certain taxa questionable. In order to address this issue, future work on plant microbiomes can employ direct sequencing metagenomics or taxonomically specific in situ hybridization approaches.
4.4 Non-fungal associates of M. polymorpha
The most abundant non-fungal eukaryotes detected by LSU amplicon sequencing include chlorophyte algae, animals such as nematodes and mites, and alveolates. Such organisms are known to be common in the limnoterrestrial habitats provided by bryophytes (Glime 2017). It is possible that some of these organisms were not inside the liverwort tissues, but rather adhering to the surface strongly enough to resist the washing and sterilization treatments. Stalked ciliates like Stentor have been reported to adhere to bryophyte surfaces (Glime 2013a), microarthropods could shelter in the air pores on the liverwort’s surface, and organisms like tardigrades and rotifers form resistant resting stages that might withstand washing. But it is also possible that some of the detected organisms could be inside the plant tissues. Nematodes might be inside galls formed in the plant (Glime 2013b) and some of the green algae may also been internal. Intracellular Coccomyxa-like algae have been reported to be widespread in Ginkgo biloba (Trémouillaux-Guiller and Huss 2007) and green algal endophytes have also been reported from a number of mosses (Reese 1981; Reese 1992). Little is known about the potential for most other non-fungal eukaryotic microbes to live as endophytes. So far, oomycete communities have been the main non-fungal eukaryotes assessed in plants, but Cercozoan diversity has also been detected in the rhizosphere and phyllosphere of Arabidopsis thaliana and other Brassicaceae (Ploch et al. 2016; Sapp et al. 2018). In addition, instead of being inside M. polymorpha directly, some of the detected taxa may be living inside each other. For example, trebouxiophycean algae can be intracellular symbionts of ciliates like Colpoda and Paramecium (Glime 2013a; Hoshina and Kusuoka 2016).
4.5 Concluding remarks
The global distribution of M. polymorpha and other Marchantia species makes this genus an attractive model for future research on host specificity and ecological factors determining microbiome assembly. Since M. polymorpha is often found in disturbed sites and is an early colonizer after fire (Bradbury 2006; Graff 1936), investigating its fungal communities along disturbance and successional gradients could elucidate how plant microbiomes respond to disturbance. The ease of growing axenic clones of Marchantia in the laboratory makes this genus a promising system for controlled experiments investigating the dynamics of complex endophytic communities that include not only of fungi and bacteria, but many other eukaryotes as well. Our study provides a foundation for such further investigations of liverwort microbiomes and use of the model M. polymorpha to answer questions about the ecology of plant microbiomes.
References
Abarenkov K, Henrik Nilsson R, Larsson KH, Alexander IJ, Eberhardt U, Erland S, Høiland K, Kjøller R, Larsson E, Pennanen T, Sen R, Taylor AFS, Tedersoo L, Ursing BM, Vrålstad T, Liimatainen K, Peintner U, Kõljalg U (2010) The UNITE database for molecular identification of fungi–recent updates and future perspectives. New Phytol 186:281–285
Afkhami ME, Strauss SY (2016) Native fungal endophytes suppress an exotic dominant and increase plant diversity over small and large spatial scales. Ecology 97:1159–1169
Aguilar-Trigueros CA, Rillig MC (2016) Effect of different root endophytic fungi on plant community structure in experimental microcosms. Ecol Evol 6:8149–8158
Akita M, Lehtonen MT, Koponen H, Marttinen EM, Valkonen JPT (2011) Infection of the Sunagoke moss panels with fungal pathogens hampers sustainable greening in urban environments. Sci Total Environ 409(17):3166–3173
Alcaraz LD, Peimbert M, Barajas HR, Dorantes-Acosta AE, Bowman JL, Arteaga-Vázquez MA (2018) Marchantia liverworts as a proxy to plants’ basal microbiomes. Sci Rep 8:12712
Arnold AE (2002) Neotropical fungal endophytes: diversity and ecology. Dissertation, The University of Arizona
Arnold AE (2007) Understanding the diversity of foliar endophytic fungi: progress, challenges, and frontiers. Fungal Biol Rev 21:51–66
Arnold AE, Maynard Z, Gilbert GS (2001) Fungal endophytes in dicotyledonous neotropical trees: patterns of abundance and diversity. Mycol Res 105:1502–1507
Berg G, Rybakova D, Grube M, Köberl M (2015) The plant microbiome explored: implications for experimental botany. J Exp Bot 67:995–1002
Bischler-Causse H, Boisselier-Dubayle M (1991) Lectotypification of Marchantia polymorpha L. J Bryol 16:361–365
Boisselier-Dubayle MC, Jubier MF, Lejeune B, Bischler H (1995) Genetic variability in the three subspecies of Marchantia polymorpha (Hepaticae): isozymes, RFLP and RAPD markers. Taxon 44(3):363
Bolger AM, Lohse M, Usadel B (2014) Trimmomatic: a flexible trimmer for Illumina sequence data. Bioinformatics 30:2114–2120
Bowman JL, Araki T, Kohchi T (2016) Marchantia: past, present and future. Plant Cell Physiol 57:205–209
Bowman JL et al (2017) Insights into land plant evolution garnered from the Marchantia polymorpha genome. Cell 171:287–304
Bradbury S (2006) Response of the post-fire bryophyte community to salvage logging in boreal mixedwood forests of northeastern Alberta, Canada. For Ecol Manag 234:313–322
Bresinsky A, Schotz A (2006) Behaviour in cultures and habitat requirements of species within the genera Loreleia and Rickenella (Agaricales). Acta Mycol 41:189–208
Bunyard BA, Nicholson MS, Royse DJ (1994) A systematic assessment of Morchella using RFLP analysis of the 28S ribosomal RNA gene. Mycologia 86:762–772
Caporaso JG, Kuczynski J, Stombaugh J, Bittinger K, Bushman FD, Costello EK, Fierer N, Peña AG, Goodrich JK, Gordon JI, Huttley GA, Kelley ST, Knights D, Koenig JE, Ley RE, Lozupone CA, McDonald D, Muegge BD, Pirrung M, Reeder J, Sevinsky JR, Turnbaugh PJ, Walters WA, Widmann J, Yatsunenko T, Zaneveld J, Knight R (2010) QIIME allows analysis of high-throughput community sequencing data. Nat Methods 7:335–336
Carbone I et al (2016) T-BAS: tree-based alignment selector toolkit for phylogenetic-based placement, alignment downloads and metadata visualization: an example with the Pezizomycotina tree of life. Bioinformatics 33:1160–1168
Chen KH, Liao HL, Arnold AE, Bonito G, Lutzoni F (2018) RNA-based analyses reveal fungal communities structured by a senescence gradient in the moss Dicranum scoparium and the presence of putative multi-trophic fungi. New Phytol 218:1597–1611
Cordier T, Robin C, Capdevielle X, Fabreguettes O, Desprez-Loustau ML, Vacher C (2012) The composition of phyllosphere fungal assemblages of European beech (Fagus sylvatica) varies significantly along an elevation gradient. New Phytol 196:510–519
Davey ML, Currah RS (2006) Interactions between mosses (Bryophyta) and fungi. Can J Bot 84(10):1509–1519
Davey ML, Currah RS (2009) Atradidymella muscivora gen. Et sp. nov.(Pleosporales) and its anamorph Phoma muscivora sp. nov.: a new pleomorphic pathogen of boreal bryophytes. Am J Bot 96:1281–1288
Davey ML, Tsuneda A, Currah RS (2009) Pathogenesis of bryophyte hosts by the ascomycete. Atradidymella muscivora. Am J Bot 96:1274–1280
Davey ML, Heegaard E, Halvorsen R, Ohlson M, Kauserud H (2012) Seasonal trends in the biomass and structure of bryophyte-associated fungal communities explored by 454 pyrosequencing. New Phytol 195:844–856
Davey ML, Heegaard E, Halvorsen R, Kauserud H, Ohlson M (2013) Amplicon-pyrosequencing-based detection of compositional shifts in bryophyte-associated fungal communities along an elevation gradient. Mol Ecol 22:368–383
Davey ML, Kauserud H, Ohlson M (2014) Forestry impacts on the hidden fungal biodiversity associated with bryophytes. FEMS Microbiol Ecol 90:313–325
Davey ML, Skogen MJ, Heegaard E, Halvorsen R, Kauserud H, Ohlson M (2017) Host and tissue variations overshadow the response of boreal moss-associated fungal communities to increased nitrogen load. Mol Ecol 26:571–588
Davis EC, Shaw AJ (2008) Biogeographic and phylogenetic patterns in diversity of liverwort-associated endophytes. Am J Bot 95:914–924
Davis EC, Franklin JB, Shaw AJ, Vilgalys R (2003) Endophytic Xylaria (Xylariaceae) among liverworts and angiosperms: phylogenetics, distribution, and symbiosis. Am J Bot 90:1661–1667
Delaux P-M, Nanda AK, Mathé C, Sejalon-Delmas N, Dunand C (2012) Molecular and biochemical aspects of plant terrestrialization. Perspect Plant Ecol Evol Syst 14:49–59
Desirò A, Duckett JG, Pressel S, Villarreal JC, Bidartondo MI (2013) Fungal symbioses in hornworts: a chequered history. Proc R Soc Lond B Biol Sci 280:20130207
Döbbeler P (1997) Biodiversity of bryophilous ascomycetes. Biodivers Conserv 6:721–738
Egertová Z, Eckstein J, Vega M (2015) Lamprospora tuberculata, Octospora ithacaensis, O. orthotrichi and O. affinis–four bryoparasitic ascomycetes new to the Czech Republic. Czech Mycol 67
Field KJ, Cameron DD, Leake JR, Tille S, Bidartondo MI, Beerling DJ (2012) Contrasting arbuscular mycorrhizal responses of vascular and non-vascular plants to a simulated. Palaeozoic CO2 decline. Nat Commun 3:835
Field KJ, Rimington WR, Bidartondo MI, Allinson KE, Beerling DJ, Cameron DD, Duckett JG, Leake JR, Pressel S (2016) Functional analysis of liverworts in dual symbiosis with Glomeromycota and Mucoromycotina fungi under a simulated Palaeozoic CO2 decline. ISME J 10:1514–1526
Garcia G, Van Vooren N (2005) Un discomycète inoperculé plutôt discret, Pezoloma ciliifera, et remarques sur le genre Pezoloma. Publications de la Société Linnéenne de Lyon 74:115–130
Gardes M, Bruns TD (1993) ITS primers with enhanced specificity for basidiomycetes-application to the identification of mycorrhizae and rusts. Mol Ecol 2:113–118
Glime JM (2013a) Protozoa diversity. In: Bryophyte ecology volume 2: byological interaction
Glime JM (2013b) Invertebrates: nematodes. In: Bryophyte ecology volume 2: bryological interaction
Glime JM (2017) The fauna: a place to call home. In: Bryophyte ecology volume 2: bryological interaction
Glynou K, Ali T, Buch AK, Haghi Kia S, Ploch S, Xia X, Çelik A, Thines M, Maciá-Vicente JG (2016) The local environment determines the assembly of root endophytic fungi at a continental scale. Environ Microbiol 18:2418–2434
Graff PW (1936) Invasion by Marchantia polymorpha following forest fires. Bull Torrey Bot Club 63:67–74
Guminska B, Mierzenska M (1992) Gerronema marchantiae Sing et Clem-a fungus associating with Marchantia polymorpha L and Nostoc sp Zeszyty Naukowe Uniwersytetu Jagiellońskiego Prace Botaniczne 24:171–177
Hamayun M, Khan SA, Khan AL, Rehman G, Sohn EY, Shah AA, Kim SK, Joo GJ, Lee IJ (2009) Phoma herbarum as a new gibberellin-producing and plant growth-promoting fungus. J Microbiol Biotechnol 19:1244–1249
Harman GE, Howell CR, Viterbo A, Chet I, Lorito M (2004) Trichoderma species—opportunistic, avirulent plant symbionts. Nat Rev Microbiol 2:43–56
Heckman DS, Geiser DM, Eidell BR, Stauffer RL, Kardos NL, Hedges SB (2001) Molecular evidence for the early colonization of land by fungi and plants. Science 293:1129–1133
Hipol R, Tamang SMA, Gargabite BF, Broñola Hipol R (2015) Diversity of fungal endophytes isolated from Marchantia polymorpha populations from Baguio City, Philippines bulletin of environment, Pharmacology and Life Sciences 4:87–91
Hoshina R, Kusuoka Y (2016) DNA analysis of algal endosymbionts of ciliates reveals the state of algal integration and the surprising specificity of the symbiosis. Protist 167:174–184
Humphreys CP, Franks PJ, Rees M, Bidartondo MI, Leake JR, Beerling DJ (2010) Mutualistic mycorrhiza-like symbiosis in the most ancient group of land plants. Nat Commun 1:103
Innis MA, Gelfand DH, Sninsky JJ, White TJ (2012) PCR protocols: a guide to methods and applications. Academic Press, London
Junker C, Draeger S, Schulz B (2012) A fine line–endophytes or pathogens in Arabidopsis thaliana. Fungal Ecol 5:657–662
Kivlin SN, Lynn JS, Kazenel MR, Beals KK, Rudgers JA (2017) Biogeography of plant-associated fungal symbionts in mountain ecosystems: a meta-analysis. Divers Distrib 23:1067–1077
Knack J et al (2015) Microbiomes of streptophyte algae and bryophytes suggest that a functional suite of microbiota fostered plant colonization of land. Int J Plant Sci 176:405–420
Kowal J, Pressel S, Duckett JG, Bidartondo MI (2016) Liverworts to the rescue: an investigation of their efficacy as mycorrhizal inoculum for vascular plants. Funct Ecol 30:1014–1023
Kowal J, Pressel S, Duckett JG, Bidartondo MI, Field KJ (2018) From rhizoids to roots? Experimental evidence of mutualism between liverworts and ascomycete fungi. Ann Bot 121:221–227
Ligrone R, Carafa A, Lumini E, Bianciotto V, Bonfante P, Duckett JG (2007) Glomeromycotean associations in liverworts: a molecular, cellular, and taxonomic analysis. Am J Bot 94:1756–1777
Lundberg DS, Yourstone S, Mieczkowski P, Jones CD, Dangl JL (2013) Practical innovations for high-throughput amplicon sequencing. Nat Methods 10:999–1002
Marks RA, Smith JJ, Cronk Q, McLetchie DN (2018) Variation in the bacteriome of the tropical liverwort, Marchantia inflexa, between the sexes and across habitats. Symbiosis 75:93–101
Martin M (2011) Cutadapt removes adapter sequences from high-throughput sequencing reads. EMBnet J 17:10–12
Nelson JM (2017) Diversity and effects of the fungal endophytes of the liverwort Marchantia polymorpha. Dissertation. Duke University
Nelson JM, Hauser DA, Hinson R, Shaw AJ (2018) A novel experimental system using the liverwort Marchantia polymorpha and its fungal endophytes reveals diverse and context-dependent effects. New Phytol 218:1217–1232
Nguyen NH, Smith D, Peay K, Kennedy P (2015) Parsing ecological signal from noise in next generation amplicon sequencing. New Phytol 205:1389–1393
Oksanen, J, Blanchet FG, Friendly M, Kindt R, Legendre P, McGlinn D, Minchin PR, O’Hara RB, Simpson GL, Solymos P, Stevens MH (2017) vegan: Community Ecology Package. R package version 2.4–3
Peay KG, Kennedy PG, Talbot JM (2016) Dimensions of biodiversity in the earth mycobiome. Nat Rev Microbiol 14:434–447
Peršoh D (2015) Plant-associated fungal communities in the light of meta’omics. Fungal Divers 75:1–25
Ploch S, Rose LE, Bass D, Bonkowski M (2016) High diversity revealed in leaf-associated protists (Rhizaria: Cercozoa) of Brassicaceae. J Eukaryot Microbiol 63:635–641
Porras-Alfaro A, Bayman P (2011) Hidden fungi, emergent properties: endophytes and microbiomes. Annu Rev Phytopathol 49:291–315
Pressel S, Bidartondo MI, Ligrone R, Duckett JG (2010) Fungal symbioses in bryophytes: new insights in the twenty first century. Phytotaxa 9:238–253
Radhakrishnan G (2017) Tracing the evolution of the arbuscular mycorrhizal symbiosis in the plant lineage. Doctoral dissertation, University of East Anglia
Reese WD (1981) " Chlorochytrium," a green alga endophytic in Musci. Bryologist 84:75–78
Reese WD (1992) More mosses with Chlorochytrium1. J Phycol 28:707–707
Rognes T, Flouri T, Nichols B, Quince C, Mahé F (2016) VSEARCH: a versatile open source tool for metagenomics. PeerJ 4:e2584
Ross A, Yasutake W, Leek S (1975) Phoma herbarum, a fungal plant saprophyte, as a fish pathogen. J Fish Res Board Can 32:1648–1652
Rudgers JA, Koslow JM, Clay K (2004) Endophytic fungi alter relationships between diversity and ecosystem properties. Ecol Lett 7:42–51
Sapp M, Ploch S, Fiore-Donno AM, Bonkowski M, Rose LE (2018) Protists are an integral part of the Arabidopsis thaliana microbiome. Environ Microbiol 20:30–43
Schloss PD, Westcott SL, Ryabin T, Hall JR, Hartmann M, Hollister EB, Lesniewski RA, Oakley BB, Parks DH, Robinson CJ, Sahl JW, Stres B, Thallinger GG, van Horn DJ, Weber CF (2009) Introducing mothur: open-source, platform-independent, community-supported software for describing and comparing microbial communities. Appl Environ Microbiol 75:7537–7541
Schulz B, Boyle C (2005) The endophytic continuum. Mycol Res 109:661–686
Shaw J, Renzaglia K (2004) Phylogeny and diversification of bryophytes. Am J Bot 91:1557–1581
Spatafora JW, Chang Y, Benny GL, Lazarus K, Smith ME, Berbee ML, Bonito G, Corradi N, Grigoriev I, Gryganskyi A, James TY, O’Donnell K, Roberson RW, Taylor TN, Uehling J, Vilgalys R, White MM, Stajich JE (2016) A phylum-level phylogenetic classification of zygomycete fungi based on genome-scale data. Mycologia 108:1028–1046
Stenroos S, Laukka T, Huhtinen S, Döbbeler P, Myllys L, Syrjänen K, Hyvönen J (2010) Multiple origins of symbioses between ascomycetes and bryophytes suggested by a five-gene phylogeny. Cladistics 26:281–300
Stone JK, Bacon CW, White J (2000) An overview of endophytic microbes: endophytism defined. Microbial Endophytes 3:29–33
Taylor TN, Krings M (2005) Fossil microorganisms and land plants: associations and interactions. Symbiosis 40:119–135
Team RC (2018) R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna
Trémouillaux-Guiller J, Huss VA (2007) A cryptic intracellular green alga in Ginkgo biloba: ribosomal DNA markers reveal worldwide distribution. Planta 226:553–557
U’Ren JM, Lutzoni F, Miadlikowska J, Laetsch AD, Arnold AE (2012) Host and geographic structure of endophytic and endolichenic fungi at a continental scale. Am J Bot 99:898–914
Venables WN, Ripley BD (2002) Random and mixed effects. In: Modern applied statistics with S Springer, pp 271–300
Vilgalys R, Hester M (1990) Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J Bacteriol 172:4238–4246
Wang Q, Garrity GM, Tiedje JM, Cole JR (2007) Naive Bayesian classifier for rapid assignment of rRNA sequences into the new bacterial taxonomy. Appl Environ Microbiol 73:5261–5267
Wickham H (2009) ggplot2: elegant graphics for data analysis. Springer-Verlag, New York
Wickham H (2011) The split-apply-combine strategy for data analysis. J Stat Softw 40:1–29
Wu L, Han T, Li W, Jia M, Xue L, Rahman K, Qin L (2013) Geographic and tissue influences on endophytic fungal communities of Taxus chinensis var. mairei in China. Curr Microbiol 66:40–48
Yuan Z-L, Chen Y-C, Yang Y (2009) Diverse non-mycorrhizal fungal endophytes inhabiting an epiphytic, medicinal orchid (Dendrobium nobile): estimation and characterization. World J Microbiol Biotechnol 25:295
Acknowledgements
The authors thank the following for assistance with field work logistics and facilities: G. Hermann and M. Metz (Lewis & Clark College); S. LaGreca and K. Hodge (Cornell University); K. McFarland (University of Tennessee); D. Allard and M. Tierney (University of Vermont); P. Ball (Oregon State University, Cascades); D. Taylor (Daniel Boone National Forest); B. Overton (Lock Haven University); A. DeMarais, M. Morrison, and B. Kirkpatrick (University of Puget Sound); C. Crisafulli (Pacific Northwest Research Station); K. Golinski (Smithsonian Institution); S. Heiney (North Carolina Botanical Garden). We thank R. Vilgalys, F. Lutzoni, F. Dietrich, and P. Manos for their advice on manuscript development. We also thank our undergraduate assistants: K. Atherton, C. Chen, R. Hinson, and S. Ou. Funding for this work was provided by grant no. DEB-1501826, U.S. National Science Foundation, an Anderson & Crum grant from the American Bryological and Lichenological Society, and a Grant in Aid of Research from the Duke University Biology Department.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Nelson, J., Shaw, A.J. Exploring the natural microbiome of the model liverwort: fungal endophyte diversity in Marchantia polymorpha L. Symbiosis 78, 45–59 (2019). https://doi.org/10.1007/s13199-019-00597-4
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s13199-019-00597-4