Ascomycetes associated with ectomycorrhizas: …...Ascomycetes associated with ectomycorrhizas: molecular diversity and ecology with particular reference to the Helotiales emi_2020
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Ascomycetes associated with ectomycorrhizas:molecular diversity and ecology with particularreference to the Helotialesemi_2020 3166..3178
Leho Tedersoo,1,2* Kadri Pärtel,1 Teele Jairus,1,2
Genevieve Gates,3 Kadri Põldmaa1,2 andHeidi Tamm1
1Department of Botany, Institute of Ecology and EarthSciences, University of Tartu, 40 Lai Street, 51005Tartu, Estonia.2Natural History Museum of Tartu University, 46Vanemuise Street, 51005 Tartu, Estonia.3Schools of Agricultural Science and Plant Science,University of Tasmania, Hobart, Tasmania 7001,Australia.
Summary
Mycorrhizosphere microbes enhance functioning ofthe plant–soil interface, but little is known of theirecology. This study aims to characterize the asco-mycete communities associated with ectomycorrhi-zas in two Tasmanian wet sclerophyll forests. Wehypothesize that both the phyto- and mycobiont,mantle type, soil microbiotope and geographical dis-tance affect the diversity and occurrence of the asso-ciated ascomycetes. Using the culture-independentrDNA sequence analysis, we demonstrate a highdiversity of these fungi on different hosts andhabitats. Plant host has the strongest effect on theoccurrence of the dominant species and communitycomposition of ectomycorrhiza-associated fungi.Root endophytes, soil saprobes, myco-, phyto- andentomopathogens contribute to the ectomycorrhiza-associated ascomycete community. Taxonomicallythese Ascomycota mostly belong to the orders Helo-tiales, Hypocreales, Chaetothyriales and Sordariales.Members of Helotiales from both Tasmania and theNorthern Hemisphere are phylogenetically closelyrelated to root endophytes and ericoid mycorrhizalfungi, suggesting their strong ecological and evolu-tionary links. Ectomycorrhizal mycobionts from Aus-tralia and the Northern Hemisphere are taxonomicallyunrelated to each other and phylogenetically distant
to other helotialean root-associated fungi, indicatingindependent evolution. The ubiquity and diversity ofthe secondary root-associated fungi should be con-sidered in studies of mycorrhizal communities toavoid overestimating the richness of true symbionts.
Introduction
Endophytic and mycorrhizosphere microbes, especiallyBacteria, Archaea and microfungi, synthesize plantgrowth regulators and vitamins facilitating the develop-ment and functioning of the mycorrhizal system in soil(Schulz et al., 2006). These root-associated microbessuch as mycorrhiza helper bacteria and the nitrogen-fixing actinobacteria and rhizobia differ substantially intheir function and ecology, including host preference pat-terns (Benson and Clawson, 2000; Sprent and James,2007; Burke et al., 2008). Of microfungi, foliar endo-phytes may considerably vary according to the special-ization to different host species and even organs(Neubert et al., 2006; Arnold, 2007; Higgins et al., 2007).On the contrary, facultative root-associating fungi suchas endophytes (e.g. the Phialocephala–Acephala andMeliniomyces–Rhizoscyphus complexes) form mostlynon-specific associations with many plant hosts (Vrålstadet al., 2002; Chambers et al., 2008), although host pref-erence may occur on the cryptic species level (Grüniget al., 2008).
Despite numerous studies on isolation and morphol-ogical identification of ascomycetous microfungi fromectomycorrhizal (EcM) root tips, their specificity for hostplants, fungi and substrate types remains unknown(Melin, 1923; Fontana and Luppi, 1966; Summerbell,1989; Girlanda and Luppi-Mosca, 1995). Molecular toolshave only recently been used to characterize and distin-guish the secondarily associated microfungi from EcMfungi in situ. These microfungi were identified from EcMroot tips by either cutting additional bands from the gel(Rosling et al., 2003; Tedersoo et al., 2006), using specificprimers (Urban et al., 2008) or cloning (Morris et al.,2008a,b; 2009; Wright et al., 2009). Cloning from DNAextracts comprising pooled individual EcM root tipsreveals many ascomycete taxa of uncertain ecologicalrole (Bergemann and Garbelotto, 2006; Smith et al.,
Received 9 April, 2009; accepted 22 June, 2009. *For correspon-dence. E-mail [email protected]; Tel. (+372) 7376222; Fax(+372) 7376222.
2007). Many of these are endophytic or rhizoplane colo-nists that are accidentally reported as forming mycorrhi-zas both in research publications and InternationalSequence Database (INSD) entries (Grünig et al., 2008).Such uncertain reports are especially common in ericoidmycorrhizas (ErM) and EcM for which universal fungal-specific primers are routinely used for the identification ofmycobionts.
Helotiales (Ascomycota) comprises the largest numberof undescribed root-associated fungi in addition toapproximately 2000 described species with contrastinglifestyles (Wang et al., 2006). Various subgroups ofHelotiales such as the Phialocephala–Acephala andRhizoscyphus–Meliniomyces complexes and Lachnumspp. are identified from EcM and arbutoid mycorrhiza inforest trees and subshrubs of the Northern Hemisphere(Vrålstad et al., 2002; Rosling et al., 2003; Tedersoo et al.,2003; 2007; 2008a; Bergemann and Garbelotto, 2006)and often erroneously reported as truly mycorrhizal.Indeed, the ecologically heterogeneous Rhizoscyphus–Meliniomyces and Phialocephala–Acephala complexesboth include distinct EcM-forming species nested withinnumerous pathogenic, ErM and root endophytic taxa(Vrålstad et al., 2002; Hambleton and Sigler, 2005;Münzenberger et al., 2009).
In previous EcM fungal community studies in Tasmania,we encountered frequent secondary colonization of EcMroot tips by Ascomycota besides the predominatelybasidiomycetous EcM fungi (Tedersoo et al., 2008b;2009). The present study was undertaken to identify anddistinguish these EcM-associated Ascomycota (EAA)from the true EcM-forming mycobionts using a culturing-independent approach. Utilizing the DNA extracts fromsingle EcM root tips with preidentified plant and EcMfungal hosts and developing several ascomycete-specificprimers, we hypothesized that these EAA have prefer-ence for either host tree, host fungus lineage, EcM mantletype, soil microbiotope, plot and site. Because of thedominance of Helotiales in this and previous studiesinvolving root-associated fungi, we addressed the phylo-genetic relations of helotialean EcM, ericoid mycorrhizal,endophytic and EAA isolates using the rDNA 28Ssequence data.
Results
Identification and distribution of EAA
Application of the newly designed taxon-specific primers(Fig. 1; Appendix 1) allowed us to specifically amplifyAscomycota from EcM root tips. Based on the rDNA ITSsequence analysis, 251 individuals of EAA were identifiedfrom 226 out of 675 (33.5%) analysed root tips (148individuals from the Mt. Field site and 103 from the Warrasite). 88.9% of the EcM root tips yielded a single ampliconof EAA. Based on the 99% ITS barcoding threshold, EAAwere assigned to 105 species, including 69 (65.7%)singletons and 15 (14.3%) doubletons (Appendix 2).
At both sites, species of EcM fungi and EAA wereaccumulating at similar rates with increasing samplingeffort (Fig. 2). The species accumulation curves hadstrongly overlapping confidence intervals (not shown)suggesting no substantial difference in EAA diversityamong mantle types, sites, plots, microsites or plant andfungal hosts. Similarly, there were no statistically signifi-cant differences in the relative frequency of colonization ofEAA among these habitats.
Trends in the distribution of eight most frequent EAAspecies were statistically analysed at both two sites(Table 1). Five of these species differed significantlyaccording to the site. Only Lecanicillium flavidum (syn.Verticillium fungicola var. flavidum) displayed a statisti-cally significant preference for EcM fungal lineage (Fish-er’s exact test: d.f. = 4; P = 0.002). This species occurredmore frequently on root tips colonized by members ofthe/cortinarius lineage, compared with the other four mostcommon EcM lineages. Due to elevated abundanceon the/cortinarius EcM, L. flavidum was more commonon plectenchymatous mantles than expected (d.f. = 1;P = 0.004). In contrast, Helotiales sp016 colonized exclu-sively mycorrhizas with pseudoparenchymatous mantles(d.f. = 1; P = 0.001).
Among the seven most common EAA species at Mt.Field, host trees and plots affected the distribution of sixand two species respectively (Table 1). Putative rootendophytes, root parasites and mycoparasites includedspecies with significant host plant preference. Forexample, Helotiales sp008 (putative endophyte, d.f. = 2;
Fig. 1. Map of primers used for amplification of the ITS and 28S rDNA. Newly designed primers are given in bold.
P = 0.001) and L. flavidum (mycoparasite; d.f. = 2;P = 0.043) occurred significantly more frequently on EcMof Eucalyptus regnans, while Hypocreales sp086 (rootparasite Neonectria cf. radicicola; d.f. = 2; P = 0.018)
preferred Pomaderris apetala compared with the othertwo hosts (Nothofagus cunninghamii is the third host). Atthe Warra site, only L. flavidum was significantly morecommon in the forest floor soil compared with decayedwood (d.f. = 1; P = 0.027).
At Mt. Field and Warra, respectively, the multivariatemodel explained 19.3% and 17.6% of the total variation inthe distribution of EAA. At both sites, the fungal lineageand mantle anatomy explained < 2% of the total variationthat remained non-significant. At Mt. Field, host plant andplot contributed 7.2% (SS = 141.2; P = 0.001) and 3.0%(SS = 59.0; P = 0.003) respectively. At Warra, substratetype contributed 3.5% to the total variation (SS = 34.7;P = 0.008).
Phylogenetic affinities of Tasmanian EAA
Based on blastN matches, the Tasmanian EAA belongedto 12 orders of Pezizomycotina. Helotiales, Hypocreales,Chaetothyriales and Sordariales comprised 54, 21, 9 and8 species respectively. Many of the hypocrealean taxawere assigned to parasitic lifestyle, including six putativemycoparasites (L. flavidum, Hypomyces spp.), four rootparasites (e.g. Neonectria radiciicola, Cylindrocarpon sp.)and two insect parasites (Cordyceps spp.). Sordarialesand Chaetothyriales, respectively, comprised mostlysaprotrophic and putatively endophytic members.
Helotiales comprised most of the dominant species thatwere assigned to the endophytic lifestyle based on ITSmatches (Appendix 2) and phylogenetic analysis (Fig. 3).The 30 species of Tasmanian EAA with available 28Ssequence formed 10 distinct, more or less supported lin-eages, including both monospecific branches and aggre-gates comprising up to 17 species (the Hyphodiscuscomplex) (Fig. 3). Helotialean EAA and root endophytesfrom Tasmania and the Northern Hemisphere were
00
10
20 40 60 80 100 120 140
20
30
40
50
60
70
80N
umbe
rof
spec
ies
Number of individuals
Fig. 2. Species accumulation curve of ectomycorrhizal (opensymbols) and ectomycorrhiza-associated ascomycetes (closedsymbols) at the Mt. Field (triangles) and Warra (circles) sites. Forclarity, the overlapping confidence intervals are not shown.
Table 1. Statistical significance of host plant, microbiotope, site and mantle type on the occurrence of eight most common species ofectomycorrhiza-associated ascomycetes. Significant P-values are indicated in bold.
Species Putative ecology
P-values of a Fisher’s exact test
Mt. Field Warra Both sitesa
Host plant(d.f. = 2)
Plot(d.f. = 2)
Micro-biotope(d.f. = 1)
Mantle type(d.f. = 1)
Site(d.f. = 1)
Helotiales sp007 Endophyte 0.018 0.091 na 0.401 < 0.001Helotiales sp008 Endophyte 0.001 0.007 na 0.241 0.342Helotiales sp013 Endophyte 0.554 0.193 na 0.540 0.004Helotiales sp016 Endophyte < 0.001 0.012 na 0.001 < 0.001Helotiales sp029 Endophyte na na 0.059 0.733 < 0.001Helotiales sp038 Endophyte 0.007 0.556 0.077 0.782 0.010Hypocreales sp086 Root parasite 0.018 0.279 na 0.542 0.635Lecanicillium flavidum Mycoparasite 0.043 0.125 0.027 0.003 0.854
a. Based on G-tests.Molecular identification of these species is shown in Appendix 2.na, not applied because of insufficient replication.
Fig. 3. Maximum-likelihood phylogram of Tasmanian ectomycorrhiza-associated ascomycetes (in bold) among identified taxa and otherplant-associated fungi within Helotiales. Fast bootstrap values > 70 and Bayesian posterior probabilities >95% are indicated below and abovethe branches respectively. Asterisks denote confirmed and putative EcM isolates, although it is possible that other EcM Helotiales exist.
phylogenetically closely related to ErM and saprobic iso-lates in the Hyphodiscus, Cryptosporiopsis–Neofabraeaand Rhizoscyphus–Meliniomyces complexes (Fig. 3,Appendix 2). Similarly, the majority of EAA sequencesfrom the Pinaceae and Fagaceae hosts in the NorthernHemisphere usually clustered with the well-recognizedendophytic and/or ErM lineages, whereas a few groupsformed monotypic lineages with yet unknown ecology(e.g. isolates AY394891, DQ273463, DQ273467 andEU563495).
The Tasmanian helotialean EcM species that have beenconfirmed by mycorrhiza anatomy and consistent molecu-lar identification (cf. Tedersoo et al., 2008b, 2009; L. Ted-ersoo, unpublished) were clustered in four distinctlineages (Fig. 3). These Tasmanian EcM lineages wereclearly distinguished from EAA, endophytes, ErM isolatesand the two EcM lineages distributed in the NorthernHemisphere (/meliniomyces and an unnamed lineage)based on both ITS (Appendix 2) and 28S (Fig. 3)sequence data.
Discussion
Diversity of EAA
Ascomycota comprises common and taxonomicallydiverse secondary colonists on EcM root tips. Our resultscorroborate previous reports of high local diversity of rootand foliar endophytes in various ecosystems (Vandenk-oornhuyse et al., 2002; Sieber and Grünig, 2006; Arnold,2007; Higgins et al., 2007). We detected no statisticaldifference in the diversity of EAA among the host plantsand fungi, microbiotope, plot and site, indicating thatnone of these substantially affect the naturally high EAAdiversity.
Among the factors investigated, host plant had thestrongest effect on the frequency of individual EAAspecies. This agrees with studies on foliar endophytes(Arnold and Lutzoni, 2007), but contrasts with previousresearch on root endophytes that detected no significantdifferentiation according to the host plant (Narisawa et al.,2002; Sieber and Grünig, 2006; Chambers et al., 2008;but see Grünig et al., 2008). However, methodologicaldifferences such as choice of a barcoding threshold andinclusion of the culturing step may account for the discrep-ancies. We speculate that host generalists, e.g. mostmembers of the Trichoderma, Phialocephala–Acephalaand Rhizoscyphus–Meliniomyces complexes, may befavoured by culturing due to their relatively rapid mycelialgrowth and non-specialized ecology. Similarly, lower DNAbarcoding thresholds may result in lumping of closelyrelated taxa that are often ecologically differentiated(Summerell and Leslie, 2004; Sharon et al., 2006; Grüniget al., 2008). Host preference among EAA (this study),
foliar endophytes (Arnold, 2007), EcM symbionts (Molinaet al., 1992; Tedersoo et al., 2008b) and arbuscularmycorrhizal symbionts (Vandenkoornhuyse et al., 2003)suggest that plant diversity, through the niche comple-mentarity effect, may promote the diversity of both com-mensal and mutualistic fungi above and below ground.The relatively stronger effect of plant host compared withplot, site and microbiotope effects on EAA species andcommunities suggests that interspecific differences inphytochemistry play a more important role in structuringthe distribution of EAA compared with qualitative differ-ences in the soil matrix and geographical distance.
Based on INSD search and phylogenetic analysis, mostof the EAA represented root endophytes, with a minorityhaving strongest affinities to plant parasites, mycopara-sites, insect parasites and soil saprobes. While the lattermay be rhizoplane fungi, root endophytes and plant para-sites are the expected root colonists (Vandenkoornhuyseet al., 2002; Neubert et al., 2006). Their ubiquitous asso-ciation with ectomycorrhizas and the negligible effect offungal host indicates that the fungal mantle is not neces-sarily an effective barrier to the colonization of endophyticand potentially parasitic fungi. Moreover, endophytic fungioften proliferate in the fungal mantle, developing hyphaewith hyaline or melanized cell walls (L. Tedersoo, pers.obs.).
Our results suggest that EcM root tips provide a habitatfor mycoparasitic fungi that normally infect fruit-bodiesabove ground. This study confirms previous reports onidentification of the Lecanicillium fungicola complex fromEcM root tips and associated soil (Summerbell, 1989) andsuggests that some mycoparasites (such as L. flavidum)may be relatively frequent in ectomycorrhizas. Theobserved preference of L. flavidum for cortinarius EcM hasnot been detected in case of the more general fungalfruit-body parasitism by L. flavidum or the closely relatedL. fungicola (Zare and Gams, 2008; K. Põldmaa, pers.obs.). The rarity of other mycoparasites among the EAAmay be ascribed to their host specificity (Põldmaa, 2000) orscarcity of their below ground associations. However,given the strongly seasonal production of suitable fruit-bodies, it is not surprising to find mycoparasites on EcMthat are active throughout the year. The ecology of EAA onEcM root tips, particularly the potential to spread along withthe EcM hyphae and rhizomorphs that give rise to fruit-body primordia, warrants further investigation. Similarly,Arnold (2008) noted the presence of certain entomopatho-gens among foliar endophytes and suggested that their lifecycle may include both insect and plant hosts.
In the perspective of EcM communities, the presence ofEAA hamper molecular identification and interpretation ofresults. Even small supplement of EAA DNA may becoamplified, resulting in double signal in sequence chro-matograms or, worse, the preferential amplification of
EAA (Rosling et al., 2003). To be able to separate thetargeted EcM fungi from these ‘contaminant’ secondarycolonizers, both morphological and molecular aspects ofEcM fungal communities need to be addressed. This cau-tions against pooling large amounts of roots prior to DNAextraction that disables post hoc morphological confirma-tion of the molecularly identified fungi (Bergemannand Garbelotto, 2006). We predict that ongoing high-throughput sequencing studies addressing mutualisticEcM fungi and soil eukaryotes will create thousands ofsequences from putative root endophytes, whose identityand actual ecology can easily be misinterpreted (Nilssonet al., 2009).
Phylogenetic affinities of EAA
Most of the EAA in the two Tasmanian sites belong to theHelotiales. This agrees well with previous sporadic reportsof EAA in EcM fungal communities (Bergemann and Gar-belotto, 2006; Smith et al., 2007; Morris et al., 2008a,b;2009; Urban et al., 2008; Wright et al., 2009) as well asthe more focused studies on root endophytes (Vandenk-oornhuyse et al., 2002; Vrålstad et al., 2002) and ErMfungi (Allen et al., 2003; Bergero et al., 2003) in theNorthern Hemisphere. Similarly, the Cladophialophora–Exophiala–Capronia group of Chaetothyriales is acommon facultative root endophytic taxon, althoughmembers of this group have been found from a variety ofsubstrates (Narisawa et al., 2002; 2007). The Hypocre-ales and Sordariales, represented here by parasitic orsaprobic members, are detected as infrequent root colo-nizers in studies on root endophytic fungi. These twoorders, as well as Dothideomycetes, dominate amongfoliar endophytes in angiosperms (Arnold, 2007; Higginset al., 2007).
We paid particular attention to phylotype distributionwithin Helotiales, because this was the most importantorder in terms of EAA frequency and diversity, andalso comprised EcM-forming fungi. The Tasmanian EAAspecies formed several mono- or multispecific lineagesthroughout the Helotiales, particularly clustering withthe Hyphodiscus, Oidiodendron and Cryptosporiopsis–Neofabraea complexes from the Northern Hemisphere.Except for the Cryptosporiopsis–Neofabraea clade, thesetaxa are not included in the 18S + 28S rDNA phylogeniesof Wang and colleagues (2006) and therefore, their phy-logenetic position within the Helotiales remains obscure.
Both the phylogenetic analysis and INSD searchessuggest substantial taxonomic overlap among Tasmanianand boreal EAA, root endophytes and ErM fungi. Previousstudies have suggested that several ErM fungi and rootendophytes may be conspecific based on identical culturemorphology (McNabb, 1961), synthesis trials (Bergeroet al., 2000) and molecular identification (Bergero et al.,
2000; Piercey et al., 2002). Moreover, many of these fungiare common saprobes in soil and peat (Piercey et al.,2002). Enzymatic tests have revealed relatively highlevels of cellulolytic activities in ErM and endophytic fungicompared with EcM symbionts (Mandyam and Jump-ponen, 2005). These elevated enzymatic activities maycontribute to the improved nutrition of ericoid plants inhighly organic, nutrient-poor soils (Read et al., 2004).Among hyperdiverse soil saprobes, members of the Cha-etothyriales and Helotiales in particular display frequentendophytic colonization. We suggest that Ericalesevolved capacities to host these endophytes in individualroot cells and stimulated the formation of coils forimproved nutrient exchange, thus giving rise to the ericoidmycorrhiza. ErM fungi and root endophytes largelyoverlap in many groups within Helotiales (Bergero et al.,2000; Chambers et al., 2008), Chaetothyriales (Usuki andNarisawa, 2005) and Sebacinales (Selosse et al., 2007).
Species of confirmed helotialean EcM fungi from Aus-tralia and Europe formed four and two distinct lineagesrespectively. There were no close relationships betweenEcM Helotiales from Australia and the Northern Hemi-sphere, suggesting that in both regions, EcM lifestylemay have evolved multiple times independently in thistaxon. Except for the /meliniomyces and /acephalamacrosclerotiorum lineages (Hambleton and Sigler, 2005;Münzenberger et al., 2009), EcM fungal taxa have noclear closely related ErM or root endophytic sister groups,suggesting different origin of EcM and other rootbiotrophic lifestyles. Therefore, taxonomic breadth of thepostulated common guild between EcM and ErM myco-bionts (Vrålstad et al., 2002; Bougoure et al., 2007) aswell as their evolutionary and ecological differencesrequire further clarification.
In conclusion, ascomycetes associated with ectomyc-orrhizas are highly diverse and comprise root endophytes,saprotrophs, myco-, phyto- and entomopathogens. Thedistribution of these microfungi is influenced by plant hostrather than EcM fungi, substrate type or geographicalvariables. Within Helotiales, EAA and the putative ecto-mycorrhizal symbionts are distantly related and probablyevolved multiple times independently in the Northernand Southern Hemisphere. In studies of mycorrhizalcommunities, the ubiquity and diversity of secondaryroot-associated fungi should be considered to avoidoverestimating the diversity of the true mycorrhizalsymbionts.
Experimental procedures
Sample preparation
Root sampling was performed in two Tasmanian wet sclero-phyll forest sites, Mt. Field (42°41′-S, 146°42′-E) and Warra(43°04′-S; 146°40′-E) as described in detail in Tedersoo and
colleagues (2008b; 2009). Briefly, EcM root tips of matureN. cunninghamii (Hook.) Oerst., E. regnans F. Muell. andP. apetala Labill. were sampled in forest floor soil from 45 soilcores (15 cm¥ 15 cm to 5 cm depth) in three 1 ha plots at Mt.Field. At Warra, root tips of only N. cunninghamii weresampled from 42 cores in decayed wood and 22 cores inforest floor soil. EcM root tips were sorted into morphotypesand anatomotypes using a stereomicroscope. Particular carewas taken to characterize and record the anatomy of EcMformed by Ascomycota (L. Tedersoo, unpublished). SingleEcM root tips, one to four from each anatomotype per soilcore, were carefully cleaned from the adhering soil and debrisand subjected to DNA extraction, PCR amplification andsequencing as described in Tedersoo and colleagues(2008b). The EcM symbionts were identified based on bar-coding of the rDNA Internal Transcribed Spacer (ITS) region.Roots from different host trees were initially distinguishedbased on morphological characters such as colour, branchingand thickness, and confirmed based on the length polymor-phism of the plastid trnL region (Tedersoo et al., 2008b).
From the DNA extracts of single root tips that were suc-cessfully ascribed to EcM plant and fungal species (Tedersooet al., 2008b; 2009), we targeted the associated Ascomycotausing the combination of a fungal-specific primer ITSO-FT(5′-acttggtcatttagaggaagt-3′) and the Ascomycota-specificLA-W (5′-cttttcatctttcgatcactc-3′) (Fig. 1). The Ascomycotawere specifically addressed, because a vast majority of soilfungi and endophytes belong to this phylum (Vandenkoorn-huyse et al., 2002; O’Brien et al., 2005; Arnold, 2007). We areaware that certain Basidiomycota such as Ceratobasidiumand Cryptococcus, and Zygomycota may also contribute tothe root endophytic fungal community especially in theculture-based studies, but form a minor component(Summerbell, 1989; Hoff et al., 2004; Neubert et al., 2006).Because most EcM root tips were associated with a single orno species of EAA (based on preliminary PCR surveys), onlydirect sequencing of the PCR products was performed,neglecting the cloning step. When several fungi were presenton root tips as revealed from double DNA bands on 1%agarose gels, the DNA fragments were cut from the geland re-amplified using the internal primers ITS5 (5′-ggaagtaaaagtcgtaacaagg-3′) or ITS1 (5′-tccgtaggtgaacctgcgg-3′)and ITS4 (5′-tcctccgcttattgatatgc-3′). Because the resultsrevealed dominance of Helotiales and Sordariomycetes, wefurther designed taxon-specific reverse primers, ITS4-Sord(5′-cccgttccagggaatct-3′), LR6-Sord (5′-gtttgagaatggatgaaggc-3′) and LR6-LS (5′-aaaatggcccactagtgttg-3′) in the 28SrDNA to specifically target these taxa. To address phyloge-netic relationships among the species of Helotiales, weamplified the 28S rDNA using a fungal specific primerLR0R (5′-acccgctgaacttaagc-3′) in combination with eitherof the newly developed Pezizomycotina-specific primerLR3-Asc (5′-cacytactcaaatccwagcg-3′) or Leotio- andSordariomycetes-specific LR6-LS (Fig. 1; Appendix 1). TheITS region of EAA was sequenced using the primers ITS4,ITS1, ITS5 and/or LF340 (5′-tacttgtkcgctatcgg-3′); 28S rDNAwas sequenced using the primers ctb6 (5′-gcatatcaataagcggagg-3′), TW13 (5′-ggtccgtgtttcaagacg-3′) and/or LR5(5′-tcctgagggaaacttcg-3′). Sequences were further trimmed,assembled and edited using Sequencher 4.7 software(Genecodes Corp., Ann Arbor, MI, USA). Based on the clus-
tering of ITS sequences belonging to the Helotiales in thepresent study and in previous research (Hambleton andSigler, 2005; Grünig et al., 2009), 99.0% was selected as auniversal barcoding threshold to distinguish between putativespecies. All unique ITS and partial 28S rDNA sequences aredeposited both in INSD (Accession Numbers FN298677–FN298803) and UNITE (UDB004100–UDB004230) publicdatabases.
Thirty 28S rDNA sequences (typically 600–900 bp span-ning divergent domains D1–D2 or D1–D3) of helotialean EAAfrom Tasmania were automatically aligned with publishedendophyte, ErM, EcM and preidentified fruit-body sequences(retrieved from INSD and UNITE public databases) usingMAFFT 5.861 (Katoh et al., 2005). Obvious alignment errorswere checked and edited manually. The final data set com-prised 167 taxa and 1335 characters. Using the onlineversion of RAxML 7.0.4 (Stamatakis et al., 2008), a maximumlikelihood phylogram with 100 fast bootstrap replicates wasconstructed following the GTR + G + P base substitutionmodel. Posterior probabilities were estimated with MrBayes3.1.2 (Ronquist and Huelsenbeck, 2003) using the samemodel (parameters: lset nst = 6, rates = invgamma). Two par-allel MCMC analyses were performed, both initiated withrandom starting trees and run for 10 000 000 generations.Every 100th generation was sampled. The first 10 000 treeswere discarded as burn-in. Posterior probabilities werecalculated from the remaining 90 000 trees sampled from9 000 000 generations.
Statistical analyses
G-tests were applied to detect statistically significant differ-ences in the occurrence of EAA among host fungal lineages,host plant species, mantle types, sites, plots and micro-biotopes. The effects of host fungus, mantle type and site onthe frequency of the eight most common EAA were furtherstudied based on the pooled sites and occurrence of EAAindividuals in each class, using Fisher’s exact tests. Morespecifically, the effects of plant host and plot were addressedat the Mt. Field site and microbiotope at the Warra site,following the same procedure. To study the differences inaccumulating species richness among each of the factors, wecalculated individual-based rarefaction curves with 95% con-fidence intervals using a computer program EstimateS 8.0(Colwell, 2006). Accumulating species richness of EAA wascompared with that of EcM fungal species identified from thesame EcM root tips.
Using a computer program DISTLM forward 1.3 (McArdleand Anderson, 2001), we studied the effect of fungal hostlineage, microsite and mantle anatomy on EAA communitycomposition of N. cunninghamii EcM at the Warra site. Inanother analysis, we addressed the effects of fungal hostlineage, plant host species, plot and mantle anatomy on EAAcommunity structure at the Mt. Field site. In both multivariateanalyses, occurrence of an EAA formed a sampling unit (i.e.individual). Singletons were removed from the analyses.Thus, the Mt. Field and Warra data sets comprised 22 and 17species as well as 112 and 65 individuals respectively. EcMfungal lineages with at least three occurrences were includedas dummy variables. The occurrence of individuals was
standardized by sums of variables (EAA species). Chi-squaredistance was used as a distance metric with 999 permuta-tions. Significance level a = 0.05 was used in all statisticalanalyses.
Acknowledgements
We thank G. Kantvilas, D. Ratkowsky, N. Ruut and D.Puskaric for support in Tasmania; H.-O. Baral for voucherspecimens; D. Ratkowsky for helpful suggestions on anearlier draft of the manuscript. This study was funded byEstonian Science Foundation Grants no. 6606, 6939, 7434and JD92, Doctoral School of Environmental Sciences, Krist-jan Jaak scholarship and FIBIR/rloomtipp. G.G. is supportedby Forestry Tasmania, Holsworth Wildlife Research Endow-ment Fund, CRC for Forestry and Bushfire.
References
Allen, T.R., Millar, T., Berch, S.M., and Berbee, M.L. (2003)Culturing and direct DNA extraction find different fungi fromthe same ericoid mycorrhizal roots. New Phytol 160: 255–272.
Arnold, A.E. (2007) Understanding the diversity of foliarendophytic fungi: progress, challenges, and frontiers. FungBiol Rev 21: 51–66.
Arnold, A.E. (2008) Endophytic fungi: hidden components oftropical community ecology. In Tropical Forest CommunityEcology. Carson, W.P., and Schnitzer, S.A (eds). Oxford,UK: Wiley-Blackwell, pp. 254–271.
Arnold, A.E., and Lutzoni, F. (2007) Diversity and host rangeof foliar fungal endophytes: are tropical leaves biodiversityhotspots? Ecology 88: 541–549.
Benson, D.R., and Clawson, M.L. (2000) Evolution of theactinorhizal plant symbioses. In Prokaryotic Nitrogen Fixa-tion: A Model System for Analysis of Biological Process.Triplett, E.W. (ed.). Wymondham, UK: Horizon ScientificPress, pp. 207–224.
Bergemann, S.E., and Garbelotto, M. (2006) High diversity offungi recovered from the roots of mature tanoak (Lithocar-pus densiflorus) in northern California. Can J Bot 84:1380–1394.
Bergero, R., Perotto, S., Girlanda, M., Vidano, G., and Luppi,A.M. (2000) Ericoid mycorrhizal fungi are common rootassociates of a Mediterranean ectomycorrhizal plant(Quercus ilex). Mol Ecol 9: 1639–1649.
Bergero, R., Girlanda, M., Bello, F., Luppi, A.M., and Perotto,S. (2003) Soil persistence and biodiversity of ericoid myc-orrhizal fungi in the absence of the host plant in a Mediter-ranean ecosystem. Mycorrhiza 13: 69–75.
Bougoure, D.S., Parkin, P.I., Cairney, J.W.G., Alexander, I.J.,and Anderson, I.C. (2007) Diversity of fungi in hair roots ofEricaceae varies along a vegetation gradient. Mol Ecol 16:4624–4636.
Burke, D.J., Dunham, S.M., and Kretzer, A.M. (2008) Molecu-lar analysis of bacterial communities associated with theroots of Douglas fir (Pseudotsuga menziesii) colonized bydifferent ectomycorrhizal fungi. FEMS Microbiol Ecol 65:299–309.
Chambers, S.M., Curlevski, N.J.A., and Cairney, J.W.G.
(2008) Ericoid mycorrhizal fungi are common root inhabit-ants of non-Ericaceae plants in a south-eastern Australiansclerophyll forest. FEMS Microbiol Ecol 65: 263–270.
Colwell, R.K. (2006) EstimateS: statistical estimation ofspecies richness and shared species from samples,version 8. [WWW document]. URL http://purl.oclc.org/estimates.
Fontana, A., and Luppi, A.M. (1966) Funghi saprofiti isolati daectomicorrize. Allionia 12: 39–46.
Girlanda, M., and Luppi-Mosca, A.M. (1995) Microfungi asso-ciated with ectomycorrhizae of Pinus halepensis Mill.Allionia 33: 93–98.
Grünig, C.R., Queloz, V., Sieber, T.N., and Holdenrieder, O.(2008) Dark septate endophytes (DSE) of the Phialo-cephala fortinii s.1. Acephala applanata species complex intree roots: classification, population biology, and ecology.Botany 86: 1355–1369.
Grünig, C.R., Queloz, V., Duó, A., and Sieber, T.N. (2009)Phylogeny of Phaeomollisia piceae Gen. et Sp. nov. a dark,septate, conifer-needle endophyte and its relationships toPhialocephala and Acephala. Mycol Res 113: 207–221.
Hambleton, S., and Sigler, L. (2005) Meliniomyces, a newanamorph genus for root-associated fungi with phyloge-netic affinities to Rhizoscyphus ericae (= Hymenoscyphusericae), Leotiomycetes. Stud Mycol 53: 1–27.
Higgins, K.L., Arnold, A.E., Miadlikowska, J., Sarvate, S.D.,and Lutzoni, F. (2007) Phylogenetic relationships, hostaffinity, and geographic structure of boreal and arctic endo-phytes from three major plant lineages. Mol Phyl Evol 42:543–555.
Hoff, J.A., Kloppenstein, N.B., McDonald, G.I., Tonn, J.R.,Kim, M.-S., Zambino, P.J., et al. (2004) Fungal endophytesin woody roots of Douglas-fir (Pseudotsuga menziesii)and ponderosa pine (Pinus ponderosa). For Path 34: 255–271.
Katoh, K., Kuma, K., Toh, H., and Miyata, T. (2005) MAFFTversion 5: improvement in accuracy of multiple sequencealignment. Nucleic Acid Res 33: 511–518.
McArdle, B.H., and Anderson, M.J. (2001) Fitting multivariatemodels to community data: a comment on distance-basedredundancy analysis. Ecology 82: 290–297.
McNabb, R.F.R. (1961) Mycorrhiza in the New Zealand Eri-cales. Aust J Bot 9: 57–61.
Mandyam, K., and Jumpponen, A. (2005) Seeking the elusivefunction of root-colonizing dark septate endophytic fungi.Stud Mycol 53: 173–189.
Melin, E. (1923) Experimentelle Untersuchungen über dieBirken-und Espenmykorrhizen und Ihre Pilzsymbionten.Svensk Bot Tidskr 17: 479–519.
Molina, R., Massicotte, H., and Trappe, J.M. (1992) Specific-ity phenomena in mycorrhizal symbiosis: community-ecological consequences and practical implications. InMycorrhizal Functioning. An Integrative Plant-FungalProcess. Allen, M. (ed.). New York, NY, USA: Chapman &Hall, pp. 357–423.
Morris, M.H., Perez-Perez, M.A., Smith, M.E., and Bledsoe,C.S. (2008a) Multiple species of ectomycorrhizal fungi arefrequently detected on idividual oak root tips in a tropicalcloud forest. Mycorrhiza 18: 375–383.
communites on the roots of co-occuring oaks (Quercusspp.) in a California woodland. New Phytol 178: 167–176.
Morris, M.H., Perez-Perez, M.A., Smith, M.E., and Bledsoe,C.S. (2009) Influence of host species on ectomycorrhizalcommunities associated with two co-occurring oaks(Quercus spp.) in a tropical cloud forest. FEMS MicrobiolEcol 69: 274–287.
Münzenberger, B., Bubner, B., Wöllecke, J., Sieber, T.N.,Bauer, R., Fladung, M., and Hüttl, R.F. (2009) The ecto-mycorrhizal morphotype Pinirhiza sclerotia is formed byAcephala macrosclerotiorum sp. nov., a close relative ofPhialocephala fortinii. Mycorrhiza (in press): doi: 10.1007/s00572-009-0239-0.
Narisawa, K., Kawamata, H., Currah, R.S., and Hashiba, T.(2002) Suppression of Verticillium wilt in eggplant by somefungal root endophytes. Eur J Plant Pathol 108: 103–109.
Narisawa, K., Hambleton, S., and Currah, R.S. (2007) Het-eroconium chaetospira, a dark septate root endophyteallied to the Herpotrichiellaceae (Chaetothyriales) obtainedfrom some forest soil samples in Canada using bait plants.Mycoscience 48: 274–281.
Neubert, K., Mendgen, K., Brinkmann, H., and Wirsel, S.G.R.(2006) Only a few fungal species dominate highly diversemycofloras associated with the common reed. ApplEnviron Microbiol 72: 1118–1128.
Nilsson, R.H., Ryberg, M., Abarenkov, K., Sjökvist, E., andKristiansson, E. (2009) The ITS region as a target forcharacterization of fungal communities using emergingsequencing technologies. FEMS Microbiol Lett 296:97–101.
O’Brien, H.E., Parrent, J.L., Jackson, J.A., Moncalvo, J.-M.,and Vilgalys, R. (2005) Fungal community analysis bylarge-scale sequencing of environmental samples. ApplEnviron Microbiol 71: 5544–5550.
Piercey, M.M., Thormann, M.N., and Currah, R.S. (2002)Saprobic characteristics of three fungal taxa from ericaleanroots and their association with the roots of Rhododendrongroenlandicum and Picea mariana in culture. Mycorrhiza12: 175–180.
Põldmaa, K. (2000) Generic delimitation of the fungicolousHypocreaceae. Stud Mycol 45: 83–94.
Porter, T.M., Schadt, C.W., Rizvi, L., Martin, A.P., Schmidt,S.K., Scott-Denton, L., et al. (2008) Widespread occur-rence and phylogenetic placement of a soil clone groupadds a prominent new branch to the fungal tree of life. MolPhyl Evol 46: 635–644.
Read, D., Leake, J.R., and Perez-Moreno, J. (2004) Mycor-rhizal fungi as drivers of ecosystem processes in heathlandand boreal forest biomes. Can J Bot 82: 1243–1263.
Ronquist, F., and Huelsenbeck, J.P. (2003) MRBAYES 3:Bayesian phylogenetic inference under mixed models. Bio-informatics 19: 1572–1574.
Rosling, A., Landeweert, R., Lindahl, B.D., Larsson, K.-H.,Kuyper, T.W., Taylor, A.F.S., and Finlay, R.D. (2003) Ver-tical distribution of ectomycorrhizal fungal taxa in a podzolsoil profile. New Phytol 159: 775–783.
Selosse, M.-A., Setaro, S., Glatard, F., Richard, F., Urcelay,C., and Weiss, M. (2007) Sebacinales are common myc-
orrhizal associates of Ericaceae. New Phytol 174: 864–878.
Sharon, M., Kuninaga, S., Hyakumachi, M., and Sneh, B.(2006) The advancing identifiction and classification ofRhizoctonia spp. using molecular and biotechnologicalmethods compared with the classical anastomosis group-ing. Mycoscience 47: 299–316.
Sieber, T.N., and Grünig, C.R. (2006) Biodiversity of fungalroot-endophyte communities and populations, in particularof the dark septate endophyte Phialocephala fortinii s.l. InMicrobial Root Endophytes. Schulz, B.J.E., Boyle, C.J.C.,and Sieber, T.N. (eds). Berlin, Germany: Springer, pp.107–132.
Smith, M.E., Douhan, G.W., and Rizzo, D.M. (2007) Ectomy-corrhizal community structure in a xeric Quercus woodlandbased on rDNA sequence analysis of sporocarps andpooled roots. New Phytol 174: 847–863.
Sprent, J.L., and James, U.K. (2007) Legume evolution:where do nodules and mycorrhizas fit in? Plant Physiol144: 575–581.
Stamatakis, A., Hoover, P., and Rougemont, J. (2008) A rapidbootstrap algorithm for the RAxML web-servers. Syst Biol75: 758–771.
Summerbell, R.C. (1989) Microfungi associated with the myc-orrhizal mantle and adjacent microhabitats within the rhizo-sphere of black spruce. Can J Bot 67: 1085–1095.
Summerell, B.A., and Leslie, J.F. (2004) Genetic diversityand population structure of plant-pathogenic species in thegenus Fusarium. In Plant Microbiology. Gillings, M., andHolmes, A. (eds). Oxford, UK: Garland Science/BIOS Sci-entific Publisher, pp. 207–223.
Tedersoo, L., Kõljalg, U., Hallenberg, N., and Larsson, K.-H.(2003) Fine scale distribution of ectomycorrhizal fungi androots across substrate layers including coarse woodydebris in a mixed forest. New Phytol 159: 153–165.
Tedersoo, L., Suvi, T., Larsson, E., and Kõljalg, U. (2006)Diversity and community structure of ectomycorrhizal fungiin a wooded meadow. Mycol Res 110: 734–748.
Tedersoo, L., Pellet, P., Kõljalg, U., and Selosse, M.-A.(2007) Parallel evolutionary paths to mycoheterotrophy inunderstorey Ericaceae and Orchidaceae: ecological evi-dence for mixotrophy in Pyroleae. Oecologia 151: 206–217.
Tedersoo, L., Suvi, T., Jairus, T., and Kõljalg, U. (2008a)Forest microsite effects on community composition of ecto-mycorrhizal fungi on seedlings of Picea abies and Betulapendula. Environ Microbiol 10: 1189–1201.
Tedersoo, L., Jairus, T., Horton, B.M., Abarenkov, K., Suvi,T., Saar, I., and Kõljalg, U. (2008b) Strong host preferenceof ectomycorrhizal fungi in a Tasmanian wet sclerophyllforest as revealed by DNA barcoding and taxon-specificprimers. New Phytol 180: 479–490.
Tedersoo, L., Gates, G., Dunk, C., Lebel, T., May, T.W.,Kõljalg, U., and Jairus, T. (2009) Establishment of ectomy-corrhizal fungal community on isolated Nothofaguscunninghamii seedlings regenerating on dead wood inAustralian wet temperate forests: does fruit-body typematter? Mycorrhiza (in press) doi: 10.1007/s00572-009-0244-3.
Urban, A., Puschenreiter, M., Strauss, J., and Gorfer, M.(2008) Diversity and structure of ectomycorrhizal and
co-associated fungal communities in a serpentine soil.Mycorrhiza 18: 339–354.
Usuki, F., and Narisawa, K. (2005) Formation of structuresresembling ericoid mycorrhizas by the root endo-phytic fungus Heteroconium chaetospira within roots ofRhododendron obtusum var. kaempferi. Mychorriza 15:61–64.
Vrålstad, T., Schumacher, T., and Taylor, A.F.S. (2002) Myc-
orrhizal synthesis between fungal strains of the Hymenos-cyphus ericae aggregate and potential ectomycorrhizaland ericoid hosts. New Phytol 153: 143–152.
Wang, Z., Johnston, P.R., Takamatsu, S., Spatafora, J.W.,and Hibbett, D.S. (2006) Toward a phylogenetic classifica-tion of the Leotiomycetes based on rDNA data. Mycologia98: 1066–1076.
Wright, S.H.A., Berch, S.M., and Berbee, M.L. (2009) Theeffect of fertilization on the below-ground diversity andcommunity composition of ectomycorrhizal fungi associ-ated with western hemlock (Tsuga heterophylla). Mycor-rhiza 19: 267–276.
Zare, R., and Gams, W. (2008) A revision of the Verticilliumfungicola species complex and its affinity with the genusLecanicillium. Mycol Res 112: 811–824.
Appendix 1
Characterization of the primers (in bold) designed for the identification of Ascomycota in this study. Multiple sequence alignmentswith target and non-target taxa are indicated.
ITS4-Sord (calculated TM = 56°C); specific to SordariomycetesPrimer 5′-CCCGTTCCAGGGAACTC-3′Sordariomycetes (45) *****************Ascomycota other (101) **Y******A**R***TBasidiomycota (95) *********A*AR***TPlants (6) ***C********G***TLA-W (calculated TM = 58°C); specific to AscomycotaPrimer 5′-CTTTTCATCTTTCGATCACTC-3′Ascomycota (incl. SCGI*) (98) *********************