Environmental Microbiology (2009) 11(12), 3166–3178 doi:10.1111/j.1462-2920.2009.02020.x Ascomycetes associated with ectomycorrhizas: molecular diversity and ecology with particular reference to the Helotialesemi_2020 3166..3178 Leho Tedersoo,1,2* Kadri Pärtel,1 Teele Jairus,1,2 to other helotialean root-associated fungi, indicating Genevieve Gates,3 Kadri Põldmaa1,2 and independent evolution. The ubiquity and diversity of Heidi Tamm1 the secondary root-associated fungi should be con- 1Department of Botany, Institute of Ecology and Earth sidered in studies of mycorrhizal communities to Sciences, University of Tartu, 40 Lai Street, 51005 avoid overestimating the richness of true symbionts. Tartu, Estonia. 2Natural History Museum of Tartu University, 46 Introduction Vanemuise Street, 51005 Tartu, Estonia. 3Schools of Agricultural Science and Plant Science, Endophytic and mycorrhizosphere microbes, especially University of Tasmania, Hobart, Tasmania 7001, Bacteria, Archaea and microfungi, synthesize plant Australia. growth regulators and vitamins facilitating the develop- ment and functioning of the mycorrhizal system in soil (Schulz et al., 2006). These root-associated microbes Summary such as mycorrhiza helper bacteria and the nitrogen- Mycorrhizosphere microbes enhance functioning of fixing actinobacteria and rhizobia differ substantially in the plant–soil interface, but little is known of their their function and ecology, including host preference pat- ecology. This study aims to characterize the asco- terns (Benson and Clawson, 2000; Sprent and James, mycete communities associated with ectomycorrhi- 2007; Burke et al., 2008). Of microfungi, foliar endo- zas in two Tasmanian wet sclerophyll forests. We phytes may considerably vary according to the special- hypothesize that both the phyto- and mycobiont, ization to different host species and even organs mantle type, soil microbiotope and geographical dis- (Neubert et al., 2006; Arnold, 2007; Higgins et al., 2007). tance affect the diversity and occurrence of the asso- On the contrary, facultative root-associating fungi such ciated ascomycetes. Using the culture-independent as endophytes (e.g. the Phialocephala–Acephala and rDNA sequence analysis, we demonstrate a high Meliniomyces–Rhizoscyphus complexes) form mostly diversity of these fungi on different hosts and non-specific associations with many plant hosts (Vrålstad habitats. Plant host has the strongest effect on the et al., 2002; Chambers et al., 2008), although host pref- occurrence of the dominant species and community erence may occur on the cryptic species level (Grünig composition of ectomycorrhiza-associated fungi. et al., 2008). Root endophytes, soil saprobes, myco-, phyto- and Despite numerous studies on isolation and morphol- entomopathogens contribute to the ectomycorrhiza- ogical identification of ascomycetous microfungi from associated ascomycete community. Taxonomically ectomycorrhizal (EcM) root tips, their specificity for host these Ascomycota mostly belong to the orders Helo- plants, fungi and substrate types remains unknown tiales, Hypocreales, Chaetothyriales and Sordariales. (Melin, 1923; Fontana and Luppi, 1966; Summerbell, Members of Helotiales from both Tasmania and the 1989; Girlanda and Luppi-Mosca, 1995). Molecular tools Northern Hemisphere are phylogenetically closely have only recently been used to characterize and distin- related to root endophytes and ericoid mycorrhizal guish the secondarily associated microfungi from EcM fungi, suggesting their strong ecological and evolu- fungi in situ. These microfungi were identified from EcM tionary links. Ectomycorrhizal mycobionts from Aus- root tips by either cutting additional bands from the gel tralia and the Northern Hemisphere are taxonomically (Rosling et al., 2003; Tedersoo et al., 2006), using specific unrelated to each other and phylogenetically distant primers (Urban et al., 2008) or cloning (Morris et al., 2008a,b; 2009; Wright et al., 2009). Cloning from DNA extracts comprising pooled individual EcM root tips Received 9 April, 2009; accepted 22 June, 2009. *For correspon- dence. E-mail [email protected]; Tel. (+372) 7376222; Fax reveals many ascomycete taxa of uncertain ecological (+372) 7376222. role (Bergemann and Garbelotto, 2006; Smith et al., © 2009 Society for Applied Microbiology and Blackwell Publishing Ltd Ectomycorrhiza-associated ascomycetes 3167 2007). Many of these are endophytic or rhizoplane colo- Results nists that are accidentally reported as forming mycorrhi- Identification and distribution of EAA zas both in research publications and International Sequence Database (INSD) entries (Grünig et al., 2008). Application of the newly designed taxon-specific primers Such uncertain reports are especially common in ericoid (Fig. 1; Appendix 1) allowed us to specifically amplify mycorrhizas (ErM) and EcM for which universal fungal- Ascomycota from EcM root tips. Based on the rDNA ITS specific primers are routinely used for the identification of sequence analysis, 251 individuals of EAA were identified mycobionts. from 226 out of 675 (33.5%) analysed root tips (148 Helotiales (Ascomycota) comprises the largest number individuals from the Mt. Field site and 103 from the Warra of undescribed root-associated fungi in addition to site). 88.9% of the EcM root tips yielded a single amplicon approximately 2000 described species with contrasting of EAA. Based on the 99% ITS barcoding threshold, EAA lifestyles (Wang et al., 2006). Various subgroups of were assigned to 105 species, including 69 (65.7%) Helotiales such as the Phialocephala–Acephala and singletons and 15 (14.3%) doubletons (Appendix 2). Rhizoscyphus–Meliniomyces complexes and Lachnum At both sites, species of EcM fungi and EAA were spp. are identified from EcM and arbutoid mycorrhiza in accumulating at similar rates with increasing sampling forest trees and subshrubs of the Northern Hemisphere effort (Fig. 2). The species accumulation curves had (Vrålstad et al., 2002; Rosling et al., 2003; Tedersoo et al., strongly overlapping confidence intervals (not shown) 2003; 2007; 2008a; Bergemann and Garbelotto, 2006) suggesting no substantial difference in EAA diversity and often erroneously reported as truly mycorrhizal. among mantle types, sites, plots, microsites or plant and Indeed, the ecologically heterogeneous Rhizoscyphus– fungal hosts. Similarly, there were no statistically signifi- Meliniomyces and Phialocephala–Acephala complexes cant differences in the relative frequency of colonization of both include distinct EcM-forming species nested within EAA among these habitats. numerous pathogenic, ErM and root endophytic taxa Trends in the distribution of eight most frequent EAA (Vrålstad et al., 2002; Hambleton and Sigler, 2005; species were statistically analysed at both two sites Münzenberger et al., 2009). (Table 1). Five of these species differed significantly In previous EcM fungal community studies in Tasmania, according to the site. Only Lecanicillium flavidum (syn. we encountered frequent secondary colonization of EcM Verticillium fungicola var. flavidum) displayed a statisti- root tips by Ascomycota besides the predominately cally significant preference for EcM fungal lineage (Fish- basidiomycetous EcM fungi (Tedersoo et al., 2008b; er’s exact test: d.f. = 4; P = 0.002). This species occurred 2009). The present study was undertaken to identify and more frequently on root tips colonized by members of distinguish these EcM-associated Ascomycota (EAA) the/cortinarius lineage, compared with the other four most from the true EcM-forming mycobionts using a culturing- common EcM lineages. Due to elevated abundance independent approach. Utilizing the DNA extracts from on the/cortinarius EcM, L. flavidum was more common single EcM root tips with preidentified plant and EcM on plectenchymatous mantles than expected (d.f. = 1; fungal hosts and developing several ascomycete-specific P = 0.004). In contrast, Helotiales sp016 colonized exclu- primers, we hypothesized that these EAA have prefer- sively mycorrhizas with pseudoparenchymatous mantles ence for either host tree, host fungus lineage, EcM mantle (d.f. = 1; P = 0.001). type, soil microbiotope, plot and site. Because of the Among the seven most common EAA species at Mt. dominance of Helotiales in this and previous studies Field, host trees and plots affected the distribution of six involving root-associated fungi, we addressed the phylo- and two species respectively (Table 1). Putative root genetic relations of helotialean EcM, ericoid mycorrhizal, endophytes, root parasites and mycoparasites included endophytic and EAA isolates using the rDNA 28S species with significant host plant preference. For sequence data. example, 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. © 2009 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 11, 3166–3178 3168 L. Tedersoo et al. 80 preferred Pomaderris apetala compared with the other two hosts (Nothofagus cunninghamii is the third host). At the Warra site, only L. flavidum was significantly more 70 common in the forest floor soil compared with decayed wood (d.f. = 1; P = 0.027). At Mt. Field and Warra, respectively, the multivariate 60 model explained 19.3% and 17.6% of the total variation in the distribution of EAA. At both sites, the fungal lineage and mantle anatomy explained < 2% of the total variation 50 that remained non-significant. At Mt. Field, host