• Ascomycetes;
  • dark septate endophytes (DSE);
  • group I intron;
  • molecular ecology;
  • Mollisia;
  • Phialocephala fortinii;
  • Tapesia


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • • 
    The diversity and phylogenetic affinities of symbiotic root-associated ascomycetes of the Helotiales are reported here based on ITS1–5.8S-ITS2 (internal transcribed spacer, ITS) nrDNA sequences.
  • • 
    Mycobionts were obtained from roots of ericoid plants and grasses and from Piceirhiza bicolorata ectomycorrhizas (pbECM) on conifers and hardwoods, predominantly in burnt and metal-polluted habitats. The mycobionts were sequenced through the ITS and compared with sequences of known helotialean taxa.
  • • 
    We recognized 132 fungal ITS-sequences with affinity to the Helotiales, of which 75% (54 different ITS-genotypes) grouped within the Hymenoscyphus ericae aggregate including Phialophora finlandia. This aggregate showed stronger affinity to members of the Hyaloscyphaceae and Dermateaceae than to Hymenoscyphus fructigenus (genus-type species; Helotiaceae). Most of the pbECM mycobionts grouped with P. finlandia, although some grouped with H. ericae. Two genotypes co-occurred in ericoid and ectomycorrhizal roots.
  • • 
    The H. ericae aggregate may be referable to a generic unit, and includes a diverse group of closely related, more or less darkly pigmented, root-associated ascomycetes where the borders between intra- and interspecific ITS-sequence variation, as well as different mycorrhizal and nonmycorrhizal root-symbioses, remains unclear.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The ascomycete order Helotiales includes an ecologically diverse group of plant pathogens, wood-, debris- and soil saprobes, plant endophytes, and mutualistic ericoid mycorrhizal (ERM)- and ectomycorrhizal (ECM) fungi. Teleomorphic species of the Helotiales are characterised by inoperculate asci and discoid, turbinate or clavate ascocarps ranging in size from the hardly visible members of the Hyaloscyphaceae to more prominent members of the Geoglossaceae and Sclerotiniaceae. The Helotiales was erected by Nannfeldt (1932), and was replaced with Leotiales (Carpenter, 1988). However, the Helotiales sensu str. and Leotiales sensu str. are currently recognised as two separate orders, the latter only comprising the Leotiaceae sensu str. (Korf & Lizon, 2000, 2001). Recent molecular studies have confirmed that several mitotic or sterile fungi of the ERM, ECM and DSE (dark septate endophytes) types have phylogenetic affinity to the Helotiales (LoBuglio et al., 1996; Monreal et al., 1999; Sharples et al., 2000; Vrålstad et al., 2000, 2002).

DSE are conidial or sterile ascomycetous fungi with darkly pigmented and septate hyphae that colonise living plant roots without causing apparent negative effect (Jumpponen & Trappe, 1998). They are widely distributed and frequently isolated from all sorts of plant roots (Wang & Wilcox, 1985; Wilcox & Wang, 1987a, 1987b; Jumpponen & Trappe, 1998). Their ecological functions are not well understood, but some DSE have been reported to occupy different positions along the mutualism-commensalism-parasitism axis (Jumpponen & Trappe, 1998). According to Jumpponen (2001), DSE may be considered as mycorrhizal if the variation in host response to mycorrhizal fungi is considered to represent a continuum ranging from parasitism to mutualism. However, the term DSE is liberally applied whenever melanised, septate hyphae are observed to colonise plant roots inter- or intracellularly (Jumpponen, 2001). It is therefore important to realise that the various DSE taxa possess different phylogenetic histories and ecological roles. Phialophora finlandia Wang and Wilcox is a mitotic fungus commonly referred to as a typical DSE (Jumpponen, 2001) that forms ectendomycorrhizal relationships with Pinus and Larix (for a review, see Yu et al., 2001), ECM with spruce and birch (Wilcox & Wang, 1987a,b), and has also been observed to form ericoid mycorrhiza with Gaultheria shallon (Monreal et al., 1999). Parsimony analysis of ITS-sequence data has grouped P. finlandia together with the well-known ERM fungus Hymenoscyphus ericae (Read) Korf & Kernan in a 100% bootstrap supported clade (Vrålstad et al., 2002). This clade, referred to as the H. ericae aggregate, also includes several sterile, darkly pigmented fungi (typically fitting the DSE concept) responsible for the Piceirhiza bicolorata ECM morphotype (pbECM), as well as a white group of root-associated, nonmycorrhizal fungi (Vrålstad et al., 2000, 2002). The different strategies observed for fungi within the H. ericae aggregate suggest that the taxonomical and ecological borders between ERM, ECM and nonmycorrhizal fungi may not be as sharp as previously thought.

This study focuses on molecular diversity and phylogenetic affinities of root-symbionts of the Helotiales isolated from boreal forest plants of burnt and metal polluted habitats. Fungi obtained from pbECM roots of conifers and hardwoods, and from roots of ericaceous plants and grasses, were sequenced through the ITS-region (ITS1-5.8S-ITS2) of the nuclear ribosomal (nr) DNA repeat and compared with known Helotiales taxa in phylogenetic analyses. The diversity and distribution of ITS-genotypes of the H. ericae aggregate within and between root systems of the different boreal forest plants were surveyed.

Materials and Methods

  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

Study area and sampling

In the present study, the two localities of main focus were a postfire site of Norway (N), Oslo, Maridalen (NOM), and two heavily metal contaminated Cu-mine spoils of N., Hedmark, Folldal (NHF). A few collections of plant individuals from a clear cut spruce forest (NAE) and a mixed oak-pine forest (NTK) were also included. The postfire site constituted 375 da of mixed coniferous forest that had burned heavily in 1992 (Vrålstad et al., 1998). Naturally regenerated seedlings of Betula pubescens (birch), Populus tremula (aspen), Picea abies (spruce) and Pinus sylvestris (pine), ericoid plants of Calluna vulgaris (heather), Vaccinium myrtillus (blueberry) and V. vitis-idaea (cowberry), and the grass Deschampsia flexuosa (hairy grass) were collected from a 20-m × 20-m plot (= NOM4[1]) within a heavily burnt area of previous 100-yr-old-spruce forest. Additional collections of birch were added from another heavily burnt spruce forest area of the postfire site (= NOM1).

Two Cu-mine spoils were investigated (NHF1 and NHF2). The mine spoils consisted of finely grained waste material (probably products from flotation) from earlier copper-mining activities that ended in the 1950s. The waste products included among others copper pyrite, iron sulphide, iron ore, sulphur pyrite, sulphurous ore, zinc sulphide, and calcareous rocks (Helge Tysland (retired director of the Folldal mines, Hjerkinn), pers. comm.). A vegetation cover was absent, but some scattered seedlings of pine, birch and Salix spp. (willow), ericoid plants of cowberry, heather and Empetrum nigrum (crowberry), and the grass Festuca ovina (fescue) were observed to colonise the spoils. Some scattered individuals of pine, crowberry and fescue were collected within the first spoil (NHF1). Within the second spoil (NHF2), plants were collected in four major areas (= NHF2[1–4]). From the largest plot (3-m × 3-m) of NHF2 (= NHF2[4]), nine specimens of four plant species (pine, birch, willow and cowberry) were collected. Due to the scattered and uneven occurrence of each plant host in the postfire and metal-polluted pioneer plant communities, sampling of an equal number of plant specimens from each host plant was not justified. Plant root systems were collected selectively, and treated as previously described in Vrålstad et al. (2000). Table 1 lists the number of plant specimens of each host species that were collected from each locality in the period from 1998 to 1999.

Table 1.  Plant samples, axenic culture root isolates and pbECM root extracts included in the study
Year1HostARON isolate/extract no3Locality4Primers5 
sample2AvintHintEMBL/GenBank accession
  • 1

    Years of sampling.

  • 2

    Plant sample: one numbered plant root-system of a given species from which axenic cultures and root extracts were derived. Abbreviation (e.g. Cv) followed by a number (e.g. 1) reads Calluna vulgaris, sample 1.

  • 3

    DNA-samples were derived from two sources (1) axenic culture root isolates from various hosts, and (2) single pbECM roots. These were given a number in the ARON (Ascomycete Research group of Oslo, Norway) collection. ARON-numbers followed by letter (i.e. S = somatic culture) refers to an axenic culture isolate derived from a surface sterilised plant root. R (= root extract) followed by a number refers to a DNA-extract of one single pbECM root.

  • indicate isolates/extracts included in previous studies by Vrålstad et al. (2000, 2002). Axenic culture isolates are deposited in the culture collection of ARON.

  • 4

    Locality: NOM = Norway, Oslo, Maridalen (postfire site). NOM4(1) is a 20-m × 20-m plot within the postfire site, and NOM1 is another area of the postfire site. NHF = Norway, Hedmark, Folldal (two Cu-mine spoils = NHF1 and NHF2). NHF2(1–4) indicate four plots within NHF2. NAE = Norway, Akershus, Eidsvoll (boreal forest clear-cut). NTK = Norway, Telemark, Kragerø (boreal oak-pine forest). SU = Sweden, Uppsala. UK = United Kingdom. Superscript: the person responsible for the sample and/or the isolation of axenic culture isolates: Anders Dahlberg (AD) and Andy Taylor (AT) at SLU (Swedish University of Agricultural Sciences, Uppsala, Sweden); Trude Vrålstad (TV) and Elin Myhre (EM) at ARON, UiO (University of Oslo, Norway).

  • 5

    PCR products were amplified either with the Avint/ITS4 or the Hint/ITS4 primer pairs, or both. + indicate positive amplification; – indicate negative amplification; d indicate double PCR bands; nt = not tested;

  • *

    asterisks () are used to indicate the chosen sequencing primer (Avint or Hint) in cases where both primer pairs yielded positive amplification and single PCR-products.

 Calluna vulgaris (L.) Hull
1998Cv12886.S, 2887.S, 2888.SNHF1TV.– ––+ + +AJ430102, AJ430103, AJ308337
1999Cv32967.S, 2968.S, 2969.S, 2970.S, 2971.SNOM4(1)TV.– + –+ ++ +* + + –AJ430106, AJ430107, AJ430108, AJ430109, AJ430110
 Empetrum nigrum L.
 Vaccinium myrtillus L.AJ430111    
1999Vm22961.S, 2962.S, 2963.SNOM4(1)TV.+ nt −– + +AJ430112, AJ430113, AJ430114
 Vaccinium vitis-idaea L.
1999Vvi32995.S, 2996.S, 2997.S, 2998.SNHF2(4)TV.– –––+ + ++AJ430115, AJ430116, AJ430117, AJ430118
 Vvi43010.S, 3012.S, 3014.SNHF2(4)TV.– + −+ +* +AJ430119, AJ430120, AJ430121
 Deschampsia flexuosa (L.) Trin.
1998Df12927.S, 2928.S, 2929.SNAETV.– − ++ + −AJ430207, AJ430216, AJ430208
 Festuca ovina L.AJ430225
 Betula pubescens Ehrh.
1998Bp1R0747, R0748NOM4(1)TV.nt nt+ +AJ430123, AJ430124
 Bp52976.S, 2977.S, 2978.SNHF2(4)EM.+ + +– ––AJ430400, AJ430401, AJ430402
 Bp73024.S, 3025.S, 3028.SNHF2(3)EM.+ + +nt nt ntAJ430126, AJ430127, AJ430128
 Bp83026.S, R0996NHF2(4)EM.+ dnt +AJ430227, AJ430129
 Bp11R1079, 3041.SNOM1EM.+ −nt +AJ430131, AJ430132
 Bp123042.S, R1007NHF2(4)EM.+ −nt +AJ430133, AJ430134
 Bp14R1124, R1125NOM1EM.– −+ +AJ430136, AJ430137
 Bp15R1133, R1135, R1136, R1137, R1138, R1139, R1140, 3060.SNOM4(1)EM.– ––– + ++ ++ + ++ +* +*+* +*AJ430138, AJ430139, AJ430140, AJ430141, AJ430142, AJ430143, AJ430144, AJ430145
 Nothofagus procera (Poepp. et Endl.) Oerst.
 Populus tremula L.
1998Pt12903.S, 2906.SNTKTV.+AJ308338, AJ308341
 Quercus robur L.
1998Qr12893.S, 2894.SNTKTV.– ++ −AJ292203, AJ308339
 Salix herbaceae L.
 Picea abies (L.) Karst.
1999Pa62953.S, 2954.S, 2955.S, 2956.S, 2958.SNOM4(1)TV.– ––– –+ + ++ +AJ430153, AJ430154, AJ430155, AJ430156, AJ430157
 Pinus sylvestris L. 
1998Ps12878.S, 2879.S, 2880.SNHF1TV.+ + −nt –+AJ430209, AJ292201, AJ430210
1999Ps3R0903, 2983.SNHF2(4)EM.nt ++ +*AJ430160, AJ430161
 Ps42985.S, 2986.S, 2989.S, R0854NHF2(4)EM.+* + +++ – +*AJ430406, AJ430408, AJ430407, AJ430162
 Ps53004.S, 3005.S, 3006.SNHF2(3)EM.+ + +nt nt ntAJ430163, AJ430164, AJ430165
 Ps63018.S, 3021.S, R1014NHF2(4)EM.+ + +nt nt +*AJ430166, AJ430167, AJ430410
 Ps73032.S, R1057, R1058, R1059NHF2(1)EM.+ d + dnt + ++AJ430168, AJ430169, AJ430170, AJ430171
 Ps8R1108, R1109NOM4(1)EM.– –+ +AJ430172, AJ430173
 Ps93034.S, 3035.S, 3066.S, R1162, R1165NHF2(1)EM.+ + ++ +nt nt nt +*AJ430174, AJ430175, AJ430176, AJ430211, AJ430411
 Ps103037.S, 3038.S, 3039.S, 3061.S, 3062.S, 3063.S, R1158NHF2(1)EM.+ + ++ + ++nt nt nt – –––AJ430413, AJ292197, AJ430212, AJ430409, AJ430412, AJ292198, AJ430414
 Ps113043.S, R1084, R1085NOM4(1)EM.– ––+ + +AJ430177, AJ430178, AJ430179
 Ps123047.S, 3048.S, 3049.S, 3050.S, 3051.S, R1113, R1114, R1115NOM4(1)EM.– −+ + d – –+ + + nt nt ++ +AJ430180, AJ430181, AJ430182, AJ430213, AJ430214, AJ430183, AJ430184, AJ430185
 Ps13 3055.S, 3057.S, R1044, R1045, R1047, R1050NHF2(4)EM.– –– +* + ++ + ++ − +*AJ430190, AJ430191, AJ430186, AJ430187, AJ430188, AJ430189

Preparation of roots and axenic culture isolation of associated fungi

Ectomycorrhizal plant root systems were rinsed carefully under tap water, cut into segments and rinsed again under tap water through a 0.5-mm sieve. The clean root segments were studied in water under a stereo magnifier. Vital ECM roots fitting the description of the Piceirhiza bicolorata morphotype (cf. Brand et al., 1992; Vrålstad et al., 2000; pbECM), were detached with a scalpel and rinsed in sterile water. For direct DNA extraction, a single pbECM root (up to 10 replicates for each root system) was transferred to a microcentrifuge tube containing 2 × CTAB-extraction buffer, and stored at −20°C. Axenic culture isolation of putative mycobionts of single pbECM root was performed as described in Vrålstad et al. (2000). Ericoid- and grass root systems were carefully rinsed under tap water. A selection of fine roots were detached and rinsed properly under tap water for a minimum of 10 min. Cleaned fine roots were cut into 2–3 mm segments that were rinsed three to four times in sterile water, surface sterilised and cultivated as described in Vrålstad et al. (2000). One axenic culture isolate from each root-fragment was cultivated. Table 1 lists the axenic culture isolates and single pbECM root extracts included in the present study, along with their hosts, origins and EMBL/GenBank/DDBJ accession numbers.

Teleomorphs of the Helotiales for taxonomic reference

Ascocarp specimens of some selected teleomorph taxa of the Helotiales known to occur in boreal coniferous forests of Norway were sampled during the spring of 2000. Cleaned, individual apothecia were transferred to tubes containing 2 × CTAB for direct DNA-amplification. Axenic culture isolates were obtained from ascospores (cf. Vrålstad et al., 1998). Table 2 lists ITS-sequences of teleomorphs included in the study. ITS-sequences retrieved from the EMBL/GenBank/DDBJ sequence databases are also included in Table 2.

Table 2.  Included reference material and sequences of ascomycetes of Helotiales (ingroup taxa) and Rhytismatales (outgroup taxa)
Taxon1Strain/source2FamilyDeterminer/reference paperHost or substrateOriginAccession no3
  • 1

    T = ex-type culture.

  • 2

    ARON = Ascomycete Research group of Oslo, Norway. ARON numbers followed by P = DNA derived from axenic poly spore culture, H = DNA derived from fresh or dried ascocarp specimen. CBS = Centraal Bureau voor Schimmelcultures, Baarn, The Netherlands. UAMH = University of Alberta Microfungus Collection and Herbarium. JHH = John H. Haines. NYS = New Your State museum. SAC = Sharon A. Cantrell. GAM = J. H. Miller Mycological Herbarium. UBC = University of British Columbia.

  • 3

    EMBL/GenBank/DDBJ accession numbers of the ITS1-5.8S-ITS2 sequences. Asterisk indicates reference sequences retrieved from EMBL/GenBank, the remaining sequences were generated by us.

Ingroup taxa of Helotiales:
Dermea acerina (Peck)  RehmCBS 161.38DermateaceaeAbeln et al. (2000)Acer rubrumCanadaAF141164*
Mollisia cinerea (Batsch: Fr.)  P. KarstARON3129.PT. SchumacherDecaying cone of Picea abiesNorwayAJ430222
M. minutella (Sacc.) RehmARON3139.HDecaying stalk of Epilobium angustifoliumAJ430223
Neofabraea malicorticis (Cordley) H. Jacks.CBS 141.22Abeln et al. (2000)Malus sylvestris (fruit)UnknownAF141161*
Pezicula heterochroma Tul. & C. Tul. (T)CBS 199.46Alnus crispaCanadaAF141167*
P. sporulosa Tul. & C. Tul. (T)CBS 262.31Cupressus lawsonianaUKAF141172*
Pyrenopeziza revincta (P. Karst.) GremmenARON3150.PT. SchumacherDecaying stalk of Epilobium angustifoliumNorwayAJ430224
Tapesia cinerella RehmARON3188.HDecaying twig/barkAJ430228
T. fusca (Pers. Fr.) FuckelARON3154.HAJ430229
Crocicreas coronatum (Bull. Fr.) S. E. Carp.ARON3093.HHelotiaceaeDecaying plant stemAJ430397
Hymenoscyphus ericae (Read) Korf & Kernan (T)UAMH 6735D.J. ReadCalluna vulgarisUKAJ319078
H. fructigenus (Bull. Fr.)  GrayARON3264.HT. SchumacherDecaying nut of Quercus roburNorwayAJ430396
H. monotropa Kernan &  FinocchioUAMH 6650M. KernanMonotropa unifloraUSAAF169309*
H. rhodoleucus (Fr. Fr.)  W. PhillipsARON2329.PT. SchumacherStalk of Equisetum sp.NorwayAJ430395
Pezizella amentii (Batsch: Fr.)  DennisARON3114.PCatkin of Salix sp.AJ430398
Cistella acuum (Alb. & Schwein.  Fr.) Raitv.JHH 3966HyaloscyphaceaeCantrell & Hanlin (1997)UnknownUSAU57492*
Hyaloscypha aureliella (Nyl.)  HuhtinenSAC NY1 GAMUnknownU57495*
Lachnellula calyciformis (Willd. Fr.) DharneJHH 4622, NYSUnknownU59145*
Lachnum bicolor (Bull. Fr.)  P. Karst.ARON 3113.PT. SchumacherDecaying hardwood twigNorwayAJ430394
L. cladestinum (Bull. Fr.)  P. Karst.JHH 4676 NYSCantrell & Hanlin (1997)UnknownUSAU58636*
L. controversumJHH 4611 NYSUnknownU58638*
L. cf. pygmaeum (Fr.)  Bres. sp1ARON 3248.PT. SchumacherMoribund fine roots of Picea abiesNorwayAJ430218
L. cf. pygmaeum (Fr.)  Bres. sp2ARON 3255.HMoribund ericoid rootsAJ430219
L. rhytismatis (W. Phillips)  Nannf.ARON3186.HDecaying leaf nerve (Vaccinium myrtillus)AJ430220
L. virgineum (Batsch: Fr.) P.  Karst.ARON3137.PDecaying stalk of Epilobium angustifoliumAJ430221
L. virgineumJHH 4312 NYS Cantrell & Hanlin (1997)UnknownUSAU59004*
Phacidium infestans P. Karst.U51980*PhacidiaceaeGernandt et al. (1997)Pinus cembraU92305*
Anamorphs and sterile strains
Dactylaria dimorphospora Veenb.-RijksCBS 256.70Liou & Tzean (1997)Agricultural soilThe NetherlandsU51980*
Dactylella lobata DuddingtonCBS 228.54Liou & Tzean, unpub.UnknownUKU51958*
Phialocephala fortinii Wang & WilcoxUAMH 9525Addy et al. (2000)Vaccinium vitis-idaeaCanadaAF214579*
UAMH 6677AF214580*
H24Luetkea pectinataAF214579*
Phialophora finlandia Wang & Wilcox (T)FAG-15C.J.K. Wang & HE WilcoxECM-roots of Pinus sylvestrisFinlandAF011327*
Phialophora sp.p3847McKemy et al. unpub.UnknownUnknownAF083200*
Scytalidium vaccinii Dalpé,  Litten & Sigler (T)UAMH 5828Y. Dalpé & L. SiglerVaccinium angustifoliumUSAAJ319077
‘Glacial ice fungus’GI685Ma et al. (2000)Isolated from glacial iceGreenlandAF177733*
Ericoid root-isolateSd9Bergero et al. (2000)Erica arboreaItalyAF269067*
GU44Sharples et al. (2000)Calluna vulgarisAustraliaAF252842*
UBCtra241Millar et al., unpub.Gaultheria shallonCanadaAF149068*
Outgroup taxa of Rhytismatales:
Meria laricis Vuill.RhytismataceaeGernandt et al. (1997)Larix occidentalisUSAU92298*
Rhabdocline pseudotsugae Parker & ReidPseudotsuga menziesiiU92290*
R. parkeri Shearwood-Pike,  Stone & CarrollU92294*

Molecular methods

DNA was extracted from single pbECM roots as well as from fresh axenic culture mycelia using the miniprep method described by Gardes & Bruns (1993) and slightly modified in Vrålstad et al. (2000). Previous works (Egger et al., 1995; Perotto et al., 2000; Vrålstad et al., 2000) and our own observations had shown that most isolates of the H. ericae aggregate contained a group I intron positioned between the ITS1 and ITS5 (White et al., 1990) primer sites of the SSU nrDNA (i.e. a 1506SSU group I intron; cf. Perotto et al., 2000). In some cases we found that the ITS1 primer produced no amplicon due to a large insert within the ITS1 primer site. Since DNA-extracts of pbECM roots and axenic culture isolates were frequently observed to hold nrDNA copies both with and without group I introns (i.e. two different-sized PCR-amplicons), we decided to avoid the primer ITS5. Two new primers were therefore designed: the universal Avint (5′-GTA-ACA-AGG-TTT-CCG-TAG-GTG-3′) and the intron-specific Hint (5′-ACA-GAC-TAA-GTG-ATT-GTG-GG-3′). The Avint primer, which starts with the nine 3′- nucleotides of ITS5 and ends with the 10 5′-nucleotides of ITS1, spans the group I intron-site 1506SSU, and consequently excludes nrDNA copies with group I intron(s). The Hint primer is the reverse primer of the group I intron specific primer IntSDom3 (Holst-Jensen et al., 1999) adjusted to match the intron sequence of H. ericae. The Hint primer is positioned in the 3′-end of the conserved catalytic core region S of group I introns (Cech & Herschlag, 1996) on the forward strand, and will only amplify nrDNA copies containing group I introns. Separate PCR-amplifications were performed using Avint and Hint in combination with ITS4 (White et al., 1990). PCR-amplifications were conducted in 40 µl volumes containing 19.5 µl 10 ×, 50 ×, or 100 × diluted template DNA and 20.5 µl reaction mix (final concentrations: 4 × 250 µmol l−1 dNTPs, 0.625 µmol l−1 of each primer, 2 mmol l−1 MgCl2 and 1 unit DyNAzymeTM. II DNA polymerase (Finnzymes Oy, Espoo, Finland)) on a Genius Operator (Techne (Cambridge) Ltd., Cambridge, UK) thermal cycler. Ice-cold reaction tubes were preheated for initial denaturation (95°C, 4 min), followed by 35 cycles (denaturation at 94°C, 15 s, annealing at 55°C, 15 s, and synthesis at 72°C, 40 s), and a final elongation step (72°C, 7 min) before storage (4°C). The complete ITS-region was sequenced manually using the ThermoSequenase radio labelled terminator cycle sequencing kit (Amersham Biosciences, Cleveland, OH, USA) with α-33P-ddNTPs. Both strands were sequenced to confirm the sequence, viz. Avint or Hint was used as sequencing primer for the forward strand and ITS4 for the reverse strand. A few amplicons obtained with the Hint-ITS4 and Avint-ITS4 primer pairs were larger than expected. In the former case (Hint-ITS4), this could be due to annealing of the Hint primer to another group I intron located upstream in the SSU, for example a 788SSU or a 943SSU intron (Holst-Jensen et al., 1999; Perotto et al., 2000). In the latter case (Avint-ITS4), it may be explained by the possible presence of a group I intron in a position not spanned by the Avint primer. In the cases of large amplicons, the primer ITS2 (White et al., 1990) was used as sequencing primer in addition to Hint-ITS4 or Avint-ITS4 in order to obtain complete sequences.

Sequence analyses

The obtained ITS-sequences were used as probes in Fasta searches (Pearson & Lipman, 1988, in the EMBL/GenBank/DDBJ databases in order to retrieve the most similar available ITS-sequence for inclusion in the phylogenetic analyses. Sequences were aligned in the computer program BioEdit (version; Hall, 1999), first automatically using the ClustalW option, and thereafter adjusted manually by visual inspection. A preliminary phylogenetic analysis recognised a great majority of root-derived sequences that fitted into the previously reported subclades of the H. ericae aggregate (Vrålstad et al., 2000). The sequences were split into two data sets, one for analysing the ITS-genotype variation of the H. ericae aggregate, and one for phylogenetic relationships among the H. ericae aggregate, unidentified root-symbionts and known taxa of the order Helotiales. The H. ericae aggregate data set included sequences of the H. ericae, S. vaccinii and P. finlandia ex-types, and all root-derived fungal genotypes within this group. The Helotiales data set included six representatives of the H. ericae aggregate, teleomorph and anamorph reference sequences (Table 2), and root-derived fungal ITS-sequence genotypes outside the H. ericae aggregate. Rhabdocline pseudotsugae, R. parkeri, and Meria laricis were used as outgroup taxa for the Helotiales data set, and Mollisia cinerea and Hyaloscypha aureliella as outgroup taxa for the H. ericae aggregate data set. The latter outgroup taxa were determined by the phylogenetic relationships inferred from the Helotiales analysis. The ITS-sequence of H. aureliella in the EMBL/GenBank/DDBJ sequence database lacked a 23 base pair motif succeeding the ‘AACTTTC’ motif at the 5′ end of the 5.8S gene. Since the vast majority of 5.8S genes of ascomycetes contain this 23 bp motif, we assumed the lack was due to an error, and in the alignment the lacking motif was substituted by question marks (?). Maximum parsimony analyses were conducted in a beta version of PAUP* 4.04b (Swofford, 1999) with legal character states being the A, C, G, T and gap (–). Polymorphisms were recoded using the standard IUPAC codes. Before phylogenetic analyses, PAUP was used to identify isolates with 100% identical ITS-genotypes. Only one representative of each ITS-genotype was included in the phylogenetic analysis in order to reduce computer runtime needed for the analysis. Due to high ITS-sequence variability, it was not possible to find a single unambiguous alignment for the Helotiales dataset. Two analyses were therefore performed for this dataset: One analysis including all (575) aligned characters treated as unordered and with equal weight (Helotiales-FULLset), and one analysis excluding a total of 233 characters from different ambiguously aligned areas of the alignment, leaving 342 characters (unordered and with equal weight) to be analysed (Helotiales-REDset). The alignments are available upon request. For all analyses, we used the heuristic search option in PAUP* with 1000 replicates and the random stepwise addition option, ‘TBR’ branch swapping, MULPARS, and the collapse zero length branches option switched on. Bootstrap analyses were performed with 1000 bootstrap replicates, using the heuristic search option as described above, and with ‘Maxtrees’ set to 1000. Percentage ITS1-5.8S-ITS2-sequence identity (% SI) was calculated group-wise for similar sequences within clades and subclades, using the ‘sequence identity matrix’-option in BioEdit, which includes gaps (–) as a fifth character state.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

A total of 132 ITS-sequences of root-associated fungi grouped within the Helotiales. Ninety-nine of these (75%) grouped within the Hymenoscyphus ericae aggregate and constituted a total of 54 different ITS-genotypes. Table 3 summarises the sequences that shared 100% sequence identity. Sixty additional sequences were excluded from further analyses for various reasons: 15 axenic culture isolates and two pbECM root extracts were identified as nonhelotialean ascomycetes; 18 pbECM root extracts were identified as basidiomycetes; and 25 pbECM root extracts yielded double, unreadable sequences. Separate amplifications with the Avint-ITS4 and Hint-ITS4 primer pairs in most cases solved the problems with double or negative PCR-amplicons. More than 70% of the DNA samples amplified with the Hint-ITS4 primer pair (Table 1). Nine culture isolates and 15 pbECM root extracts were observed to contain nrDNA copies both with and without group I introns (Table 1). Some sequences of the pbECM root extracts, culture isolates and apothecia were polymorphic at single nucleotide sites, mainly as simultaneous occurrence of C and T = Y (transition polymorphism).

Table 3.  ITS1-5.8S-ITS2 sequences sharing 100% nucleotide sequence identity
Clade1ARON numbers2, shared genotype3/(plant sample4, locality5)
  • 1

    The Helotiales and H. ericae aggregate trees refer to six (I–VI) and four (1–4) major clades, respectively. Clade I of the Helotiales-tree and Clade 4 of the H. ericae aggregate-tree were subdivided into four (A–D) and five (A–E) groups, respectively.

  • 2

    ARON-numbers followed by letter (i.e. S = somatic culture) refers to an axenic culture isolate derived from a surface sterilised plant root. R (= root extract) followed by a number refers to a DNA-extract of one single ectomycorrhizal root.

  • 3

    Bold numbers correspond to sequences denoted with asterisks (*) in the Helotiales and the H. ericae aggregate trees. These share 100% sequence identity with the succeeding sequences. Numbers separated by indicate that isolates or root-extracts originate from the same plant specimen, while = indicate that isolates or root-extracts originate from different plant specimens.

  • 4

    Plant sample: one numbered root system of a given plant species from which axenic cultures and root extracts were derived. Bp=Betula pubescens, Cv=Calluna vulgaris; Df=Deschampsia flexuosa; Pa=Picea abies; Ps=Pinus sylvestris; Qr=Quercus robur; Sh=Salix herbaceae; Vm=Vaccinium myrtillus; Vvi=Vaccinium vitis-idaea. Sequences derived from fungal root-isolates of Gaultheria shallon (Gs) and Luetkea pectinata (Lp) were retrieved from the EMBL/GenBank.

  • 5

    Locality: NOM = Norway, Oslo, Maridalen (postfire site). NHF = Norway, Hedmark, Folldal (Cu-mine spoils). NAE = Norway, Akershus, Eidsvoll (boreal forest clear-cut). NTK = Norway, Telemark, Kragerø (boreal oak-pine forest).

Helotiales/Fig. 1
 IA2976.S≡ 2977.S ≡ 2978.S (Bp5, NHF2[4])
 IB2927.S (Df1, NAE) = 2878.S (Ps1, NHF1)
 IB2880.S (Ps1, NHF) = AF214580 (= P. fortinii; Vvi, Canada)
 IB3051.S (Ps12, NOM4[1]) = AF214582 (= P. fortinii; Lp, Canada)
 II2986.S (Ps4, NHF2[4]) = 3061.S (Ps10, NHF2[1])
 III3038.SR1158 (Ps10, NHF2[1])
 IIIR1165 (Ps9, NHF2[1]) ≡ 3062.S ≡ 3063.S (Ps10, NHF2[1])
H. ericae aggregate/Fig. 2
 12894.S (Qr1, NTK) = 2971.S (Cv3, NOM4[1])
 2888.S≡ 2886.S (Cv1, NHF1) = 2884.S (En1, NHF1) = 2996.S ≡ 2997.S ≡ 2998.S (Vvi3, NHF2[4])
 R1058≡ R1057≡ R1059 (Ps7, NHF2[4])
 2970.S≡ 2967.S ≡ 2969.S (Cv3, NOM4[1])
 33025.S≡ 3024.S ≡ 3028.S (Bp7, NHF2[3])
 4C2810.S (Pa1, NOM4[1]) = 2938.S (Pa4, NOM4[1]) = R0748 (Bp1, NOM4[1]) = R1115≡ R1114 (Ps12, NOM4[1])
 4C2893.S (Qr1, NTK) = 2954.S ≡ 2955.S (Pa6, NOM4[1])
 4E2948.S (Pa5, SU) = 3055.S ≡ 3057.S ≡ R1044≡ R1050 (Ps13, NHF2[4])
 4B2963.S (Vm2, NOM4[1]) = AF149070 (Gs, Canada)
 4C3003.S (Pa7, NOM4[1]) = 2953.S (Pa6, NOM4[1])
 4E3005.S≡ 3004.S ≡ 3006.S (Ps5, NHF2[3])
 4E3010.S (Vvi4, NHF2[4]) = 3015.S (Sh1, NHF2[4])
 4B3012.S≡ 3014.S (Vvi4, NHF2[4])
 4C3021.S≡ 3018.S (Ps6, NHF2[4])
 4C3034.S≡ 3035.S (Ps9, NHF2[1])
 4E3048.S≡ 3049.S ≡ R1113 (Ps12, NOM4[1])
 4CR0793 (Bp4, NOM4[1]) = 2956.S ≡ 2958.S (Pa6, NOM4[1])
 4CR1124 (Bp14, NOM1) = R1079 (Bp11, NOM1) = R1084≡ R1085 (Ps11, NOM4[1]) = R1108≡ R1109 (Ps8, NOM4[1]) = R1133≡ R1135≡ R1139 (Bp15, NOM4[1])
 4CR1140≡ R1136R1137 (Bp15, NOM4[1])
 4ER1045≡ R1047 (Ps13, NHF2[4])
 4ER1176 (Bp17, NHF2[4]) = 2983.S (Ps3, NHF2[4]) = R0854 (Ps4, NHF2[4])

The Helotiales

Parsimony analyses of the Helotiales FULLset yielded 120 most parsimonious trees (MPTs; length 1813, consistency index [CI] 0.4335; CI excluding autapomorphies [CIx] 0.4184; retention index [RI] 0.7695; rescaled consistency index [RC] 0.3336), while analysis of the Helotiales REDset yielded 24 MPTs (length 663, CI 0.4480; CIx 0.4209; RI 0.7951; RC 0.3562). The bootstrap consensus trees obtained from these analyses collapsed several branches and yielded two nearly identical polytomies of the Helotiales: the FULLset retained nine major clades while the REDset collapsed a few more branches including branches internal to the nine clades, resulting in 14 major clades. The bootstrap consensus tree obtained with the REDset analysis is shown in Fig. 1(a). Six of the major clades (denoted I–VI) contained unidentified root-associated fungi. In four of these clades root-associated fungi grouped with identified helotialean taxa (Fig. 1a; clade I, II, V, VI). Fig. 1(b) shows the phylogenetic relationships within these clades inferred from the Helotiales FULLset analysis. Clade I included the vast majority of root-associated fungi of this study, and comprised four main groups (referred to as A–D; cf. Fig. 1a,b). Three of these groups (B–D) contained identified helotialean teleomorphs, while group A only consisted of pbECM root isolates. Group B includes the genera Pyrenopeziza, Tapesia and Mollisia of the Dermateaceae, as well as the mitotic DSE Phialocephala fortinii. Several root-associated fungi of heather, blueberry, hairy grass and pine shared 97–100% SI with P. fortinii, and grouped in a terminal clade of the B group (Fig. 1b). The FULLset analysis recognised Tapesia cinerella and T. fusca as sister taxa to P. fortinii (Fig. 1b), but this relationship yielded no bootstrap support in the REDset analysis (Fig. 1a). The B group of clade I also included a terminal clade with root isolates derived from fescue and birch sharing 96.4–96.8% SI with Pyrenopeziza revincta. Another terminal clade included a root isolate of hairy grass sharing 98.5% SI with an ancient ascomycete isolated from the glacial ice of Greenland (Ma et al., 2000). Group C included Hyaloscypha aureliella of the Hyaloscyphaceae and two fungal root-associates derived from pine and salal. Both analyses recognized the H. aureliella group as a sister group to the Hymenoscyphus ericae aggregate (group D; Fig. 1). The H. ericae aggregate includes 75% of the sequenced root associated fungi of this study, and is treated separately in Fig. 2 (see next section The Hymenoscyphus ericae aggregate). Clade II comprised pbECM root isolates and Dactylaria dimorphospora, clade III pbECM root isolates and extracts, and clade IV consisted of a single confirmed ericoid mycorrhizal isolate (Bergero et al., 2000). In clade V, two pbECM root isolates grouped with Cistella acuum and Phialophora sp. Clade VI comprised several species of Lachnum spp., and included a terminal clade with fungal root isolates of hairy grass and cowberry sharing 97.6–98.5% SI with L. pygmaeum (Fig. 1b).


Figure 1. Phylogenetic recognition of fungal root-symbionts derived from boreal forest plants in burnt and metal polluted habitats. Phylogenetic affinities of unknown fungal root-symbionts to known taxa of the Helotiales were identified with parsimony analyses of ITS1-5.8S-ITS2 nrDNA sequences using the heuristic search algorithm in PAUP* 4.0b4 (random addition) and bootstraping (1000 bootstrap replicates). Rhabdocline pseudotsugae, R. parkeri and Meria laricis (Rhytismatales) were used as outgroup taxa. Bootstrap support > 50% is indicated. (a) The bootstrap consensus tree shown here is based on parsimony analysis of the Helotiales REDset (excluding 233 characters from highly variable and ambiguously aligned areas). Bold, Roman numerals (I–VI) denote the six major clades containing unknown root-associated fungi. Clade I was subdivided into four major groups (A–D). (b) The phylogenetic relationships within the clades (I–VI) and groups (A–D) based on parsimony analysis of the Helotiales FULLset (including all characters). The clades shown represent the clades in one of the 120 most parsimonious trees obtained in this analysis. Branch lengths are drawn proportional to the number of character state changes (csc; bar = 10 csc) to give a visual impression of the degree of similarity between ITS-genotypes. The following apply for both for (a) and (b): Asterisks (*) indicate ITS-genotypes recorded more than once. ARON (Ascomycete Research group of Oslo, Norway) numbers refer to ITS-sequences generated during this study (plain = axenic culture isolates; italic=ECM root extracts). Accession numbers refer to sequences retrieved from the EMBL/GenBank/DDBJ sequence databases. T: sequences derived from ex-type cultures. Teleomorph and anamorph names are in black. Plant species names indicate the hosts of the root-associated fungi (brown = grasses; purple = ericoid plants; green = hardwoods; blue = conifers). Previously confirmed ectomycorrhizal (ECM) and ericoid mycorrhizal (ERM) isolates have blue and purple highlight colours, respectively. The H. ericae aggregate is highlighted in a yellow box.

Download figure to PowerPoint


Figure 2. One out of the 9017 most parsimonious trees yielded with the heuristic search algorithm in PAUP* and the H. ericae aggregate data set, using Mollisia cinerea and Hyaloscypha aureliella as outgroup taxa. Bar = five character state changes (csc). Numbers above lines =% bootstrap support. Bold lines denote the four major clades (clade 1–4) retained in the strict consensus tree. The five major groups of clade 4 are denoted A–E. Asterisks (*) indicate ITS-genotypes recorded more than once. Red arrows indicate ITS-genotypes recorded in both ericoid and ECM-host plants. Diamond (◆), Phialophora finlandia (isolate FAG15) has previously been reported to form both ECM (Wilcox & Wang, 1987a) and ERM (Monreal et al., 1999). ARON (Ascomycete Research group of Oslo, Norway) numbers refer to ITS-sequences generated during this study (plain = axenic culture isolates; italic=ECM root extracts). Accession numbers refer to sequences retrieved from the EMBL/GenBank/DDBJ sequence databases. Teleomorph and anamorph names are in black. Plant species names denote the hosts of the root-associated fungi (brown = grasses; purple = ericoid plants; green = hardwoods; blue = conifers). Previously confirmed ectomycorrhizal (ECM) and ericoid mycorrhizal (ERM) isolates have blue and purple highlight colours, respectively. For each sequence genotype, the locality is indicated by squares (black = NHF – Cu-mine spoils; grey = NOM – postfire site; white = other localities).

Download figure to PowerPoint

The Hymenoscyphus ericae aggregate

Parsimony analysis of the H. ericae aggregate dataset yielded 9107 MPTs that varied in the relative branching order of similar isolates within the clades. Both the strict and bootstrap consensus trees retained four major clades (1–4) that were robust to alternative alignments. Fig. 2 shows one of the MPTs with the four major clades indicated. Clade 1 comprised isolates derived from various ericoid and pbECM roots sharing 97–100% SI. Clade 2 comprised two pbECM root isolates sharing 95.6% SI, one of which confirmed to form ECM (Vrålstad et al., 2002). Clade 3 comprised the ex-types of H. ericae and S. vaccinii, along with several ericoid root isolates, and pbECM root isolates and extracts (Fig. 2 and Table 3). These sequences shared 96.8–100% SI. Finally, the large and heterogeneous Clade 4 was subdivided into five groups (A–E). The vast majority of pbECM root isolates and extracts grouped in clade 4, predominantly in the groups C (97.2–100% SI) and E (97–100% SI), the latter including P. finlandia (Fig. 2). A few ericoid- and grass root isolates also grouped in clade 4. The ex-types of H. ericae and S. vaccinii shared 97.8% SI, and those of H. ericae and P. finlandia shared 92% SI.

Mycobionts of the Piceirhiza bicolorata ECM morphotype

Overall, c. 61% of the putative pbECM mycobionts grouped in clade 4, group C and E, of the H. ericae aggregate. These groups also included P. finlandia and five other confirmed ECM-forming isolates (Figure 2; Vrålstad et al., 2002). From the postfire site (NOM), a total of 38 ITS-sequences were obtained from pbECM root isolates and extracts. Two of these were recognised as Phialocephala fortinii (Fig. 1; clade I, group B), and the remaining 36 sequences grouped within the H. ericae aggregate; one within the H. ericae group (clade 3), 28 (= 10 ITS-genotypes) within group C of clade 4, and seven (= six ITS-genotypes) within the P. finlandia group (group E of clade 4). By contrast, ITS-sequences obtained from pbECM root isolates and extracts from mine-spoil plants were more heterogeneous. Thirty-six ITS-sequences grouped within the H. ericae aggregate, one within clade 1, eight (= four ITS-genotypes) within clade 3 (the H. ericae clade), and the remaining sequences grouped within clade 4: one within group A, four (= two ITS-genotypes) within group C, and 17 (= nine ITS-genotypes) within group E. Additionally, 22 pbECM sequences (= 15 ITS-genotypes) derived from mine-spoil plants grouped within the Helotiales, but outside the H. ericae aggregate (Fig. 1), and 30 excluded pbECM sequences (first section of the results) originated also from mine-spoil plants.

Regional and local distribution of ITS-genotypes of the H. ericae aggregate

Identical ITS-genotypes (100% SI) were recorded from geographically separated locations in four cases, and all included different host plants (Fig. 2; Table 3): ARON2894.S (oak, NTK) and ARON2971.S (heather, NOM4[1]); ARON2963.S (blueberry, NOM4[1]) and AF149070 (salal, Canada); ARON2948.S (spruce, Sweden) and ARON3055.S (+ three other pbECM sequences; pine, NHF2[4]); and ARON2893.S (oak, NTK) and ARON2954.S + ARON2955.S (spruce, NOM4[1]).

Identical ITS-genotypes were frequently recognised within individual plant root systems and in some cases also on roots from different plant individuals of the same locality (Table 3, Fig. 3). Fig. 3(a) summarises the local distribution of ITS-genotypes of the H. ericae aggregate within a 20-m × 20-m plot of the postfire site (NOM4[1]). Six neighbouring plant species, that is spruce, birch, aspen, heather, blueberry and hairy grass within a 2-m diameter circle of the NOM4(1) (Fig. 3a; ‘magnifying glass’) hosted altogether 11 ITS-genotypes of the H. ericae aggregate, of which nine were observed only within individual root systems. One ITS-genotype was observed in two neighbouring specimens of spruce, and one in two neighbouring specimens of spruce and birch. Another nine ECM-seedlings of birch, pine and spruce scattered within the NOM4(1) plot (but outside the circle), harboured 10 ITS-genotypes. Eight of these were observed only from one root system each, while two ITS-genotypes were observed from different ECM root systems. A maximum of four ITS-genotypes were recorded within the same root system (Fig. 3a). A fungal isolate that shared 98.5% SI with an ancient ‘glacial ice fungus’ (Fig. 1), originated from roots of the hairy grass (Df4) within the NOM4(1) plot.


Figure 3. Local distribution and variation of ITS-genotypes of the Hymenoscyphus ericae aggregate in plant root systems from two geographically distant plots. (a) A 20-m × 20-m plot (NOM4[1]) of the postfire site in a previous 100-yr-old spruce forest. An ITS-phylogeny based on the 21 observed genotypes, is shown to the left. Hymenoscyphus ericae and Phialophora finlandia are included as reference taxa. Yellow squares (1–4) denote genotypes of the H. ericae clade. Blue shapes (5–20, i.e. circle, oval, lying and standing rhombus) illustrate subgroups of similar ITS-genotypes of the P. finlandia clade. The approximate location of each plant individual within the plot is indicated along with the recorded ITS-genotypes. The plant species include: Betula pubescens (Bp), Populus tremula (Pt), Picea abies (Pa), Pinus sylvestris (Ps), Calluna vulgaris (Cv), Vaccinium myrtillus (Vm), and Deschampsia flexuosa (Df). Numbering of plant individuals (e.g. Pa1) refer to the sample number of a plant specimen. Plants specimens of the ‘magnifying glass’ were collected within a small circle (2 m in diameter) of the plot. Coloured, dotted lines connect 100% identical ITS-genotypes observed in the roots of different plant specimens. The numbered ITS-genotypes (1–19) represent the ARON (Ascomycete Research group of Oslo, Norway) numbers listed below (numbers separated by indicate that root isolates or extracts originate from the same plant specimen, while = indicate root isolates or extracts originating from different plant specimens): (1) 2968.S; (2) R0740; (3) 2970.S  2967.S  2969.S; (4) 2961.S; (5) R0747; (6) 3048.S  3049.S  R1113 (7) 3008.S; (8) 3043.S; (9) 3060.S; (10) 3041.S; (11) 2965.S; (12) R0793 = 2956.S  2958.S; (13) 2954.S  2955.S; (14) R1138; (15) R1140  R1136R1137; (16) 3003.S = 2953.S; (17) 2810.S = 2938.S = R0748 = R1115R1114; (18) R1084R1085 = R1108R1108 = R1133R1135R1139; (19) 2963.S, 20 = 2962.S and 21 = 2971.S. (b) The 3-m × 3-m plot (NHF2[4]) of the Cu-mine spoil. An ITS-phylogeny based on the 13 observed genotypes is shown to the left. Symbols are as described in (a). Plant species not listed in (a) include: Salix herbaceae (Sh) and Vaccinium vitis-idaea (Vvi). 100% identical genotypes recorded in different plant individuals are connected by coloured, dotted lines. The numbered ITS-genotypes (22–33) represent the following ARON-numbers (the designation and = as explained in a): (22) 2995.S; (23) 2996.S; (24) R0892; (25) R0996; (26) R1176 = 2983.S = R0854; (27) R0903; (28) R1045R1947; (29) 3010.S = 3015.S; (30) 3042.S; (31) 3055.S  3057.S R1044R1050; (32) R1007; (33) 3021.S  3018.S, 34 = 3012.S  3014.S. None of the numbered ITS-genotypes of the two plots (a) and (b) were overlapping.

Download figure to PowerPoint

Fig. 3(b) summarises the local distribution of ITS-genotypes of the H. ericae aggregate within a 3-m × 3-m plot (NHF2[4]) of the Cu-mine-spoil. Eleven plant individuals of birch, pine, cowberry and willow (four, four, two and one specimens, respectively) harboured altogether 13 ITS-genotypes, of which 11 were observed only from one root system each. One ITS-genotype co-occurred in the root systems of one birch and two pine seedlings, and one co-occurred in two neighbouring specimens of cowberry and willow (Fig. 3b). Nine Helotiales sequences (six ITS-genotypes; Fig. 1), were obtained from pbECM-roots of birch and pine within the NHF2(4) plot (Tables 1 and 3).


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

The traditional view that ectomycorrhizal and ericoid mycorrhizal fungi are taxonomically distinct (Smith & Read, 1997) has been challenged by a number of recent papers, using molecular approaches and synthesis experiments in order to study fungal diversity, taxonomy and mycorrhizal relationships (Monreal et al., 1999; Bergero et al., 2000; Vrålstad et al., 2000, 2002). Bergero et al. (2000) demonstrated that identical fungal genets coexisted in ECM roots of Quercus ilex and ERM roots of Erica arborea. One of these isolates (Sd9 = AF269067; Fig. 1) formed ERM with E. arborea in vitro, and also formed a poorly developed mantle, and sometimes a rudimentary Hartig net with roots of Q. ilex in vitro. Several other isolates derived from ECM roots of Q. ilex in pure forest stands devoid of E. arborea formed ERM with E. arborea in vitro. The authors hypothesised that the recorded ECM associated fungi may represent an ericoid mycorrhizal fungal ‘reservoir’ in postcutting and postfire Mediterranean forest areas, serving as a source of mycorrhizal inoculum for E. arborea seedlings that recolonise disruptively disturbed habitats (Bergero et al., 2000). The present study adds credence to the view that fungal mycobionts may coexist in ericoid and ectomycorrhizal hosts. In two cases, 100% identical ITS-genotypes were observed from both ericoid and ectomycorrhizal roots, one involving geographically separated plants of oak and heather (clade 1 Fig. 2 and Table 3), and one involving neighbouring plants of cowberry and willow from the Cu-mine spoil (NHF2; Fig. 3b).

In a previous study (Vrålstad et al., 2002) we found that isolates of the H. ericae group (clade 3, Fig. 2) of ERM origin formed classic ERM associations, while P. finlandia-like strains of ECM origin (clade 4C–D; Fig. 2) as well as a single isolate of clade 2 (Fig. 2), formed ECM. Tested isolates of clade 1 (Fig. 2) did not form ERM or ECM, and their mycorrhizal potentials remain unclear (Vrålstad et al., 2002). The results could possibly indicate an evolution of strict ericoid-, ecto-, and nonmycorrhizal lineages within the H. ericae aggregate. In the present study, the majority of putative pbECM mycobionts grouped in the heterogeneous and poorly supported clade 4, including P. finlandia, and the majority of ericoid isolates grouped in the 100% supported H. ericae group (Fig. 2, clade 3). However, also identical and nearly identical ITS-genotypes were detected in ectomycorrhizal and ericoid mycorrhizal plants (Figs 1, 2 and 3). The predominantly ericoid H. ericae group included genotypes of pbECM root isolates and extracts that shared 97.8–99.5% SI with the H. ericae ex-type (Fig. 2). This is within the level of ITS-sequence discrepancy observed between the ex-types of H. ericae and S. vaccinii. Hence, the host range of fungal strains within the H. ericae-S. vaccinii species complex may as well include ectomycorrhizal plants. Similarly, some of the ericoid isolates grouped in the predominantly ectomycorrhizal clade 4 (Fig. 2). However, the lack of mycorrhizal synthesis studies prevents us from drawing conclusions about the mycorrhizal behaviour of these strains. The co-occurrence of identical and nearly identical ITS-genotypes in ECM and ERM hosts could possibly indicate a presence of hyphal networks between ERM and ECM plants. On the other hand, the ITS region may not necessarily track host specificity within the H. ericae aggregate. Genes involved in host specificity in this group could have a reticulate phylogenetic relationship to ITS sequences, as would be expected if sexual reproduction or hybridisation occurred. Only a small amount of genetic exchange via hyphal fusion, or via occasional sexual reproduction, is required to induce linkage disequilibrium between ITS loci and loci responsible for host specificity. Under such conditions, ITS can not serve as a marker for different host groups.

As more sequences have become available, the view of diversity in the group of H. ericae-like fungi has changed (Chambers et al., 1999; McLean et al., 1999; Monreal et al., 1999; Sharples et al., 2000). ITS fingerprinting approaches also suggest that the molecular diversity in this group of fungi is high (Stoyke et al., 1992; Perotto et al., 1996; Monreal et al., 1999). Our data indicate that the ITS-variability of putatively ectomycorrhizal representatives of the H. ericae aggregate may be equally diverse (Fig. 2). We experienced that it was easier to record new genotypes than to trace previously discovered ones, suggesting that only a small fraction of the total genotype diversity has been detected. Interestingly, a 100% identical ITS-genotype of the H. ericae aggregate was observed from different continents, that is Norway and Canada (Fig. 2; clade 4B), even though the local diversity within a few square meters was substantial (Fig. 3). In a previous study by Vrålstad et al. (2000), the number of recorded pbECM ITS1-genotypes within the postfire site (NOM) was two in 1997 (viz. G10 and G11) and four in 1998 (viz. G7, G12, G14, G15). None of the recorded genotypes were identical in the two subsequent years. The pbECM sample from the same postfire site presented in the present study (from 1999) yielded 14 ITS1-5.8S-ITS2-genotypes. Looking at the ITS1 portion of the sequences in isolation, they represent nine new and two previously reported ITS1-genotypes. Perhaps the pbECM diversity increased and the genotype composition changed successively with the time after the fire. This could be due to recombinations or mutations within the population of pbECM mycobionts at the postfire site. Another explanation may be that new fungal genotypes were introduced from outside the sampling site after 1997, or that they were overlooked in the previous years because the sampling effort was far too low to yield a reliable picture of the postfire pbECM community (Taylor, 2002). Some of the sequence divergence could be due to sequencing or PCR-amplification artefacts. However, the PCR-products were sequenced manually in both directions, all sequences were controlled at least twice, and all variable sites were inspected and verified on the gels. We therefore believe that artefact-introduced diversity has been reduced to a minimum. The recent years’ molecular revolution in ectomycorrhizal ecology has revealed that the ECM communities are impressively diverse and that species below ground are patchily distributed at a fine scale (Horton & Bruns, 2001). A similar picture may arise when we look at ECM morphotypes separately, for example the Piceirhiza bicolorata. Hence, caution must be exercised when ECM diversity data are interpreted, and a critical future challenge is to establish sampling protocols that can accurately determine ECM diversity (Taylor, 2002). Grogan et al. (2000) studied the below ground ECM-community structure in a recently burnt bishop pine forest in northern California. As much as 50% of the observed ECM fungal taxa were recorded only once. ECM fungi recorded more than once were randomly distributed on the seedlings across the study site. The patterns suggested that the postfire mycorrhizal community primarily arose from a successful colonisation by randomly distributed point_source fungal inocula within the upper mineral soil layer of the forest floor (Grogan et al., 2000). The high degree of ITS-genotype variation combined with the low prevalence of each ITS-genotype in the habitats in our study may as well indicate that plants have been inoculated from such randomly distributed and low-frequency point sources of fungal strains of the H. ericae aggregate. Alternatively, closely related ITS-genotypes may possibly form part of a common, haploid mycelium, where recent mutations have led to local divergence in the genotype composition within the vegetative mycelium of a fungal strain.

The observed discrepancy between the putative pbECM mycobionts of the postfire site and the Cu-mine spoils is striking and needs some comments. The mycobionts of pbECM observed from the postfire site grouped, with three exceptions, in clade 4 (group C and E) of the H. ericae aggregate (Figure 2) together with previously confirmed ECM-forming isolates. By contrast, only 36 of 88 pbECM sequences from the mine spoils were referable to the H. ericae aggregate. The morphological variation of the pbECM-like morphotypes of the mine spoil plants was rather large compared with that of the postfire site plants. Hence, fungi other than members of the H. ericae aggregate with partially melanised mantles characteristic of the pbECM may be involved as well. Melanised cell walls are not commonly produced by ectomycorrhizal fungal species, but tomentelloid basidiomycetes (Kõljalg et al., 2000) and some ectomycorrhizal ascomycetes (Wilcox & Wang, 1987a,b) in addition to the ectomycorrhizal members of the H. ericae aggregate, possess this feature. A variety of heavy metals may induce or accelerate the production of fungal melanins (Gadd, 1993; Martino et al., 2000), and it is not unlikely that a broader range of fungi may develop melanised pbECM-like morphotypes as a response to heavily metal-polluted environments.

Several axenic culture isolates and single pbECM root extracts in the study sample contained nrDNA versions both with and without introns (Table 1), and some isolates contained introns in more than one position (data not shown). Previous investigations suggest that the diversity of group I introns in fungi of the Helotiales is high (Holst-Jensen et al., 1999; Abeln et al., 2000; Perotto et al., 2000). In three instances, minor discrepancies (1–2 nucleotide differences) were observed between the previously published ITS1-sequences (Vrålstad et al., 2000) and reamplified sequences of this study (data not shown). This may be explained by different PCR-regimes; in the former study a H. ericae aggregate specific PCR primer (HericaeITS1) was used, while an intron-specific (Hint) primer that may amplify different nrDNA versions, that is those with introns, was used in the present study.

For the purpose of identifying ectomycorrhizal candidates, sequences obtained from axenic culture isolates alone may be of limited value (cf. clade I (A + B), II, V and VI of Fig. 1). There is always a possibility that isolates obtained from surface sterilised ECM roots may represent occasional root-endophytes rather than true ECM fungi, since the process of isolation may favour the fastest growing fungus that is present in the root. By contrast, a fungal ITS-genotype obtained from an ECM root by direct PCR-amplification using universal primers will normally represent the dominant fungal component of the ECM root (the presumed ECM mycobiont). We therefore believe that the true pbECM mycobiont has been detected at least in those cases where the same ITS-genotype was obtained from universally amplified pbECM root isolates and extracts (cf. Fig. 1, clade III; Fig. 2, clade 4, group C, E).

In this study, we used ITS-sequences and parsimony analyses for the purpose of studying the diversity and phylogenetic affinities of root-symbiotic helotialean fungi from burnt and metal polluted boreal forest habitats. The majority of the observed helotialean root-associated fungi not referable to the H. ericae aggregate were in most cases represented by axenic culture root isolates obtained from various boreal forest plants. Little is known about the ecology and possible symbiotic role of representatives of the Helotiales. We detected nearly identical isolates in roots of fescue and birch that showed strong affinity to Pyrenopeziza revincta of the Dermateaceae (> 96% SI), and some isolates derived from hairy grass and cowberry roots showed strong affinity to Lachnum cf. pygmaeum of the Hyaloscyphaceae (c. 98% SI). The sequence of a confirmed ericoid mycorrhizal isolate (AF252842; Sharples et al., 2000) grouped in the B group of clade I (Fig. 1), suggesting that some sterile ericoid mycorrhizal mycelia may be related to some of the dermateacean taxa (e.g. Tapesia and Mollisia). Another fungus recognized in this group is the dark septate endophyte Phialocephala fortinii, commonly isolated from a broad range of plant roots (Jumpponen & Trappe, 1998; Jumpponen et al., 1998; Addy et al., 2000; Jumpponen, 2001). The species is also readily amplified from ECM roots directly (Jonsson et al., 1999). The symbiotic potentials of P. fortinii range from pathogenic to commensal to mutualistic (Fernando & Currah, 1996; Jumpponen & Trappe, 1998; Jumpponen et al., 1998). We obtained fungal isolates from surface sterilised roots of hairy grass, heather, cowberry and pine that shared 97.7–100% SI with previously reported ITS-sequences of P. fortinii (Addy et al., 2000). These isolates were distributed in boreal oak-pine forest, clear-felled spruce forest, and postfire and metal-polluted habitats, which corroborates the reported ubiquitous occurrence of P. fortinii in a wide range of plants and habitats worldwide (Jumpponen & Trappe, 1998; Addy et al., 2000). Previous studies have connected P. fortinii to the Helotiales based on observations of sterile discocarps (Currah & Tsuneda, 1993) and SSU nrDNA sequences (LoBuglio et al., 1996). Our analyses revealed that P. fortinii show phylogenetic affinity to members of the Dermateaceae (Fig. 1, clade 1, group B), and suggests that a teleomorph of P. fortinii, if extant, may be found among Tapesia, Mollisia or other genera of this family. The analyses also suggest that H. ericae is not congeneric with H. fructigenus, the type species of Hymenoscyphus, and probably not with any of the Hymenoscyphus spp. included in the analyses (Fig. 1). Instead, H. ericae, P. finlandia and a diverse group of nameless ± darkly pigmented root-associated fungi constitute a 100% supported monophyletic group (Fig. 2). The observed ITS-sequence divergence of the H. ericae aggregate is within the level of variation observed in various genera of the Sclerotiniaceae (Holst-Jensen et al., 1997a, 1997b, 1998), Rutstroemiaceae (Schumacher & Holst-Jensen, 1997), Hyaloscyphaceae (Cantrell & Hanlin, 1997) and Dermateaceae (Abeln et al., 2000) of the Helotiales. This may suggest that the monophyletic H. ericae aggregate comprises a group of species comparable with a generic unit that should not be classified as Hymenoscyphus. The aggregate groups with members of the Hyaloscyphaceae and the Dermateaceae (Fig. 1, clade1). However, highly variable ITS-data cannot be used to infer phylogenetic relationships at suprageneric taxonomical levels (Holst-Jensen et al., 1997b; Abeln et al., 2000), and the phylogenetic position of the H. ericae aggregate within the Helotiales is still unresolved.

Based on ITS1-sequence similarity, Egger & Sigler (1993) suggested that S. vaccinii may represents the anamorphic state of H. ericae, a view that has been adopted by most subsequent authors (Straker, 1996; Smith & Read, 1997; Hambleton et al., 1999; Monreal et al., 1999; Sharples et al., 2000). While the ITS-sequence discrepancy between the H. ericae and S. vaccinii ex-types is as much as 2.2%, the level of intraspecific ITS-variation in teleomorphic species of the helotialean families Sclerotiniaceae (Holst-Jensen et al., 1997a, 1998), Rutstroemiaceae (Schumacher & Holst-Jensen, 1997) and Dermateaceae (Abeln et al., 2000) is very low. From the limited knowledge we have of sexuality and teleomorph formation within this group of fungi, it is not possible to discern whether the isolates recognized within the H. ericae-S. vaccinii range (i.e. clade 3; Fig. 2) belong to a single species or not. In the absence of teleomorph records, a phylogenetic species recognition that relies on concordance between several independent gene genealogies (cf. Taylor et al., 2000) may be the most reasonable way to resolve the taxonomy of the H. ericae aggregate, although such an approach is evidently a laborious task. However, a phylogenetic species recognition may provide reliable answers to where the borders between intra- and interspecific variation of the H. ericae aggregate should be drawn.


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References

This work partially fulfils the requirements for T. Vrålstad’s Ph.D. dissertation on ‘molecular ecology of root-associated mycorrhizal and nonmycorrhizal ascomycetes’, and was financially supported by the University of Oslo (UiO), Norway, a scholarship from the Research Council of Norway (NFR), and grants from the Nansen foundation (Norway) and the Sønnerland foundation (Norway) to T. Vrålstad. The laboratory work was conducted at the Mycolab and the DNA laboratory for biosystematics and ecology, UiO. We wish to thank Anne-Cathrine Scheen and Mia Knudsen for technical assistance. We are most grateful to Arne Holst-Jensen for fruitful discussions, critical reading and valuable comments on the manuscript. We acknowledge Keith Egger and another anonymous referee for their insightful criticisms, comments and contributions to the manuscript. The study is part of T. Schumacher’s project on molecular ecology of mycorrhizal and nonmycorrhizal ascomycetes supported by the NFR (Grant 115538/410).


  1. Top of page
  2. Summary
  3. Introduction
  4. Materials and Methods
  5. Results
  6. Discussion
  7. Acknowledgements
  8. References
  • Abeln ECA, De Pagter MA, Verkley GJM. 2000. Phylogeny of Pezicula, Dermea and Neofabraea inferred from partial sequences of the nuclear ribosomal RNA gene cluster. Mycologia 92: 685–693.
  • Addy HD, Hambleton S, Currah RS. 2000. Distribution and molecular characterisation of the root endophyte Phialocephala fortinii along an environmental gradient in the boreal forest of Alberta. Mycological Research 104: 12131221.
  • Bergero R, Perotto S, Girlanda M, Vidano G, Luppi AM. 2000. Ericoid mycorrhizal fungi are common root associates of a Mediterranean ectomycorrhizal plant (Quercus ilex). Molecular Ecology 9: 16391649.
  • Brand F, Gronbach E, Taylor AFS. 1992. Piceirhiza bicolorata. In: AgererR, ed. Colour atlas of ectomycorrhizae. Schwäbisch Gmünd, Germany: Einhorn-Verlag, plate 73.
  • Cantrell SA, Hanlin RT. 1997. Phylogenetic relationships in the family Hyaloscyphaceae inferred from sequences of ITS regions, 5.8S ribosomal DNA and morphological characters. Mycologia 89: 745755.
  • Carpenter SE. 1988. Leotiales, a name to replace Helotiales (Ascomycotina). Mycologia 80: 127130.
  • Cech TR, Herschlag D. 1996. Group I ribozymes: substrate recognition, catalytic strategies and comparative mechanistic analysis. In: EcksteinF Lilley DMJ, eds. Catalytic RNA. Berlin, Heidelberg, Germany: Springer Verlag, 118.
  • Chambers SM, Williams PG, Seppelt RD, Cairney WG. 1999. Molecular identification of Hymenoscyphus sp. from rhizoids of the leafy liverwort Cephaloziella exiliflora in Australia and Antarctica. Mycological Research 103: 286288.
  • Currah RS, Tsuneda A. 1993. Vegetative and reproductive morphology of Phialocephala fortinii (Hyphomycetes, Mycelium radicis atrovirens) in culture. Transactions of the Mycological Society of Japan 34: 345356.
  • Egger KN, Osmond G, Goodier JL. 1995. Sequence and putative secondary structure of group I introns in the nuclear-encoded ribosomal RNA genes of the fungus Hymenoscyphus ericae. Biochimica et Biophysica Acta 1261: 275278.
  • Egger KN, Sigler L. 1993. Relatedness of the ericoid endophytes Scytalidium vaccinii and Hymenoscyphus ericae inferred from analysis of ribosomal DNA. Mycologia 85: 219230.
  • Fernando AA, Currah RS. 1996. A comparative study of the effects of the root endophytes Leptodontidium orchidicola and Phialocephala fortinii (Fungi Imperfecti) on the growth of some subalpine plants in culture. Canadian Journal of Botany 74: 10711078.
  • Gadd GM. 1993. Interactions of fungi with toxic metals. New Phytologist 124: 2560.
  • Gardes M, Bruns T. 1993. ITS primers with enhanced specificity for basidiomycetes – application to the identification of mycorrhiza and rusts. Molecular Ecology 2: 113118.
  • Gernandt DS, Camacho FJ, Stone JK. 1997. Meria laricis, an anamorph of Rhabdocline.
  • Grogan P, Baar J, Bruns TD. 2000. Below-ground ectomycorrhizal community structure in a recently burned bishop pine forest. Journal of Ecology 88: 10511062.
  • Hall TA. 1999. BioEdit: a user-friendly biological sequence alignment editor and Analysis Program for Windows 95/98/NT. BioEdit (Biological Sequence Alignment Editor). Nucl. Acids. Symposium. Series 41: 9598.
  • Hambleton S, Huhtinen S, Currah RS. 1999. Hymenoscyphus ericae: a new record from western Canada. Mycological Research 103: 13911397.
  • Holst-Jensen A, Kohn LM, Jakobsen KS, Schumacher T. 1997a. Molecular phylogeny and evolution of Monilinia (Sclerotiniaceae) based on coding and noncoding rDNA sequences. American Journal of Botany 84: 686701.
  • Holst-Jensen A, Kohn LM, Schumacher T. 1997b. Nuclear rDNA phylogeny of the Sclerotiniaceae. Mycologia 89: 885899.
  • Holst-Jensen A, Vaage M, Schumacher T. 1998. An approximation to the phylogeny of Sclerotinia and related genera. Nordic Journal of Botany 18: 705719.
  • Holst-Jensen A, Vaage M, Schumacher T, Johansen S. 1999. Structural characteristics and possible horizontal transfer of group I introns between closely related plant pathogenic fungi. Molecular Biology and Evolution 16: 114126.
  • Horton TR, Bruns TD. 2001. The molecular revolution in ectomycorrhizal ecology: peeking into the black-box. Molecular Ecology 10: 18551871.
  • Jonsson L, Dahlberg A, Nilsson M-C, Zackrisson O, Kårén O. 1999. Ectomycorrhizal fungal communities in late-successional Swedish boreal forests, and their composition following wildfire. Molecular Ecology 8: 205215.
  • Jumpponen A. 2001. Dark septate endophytes – are they mycorrhizal? Mycorrhiza 11: 207211.
  • Jumpponen A, Mattson KG, Trappe JM. 1998. Mycorrhizal functioning of Phialocephala fortinii with Pinus contorta on glacier forefront soil: Interactions with soil nitrogen and organic matter. Mycorrhiza 7: 261265.
  • Jumpponen A, Trappe JM. 1998. Dark septate endophytes: a review of facultative biotrophic root-colonising fungi. New Phytologist 140: 295310.
  • Kõljalg U, Dahlberg A, Taylor AFS, Larsson E, Hallenberg N, Stenlid J, Larsson KH, Fransson PM, Kårén O, Jonsson L. 2000. Diversity and abundance of resupinate thelephoroid fungi as ectomycorrhizal symbionts in Swedish boreal forests. Molecular Ecology 9: 19851996.
  • Korf RP, Lizon P. 2000. Validation of Nannfeldt’s ordinal name Helotiales. Mycotaxon 75: 501502.
  • Korf RP, Lizon P. 2001. The status of the ordinal name Leotiales. Czech Mycology 52: 255257.
  • Liou GY, Tzean SS. 1997. Phylogeny of the genus Arthrobotrys and allied nematode-trapping fungi based on rDNA sequences. Mycologia 89: 876884.
  • LoBuglio KF, Berbee ML, Taylor JW. 1996. Phylogenetic origins of the asexual mycorrhizal symbiont Cenococcum geophilum Fr. & other mycorrhizal fungi among the ascomycetes. Molecular Phylogenetics and Evolution 6: 287294.
  • Ma LJ, Rogers SO, Catranis CM, Starmer WT. 2000. Detection and characterisation of ancient fungi entrapped in glacial ice. Mycologia 92: 286295.
  • Martino E, Turnau K, Girlanda M, Bonfante P, Perotto S. 2000. Ericoid mycorrhizal fungi from heavy metal polluted soils: their identification and growth in the presence of zinc ions. Mycological Research 104: 338344.
  • McLean CB, Cunnington JH, Lawrie AC. 1999. Molecular diversity within and between ericoid endophytes from the Ericaceae and Epacridaceae. New Phytologist 144: 351358.
  • Monreal M, Berch SM, Berbee M. 1999. Molecular diversity of ericoid mycorrhizal fungi. Canadian Journal of Botany 77: 15801594.
  • Nannfeldt JA. 1932. Studien über die Morphologie und Systematik der nicht-lichenisierten inoperculaten Discomyceten. Nova Acta Regiae Societatia Scientiarum Uppsaliensia Seriea 4: 1368.
  • Pearson WR, Lipman DJ. 1988. Improved tools for biological sequence comparison. Proceedings of National Academy of Sciences, USA 85: 24442448.
  • Perotto S, Actis-Perino E, Perugini J, Bonfante P. 1996. Molecular diversity of fungi from ericoid mycorrhizal roots. Molecular Ecology 5: 123131.
  • Perotto S, Nepote-Fus P, Saletta L, Bandi C, Young JPW. 2000. A diverse population of introns in the nuclear ribosomal genes of ericoid mycorrhizal fungi includes elements with sequence similarity to endonuclease-coding genes. Molecular Biology and Evolution 17: 4459.
  • Schumacher T, Holst-Jensen A. 1997. A synopsis of the genus Scleromitrula (=Verpatinia) (Ascomycotina: Helotiales: Sclerotiniaceae). Mycoscience 38: 5569.
  • Sharples JM, Chambers SM, Meharg AA, Cairney JWG. 2000. Genetic diversity of root-associated fungal endophytes from Calluna vulgaris at contrasting field sites. New Phytologist 148: 153162.
  • Smith SE, Read DJ. 1997. Mycorrhizal symbiosis. London, UK: Academic Press.
  • Stoyke G, Egger KN, Currah RS. 1992. Characterisation of sterile endophytic fungi from the mycorrhizae of subalpine plants. Canadian Journal of Botany 70: 20092016.
  • Straker CJ. 1996. Ericoid mycorrhiza: Ecological and host specificity. Mycorrhiza 6: 215225.
  • Swofford DL. 1999. Paup *: Phylogenetic Analysis Using Parsimony (* and Other Methods). Sunderland, Massachusetts: Sinauer Associates.
  • Taylor AFS. 2002. Fungal diversity in ectomycorrhizal communities: Sampling effort and species detection. Plant and Soil (In press.)
  • Taylor JW, Jacobson DJ, Kroken S, Kasuga T, Geiser DM, Hibbett DS, Fisher MC. 2000. Phylogenetic species recognition and species concepts in fungi. Fungal Genetics and Biology 31: 2132.
  • Vrålstad T, Fossheim T, Schumacher T. 2000. Piceirhiza bicolorata– the ectomycorrhizal expression of the Hymenoscyphus ericae aggregate? New Phytologist 145: 549563.
  • Vrålstad T, Holst-Jensen A, Schumacher T. 1998. The postfire discomycete Geopyxis carbonaria (Ascomycota) is a biotrophic root associate with Norway spruce (Picea abies) in nature. Molecular Ecology 7: 609616.
  • Vrålstad T, Schumacher T, Taylor AFS. 2002. Mycorrhizal synthesis between fungal strains of the Hymenoscyphus ericae aggregate and potential ectomycorrhizal and ericoid hosts. New Phytologist 153: 143152.
  • Wang CJK, Wilcox HE. 1985. New species of ectendomycorrhizal and pseudomycorrhizal fungi: Phialophora finlandia, Chloridium paucisporum, and Phialocephala fortinii. Mycologia 77: 951958.
  • White TJ, Bruns T, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: InnisMA, GelfandDH, SninskyJJ, WhiteTJ, eds. PCR protocols: a guide to methods and applications. San Diego, CA, USA: Academic Press, 315322.
  • Wilcox HE, Wang CJK. 1987a. Ectomycorrhizal and ectendomycorrhizal associations of Phialophora finlandia with Pinus resinosa, Picea rubens, and Betula alleghaniensis. Canadian Journal of Forest Research 17: 976990.
  • Wilcox HE, Wang CJK. 1987b. Mycorrhizal and pathological associations of dematiaceous fungi in roots of 7-month-old tree seedlings. Canadian Journal of Forest Research 17: 884899.
  • Yu TEJ-C, Egger KN, Peterson RL. 2001. Ectendomycorrhizal associations – characteristics and functions. Mycorrhiza 11: 167177.