Divergent arbuscular mycorrhizal fungal ... - Wiley Online Library

Research

Divergent arbuscular mycorrhizal fungal communities colonize roots of Pulsatilla spp. in boreal Scots pine forest and grassland soils Blackwell Publishing Ltd.

Maarja Öpik1,2, Mari Moora1, Jaan Liira1, Urmas Kõljalg1, Martin Zobel1 and Robin Sen2 1

Institute of Botany and Ecology, University of Tartu, 40 Lai Street, 51005 Tartu, Estonia; 2Department of Biosciences, Division of General Microbiology, Viikki

Biocenter, PO Box 56, 00014 University of Helsinki, Finland

Summary Author for correspondence: Maarja Öpik Tel: +372 7 376224 Fax: +372 7 376222 Email: [email protected] Received: 28 May 2003 Accepted: 11 August 2003 doi: 10.1046/j.1469-8137x.2003.00917.x

• Communities of arbuscular mycorrhizal (AM) fungi were characterized in roots of rare Pulsatilla patens and common P. pratensis native adults and seedlings grown in soils from Estonian boreal forest and grassland habitats. Since establishment of Pulsatilla species predominantly occurs in vegetation-free gaps, seedling baiting experiments were aimed at gap simulation. • The AM fungal small subunit ribosomal RNA gene (SSU rDNA) sequences amplified from roots were subjected to denaturing gradient gel electrophoresis (DGGE), cloning, restriction fragment length polymorphism (RFLP) grouping, sequence phylogenetic and multivariate analyses. • Nineteen identified sequence groups comprised 14 putative Glomus, two Acaulospora, two Scutellospora and one Gigaspora groupings. Four and six groupings, respectively, contained previously described species and root-derived AM fungal sequences. Sequence groups were identified in seedling roots that were more abundant in a grassland (Glomus sp. MO-G3) or a forest soil (Glomus spp. MO-G2 and MO-G5). • Our data showed site-dependent differences in AM fungal community composition, but we failed to identify AM fungi specifically or preferentially colonizing the rare plant species. Key words: endangered plants, Pulsatilla, arbuscular mycorrhizal (AM) fungi, SSU rDNA sequences, denaturing gradient gel electrophoresis (DGGE), phylogenetics, boreal forest, seedling. © New Phytologist (2003) 160: 581–593

Introduction In recent years there has been an increasing awareness of the central importance of root symbiotic mycorrhizal fungi in plant ecology through their control of soil nutrient uptake, root pathogens and intra- and inter-plant species linkages (Zobel et al., 1997; Watkinson, 1998; Sen, 2003). Interactions between arbuscular mycorrhizal (AM) and pathogenic fungi have, in particular, been shown to influence plant abundance and invasiveness ( Newsham et al., 1995; Bever et al., 1997; Klironomos, 2002; Bever, 2003). Although AM fungi in general colonize the roots of a taxonomically diverse range of plants, ecological specificity does

© New Phytologist (2003) 160: 581–593 www.newphytologist.com

occur in these symbiotic associations (McGonigle & Fitter, 1990; Helgason et al., 2002). Certain plant–AM fungus combinations, as opposed to others, provide relatively more benefit to symbiotic partners in terms of overall fitness and production (Sanders, 1993; van der Heijden et al., 1998). Thus, in relevant plant community studies the traditional comparative ‘presence / absence’ assessment via root staining is now being replaced by the more informative study of the distribution and impact of natural AM fungi (Clapp et al., 1995; van der Heijden et al., 1998; Helgason et al., 2002). Potentially concomitant variation in plant species composition and AM fungal spore communities in soils has been identified between (Johnson, 1993; Merryweather & Fitter,

581

582 Research

1998) and within (Eom et al., 2000) plant communities. If specific compatible relationships involving certain AM fungal and plant taxa are required for mutual symbiont survival, the loss of compatible fungal species or individuals may select against, or limit the distribution of dependent host plant species or individuals. A better appreciation of the outcome of host–AM fungal community interactions on plant establishment and productivity in native habitats can only be gained through improved identification of fungal root occupancy in both time and space. The major impasse, in achieving this, was initially overcome through fungal isozyme (allozyme) identification (Rosendahl & Sen, 1992) that has now been superseded by high-resolution molecular DNA identification methodology (Clapp et al., 2002a). The main targets have been the fungal ribosomal RNA gene sequences, particularly the small subunit (SSU) rDNA, where specific AM fungal sequence polymorphisms have been identified and subjected to phylogenetic analyses in order to resolve relationships at approximately the species level. An increasing number of studies are now being published, using the molecular approach, that describe in planta distribution of AM fungi in arable fields and diverse natural ecosystems, including grasslands together with temperate, rain and sub-Antarctic forests (Helgason et al., 1998, 1999, 2002; Daniell et al., 2001; Bidartondo et al., 2002; Husband et al., 2002a,b; Russell et al., 2002; Vandenkoornhuyse et al., 2002a,b). Host plants investigated are also widely represented e.g. forbs, grasses, native and cultivated, woody plants, including tree seedlings, temperate and rain forest species, those forming nodulating symbioses, and epiparasitic plants. What has been shown is the existence of striking levels of variation in AM fungal species/taxa richness in these different ecosystems: from limited numbers in arable fields (Helgason et al., 1998; Daniell et al., 2001) to over 20 AM fungal taxa associated with temperate grassland and tropical rain forest plants (Husband et al., 2002a,b; Vandenkoornhuyse et al., 2002b). A considerable diversity of novel fungal taxa, on the basis of highly divergent SSU sequences, have been recovered from roots in the native vegetated systems. Most significantly, the more commonly described and studied sporulating species (e.g. Glomus mosseae) were often poorly represented in the root-colonizing AM fungal communities. Based on these relatively recent findings, it is obvious that AM fungal diversity is far higher (Helgason et al., 2002) and root colonization community dynamics more complex than previously thought. The present study was aimed at a complementary SSU sequence-based investigation of the distribution of AM fungal species in the roots of a congeneric pair of perennial Estonian plant species of divergent local abundance, Pulsatilla patens (L.) Mill. and Pulsatilla pratensis (L.) Mill. (Ranunculaceae). The former is now rare, with a limited distribution mainly restricted to boreal forest locations while the latter is more common in Estonia, being found in both forest and grassland locations in Estonia (Pilt & Kukk, 2002). The habitats in

question form integral parts of Northern boreal and boreonemoral ecosystems where there still remains a lack of information on AM fungal community diversity as effort, to date, by us and others has mainly focused on characterization of root symbiotic ectomycorrhizal fungal communities of predominating over-storey tree species (Dahlberg, 2001; Kõljalg et al., 2002). Following seed germination and subsequent exhaustion of seed reserves, successful seedling establishment requires efficient plant acquisition of soil nutrients as well as protection against pathogens, which in the majority of plant families is controlled, or at least influenced by root symbioses with mycorrhizal fungi (Smith & Read, 1997). Plant establishment from seeds, in the case of both Pulsatilla species, takes place almost exclusively in disturbed local gaps and areas (e.g. exposed coast-line habitats or gravel pits) (Uotila, 1996; Pilt & Kukk, 2002). Because of their predominant establishment in microsites that lack vegetation, it could be hypothesized that spores of AM fungi are the main source of infective propagules for Pulsatilla spp. seedlings and that colonization via functioning hyphal network is less significant. We propose that a rare plant species is associated with specific fungal symbiont(s), and that presence of the fungus/ fungi is therefore required for successful seedling establishment. Accordingly, we hypothesize that sites where the rare plant is naturally growing, differ in their AM fungal communities from sites where only the common congener is present. To test for the presence of host- and/or site-related AM fungi, we performed a reciprocal seedling establishment experiment involving the rare and common Pulsatilla spp. growing on soils from sites supporting both or only the common plant species. We tried, in the experimental design, to simulate prevalent conditions during early stages of seedling establishment. Seeds of both Pulsatilla species were sown in pots containing two different natural soil inocula, but lacking other vegetation, which allowed relevant investigation of AM fungal species composition in roots of emerging seedlings. In particular, we aimed to identify differences in AM fungal community structure associated with Pulsatilla spp. roots that relate to (1) the contrasting source of soil inoculum (i.e. a Scots pine forest vs a dry grassland) and (2) host species in these soil inocula.

Materials and Methods Seedling establishment experiments and sampling As sufficient numbers of seedlings of P. pratensis and the rare P. patens, in particular, are almost impossible to locate in the natural vegetation, seed germination and seedling establishment experiments were conducted under controlled conditions in pots. Mature seeds of both species were collected at the end of June and beginning of July 1999 from three local populations

www.newphytologist.com © New Phytologist (2003) 160: 581–593

Research

Fig. 1 Map of Estonia showing the locations of sampled populations of Pulsatilla spp. from each of two forest (F), grassland (G) and borderline (B) communities. Soil inocula from forest F1 and grassland G1, supporting both species and just the common Pulsatilla species, respectively, were used for seedling establishment experiments. G1, dry meadow within fragmented agricultural landscape, F1, extensive boreal pine forest area, B1, open sandy area bordered by boreal pine forest, G2, dry meadow, F2, boreal pine forest, B2, roadside bordered by boreal pine forest. Open circles, plant roots; closed circles, soil and plant roots; C, common Pulsatilla pratensis; R, rare Pulsatilla patens.

of both Pulsatilla species in Estonia and were then pooled. Seeds, visually examined and carefully selected to avoid those attacked by herbivores or pathogenic fungi, were sown on 16 July 1999 (hereafter called the summer experiment) and on 8 February 2000 (hereafter called the spring experiment). These dates approximate the actual time of seedling establishment in nature – either immediately after seed set ( June–July), or from the transient over-wintered seed bank in spring (March–April). The natural soils used for the seedling establishment experiments originated from two sites – a dry meadow within an agricultural landscape at Pangodi, Central Estonia, hereafter called grassland G1, and an extensive boreal Scots pine forest at Soomaa, Central Estonia (forest F1, Fig. 1). The soils, dry arenosols with weakly differentiated horizons, were collected from 10 random locations in both target ecosystems in the first half of July 1999 and in the second half of August 1999, for the summer and spring experiment, respectively. Topsoil samples (a grey mineral layer at a depth of 2–10 cm underlying the thin litter layer) from each site were pooled for use as inoculum in the experiments. In the summer experiment, a 1 : 1 mixture of natural soil and sterile sand was used as an establishment substrate. In the spring experiment, a 1 : 1 mixture of the two natural soils was used, where one of the soils had been autoclaved (40 min at 121°C). Sterile sand (summer experiment) or a 1 : 1 mixture of two autoclaved soils (spring experiment) served as nonmycorrhizal controls. The summer experiment was conducted under natural outdoor conditions in the Botanical Garden of the University of Tartu and the spring experiment in a greenhouse of the Viikki Biocentre of Helsinki University in daylight (daylength 16 h) for 14 wk. Seeds were sown at a constant density (1.2 seeds cm−2) into rectangular pots (4 × 13 × 18 cm, depth × width × length) and circular pots (9 × 12 cm, depth × diameter), in the summer and spring experiments, respectively. In the


summer experiment, the seedlings were allowed to grow en masse until sampled. In the spring experiment, seedlings were later thinned to one individual per pot. Pots were carefully watered with tap water as required. Each treatment was replicated in six or 10 pots, in the respective summer and spring experiments. In the summer experiment, two seedlings from each pot were sampled (in two cases, only one seedling was available) at the age of 9 wk (time after mass germination) (Table 1). In the spring experiment, four to five seedlings (depending of the availability of seedlings) were harvested five and 14 wk after germination. The root AM fungal colonization rate ranged between 3% and 45% in 5-wk-old seedlings and 35–85% in 14-wk-old seedlings, respectively, in the spring experiment, while both species showed strong positive growth response in the presence of AM fungal colonization (M. Moora et al. unpubl. data). Sampling of field plant roots To provide background information on indigenous AM fungi inhabiting local Pulsatilla spp. roots, individual P. patens and P. pratensis plants were directly sampled from the sites where soils for experiments were collected in addition to a few other sites in August 1999 (Fig. 1). Sampling sites included two where both target species coexist, a Scots pine forest F1 (described above) and a dry open area alongside a railway line bordered by a boreal Scots pine forest in south-eastern Estonia (Piusa, site B1, from ‘border’), two sites supporting only the common P. pratensis, dry meadows within agricultural landscapes in central (G1, described above) and western Estonia (Varbla, site G2) and two sites supporting only the rare P. patens, a boreal Scots pine forest in southern Estonia (Vastseliina, site F2) and a roadside area bordered by boreal pine forest in southern-central Estonia (Palo, site B2) (Fig. 1).

583

584 Research

Table 1 Number of arbuscular mycorrhizal (AM) fungal sequence groups in roots of seedlings from two pot experiments and of field plants from six sites detected by denaturing gradient gel electrophoresis (DGGE), cloning of excised bands, restriction fragment length polymorphism (RFLP) typing and sequencing of representative clones

Sample set

No of AMF AMF sequence group No of sequence plants groups MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MO- MOSite/ Plant amplified (sampled) Total Meanb G1 G2 G3 G4 G5 G6 G7 G8 G9 G10 G11 G12 G13 G14 A1 A2 S1 S2 Gi1 soila sp.

1. Summer, 9 wk

G1

2. Spring, 5 wk 3. Spring, 14 wk Subtotal Relative abundanced 1. Summer, 9 wk F1 2. Spring, 5 wk 3. Spring, 14 wk www.newphytologist.com © New Phytologist (2003) 160: 581–593

Subtotal Relative abundance Subtotal Relative abundance 4. Field plants

G1 F1 B1 G2 F2 B2

Subtotal Relative abundance Grand total Relative abundance a

R C R C R C

R C R C R C

C R C R C C R R

1 (10) 4 (12) 4 (4) 4 (5) 5 (5) 4 (4) 22 (40)

1 3 5 9 2 3 13

1 1.5 1.75 3.5 1 1.5 1.77

5 (12) 5 (12) 2 (5) 3 (4) 5 (5) 4 (4) 24 (42)

3 3 3 6 4 3 10

1.6 1.4 1.5 3.33 1.6 1.5 1.75

46 (82)

18

1.76

2 (3) 1 (1) 1 (1) 2 (3) 3 (3) 1 (3) 1 (3) 1 (3) 12 (20)

3 1 1 2 2 1 1 1 6

2 1 1 1.5 1.33 1 1 1 1.33

58 (102)

19

1.67

1 1

2 1 2 1

1c 2

1 1 2

1 1 1 2 10 3 3 4.5 9.1 45.5 13.6 13.6 5 2 4 2 1 1 2 1 3 2 4 2 3 3 14 – – 15 12.5 58.3 0 0 62.5 4 16 10 3 18 8.7 34.8 21.7 6.5 39.1 2 1 1 2 1 3 1 1 1 1 9 1 3 1 – 75.0 8.3 25.0 8.3 0 13 17 13 4 18 22.4 29.3 22.4 6.9 31.0

1

1 4.5

1 2

1

– 0 1

2 9.1

– 0

3 2

3 – 13.6 0

1 1 4.5

1

4 5 4 1 22.7 18.2 0

1 4.5

2

– 0

– 0

2 9.1

– 0 2 4.3

– 0 2 3.4

1 1 1

1

2 8.3 3 6.5 1

1 4.2 1 2.2

– 0 2 4.3

– 0 – 0 1

–

1 8.3 4 6.9

– 0 1 1.7

– 0 2 3.4

1 8.3 1 1.7

–

2

1 0 4.2 3 1 6.5 2.2

– 0 1 2.2

–

– 0 0 3 1 5.2 1.7

– 0 1 1.7

– 0 5 8.3

–

1 0 4.2 4 1 8.7 2.2

2 8.3 3 6.5

1 4.2 1 2.2

1 1 2 8.3 2 4.3

–

– 0 0 4 1 6.9 1.7

– 0 3 5.2

– 0 1 1.7

– 0 2 3.4

0 5 10.9

Study sites as in Fig. 1. bAM fungal sequence groups per plant. cNo. of plants with given sequence group detected. dCalculated over plants amplified.

1

Research

The distance between sites was, in all cases, at least 60 km. Roots of three plant individuals (Table 1) of one or both species from each site (except F1, where only one individual was sampled, due to the small population size) were excavated from a depth of 15–45 cm and stored in 40% ethanol until processed. Control AM fungal strains Spore material from the following AM fungal strains was used as a control in denaturing gradient gel electrophoresis (DGGE): Acaulospora laevis Gerd. & Trappe (BEG 13), and Glomus intraradices N.C. Schenck & G.S. Sm. (IMA 6) provided by Prof. M. Giovannetti, Glomus mosseae (T.H. Nicolson & Gerd.) Gerd. & Trappe (BEG 84) from Prof. S. Rosendahl, Glomus geosporum (T.H. Nicolson & Gerd.) C. Walker (BEG 11) and Scutellospora castanea C. Walker (BEG 1) from the BEG collection provided by Prof. V. Gianinazzi-Pearson (taxon authority nomenclature follows Brummitt & Powell, 1992). Molecular AM fungal community analysis of roots DNA extraction from roots and spores DNA was extracted from the whole root system of an experiment seedling or several randomly sampled root segments from established field plants (total length c. 5 cm). The DNA extraction procedure, involving a modified chloroform extraction–isopropanol precipitation method, was performed as in Heinonsalo et al. (2001). DNA was also extracted from batches of 5–20 spores of control strains precleaned by sonication for 2–3 s twice in sterile water and once in TE buffer (10 mM Tris-HCI pH8.0, 1 mM EDTA). Spores were crushed in 50 µl TE and centrifuged for 5 min at 16 000 g to remove spore debris. Polymerase chain reaction and DGGE An approximately 590 bp fragment of the small subunit ribosomal RNA gene (SSU rDNA) was amplified by polymerase chain reaction (PCR) using the universal eukaryotic primer NS31 (Simon et al., 1992), extended to include a GC-clamp (underlined) (5′-CGCCCGCCGCGCGCGGCGGGCGGGGCGGGGGCACGGGTTGGAGGGCAAGTCTGGTGCC-3′), paired with a more AM fungal-specific fungal primer AM1 (Helgason et al., 1998) designed to exclude plant DNA sequences. All DNA template, primer, and reagent concentrations in the PCR cocktail (total volume 50 µl) were as described by Timonen et al. (1997), with the addition of dried nonfat milk (De Boer et al., 1995; Edwards et al., 1997) to a final concentration of 0.1%. The DNA polymerase used was DyNAzyme II DNA polymerase (Finnzymes OY, Espoo, Finland). Thermocycling was carried out as follows: 3 min at 95°C followed by 30 or 40 cycles of 45 s at 94°C, 1 min at 60°C, 3 min at 72°C, and a final extension for 7 min at 72°C in a PTC-100 thermocycler (MJ Research, Waltham, MA, USA). Negative TE buffer controls were included to check for contamination of


reagents. Amplified SSU rDNA fragments were identified in 1.6% agarose gels (Sambrook et al., 1989). Where required, two or three 50-µl PCR reactions were performed from a DNA template, the products pooled, and DNA concentrated by isopropanol precipitation and ethanol washing as before. DGGE was performed with the DCode Universal Mutation Detection System (Bio-Rad, Hercules, CA, USA). Electrophoretic parameters were as follows: 6% (wt : vol) polyacrylamide gel (37.5 : 1 acrylamide–bis-acrylamide, 1 × TAE buffer (40 mM Tris-acetate, 1 mM EDTA), 1 mm thick, 16 × 16 cm) containing a gradient of a denaturant from 22 to 35%, generated with urea-formamide (Muyzer et al., 1993). Electrophoresis was carried for 4 h at 150 V in 1 × TAE buffer at a constant temperature of 60°C. Gels were stained with ethidium bromide and visualized under UV epifluorescence light using a Fluor-S Imager and Quantity One software (both Bio-Rad). Cloning and sequencing All fragments clearly resolved in DGGE were excised from gels and stored in 50 µl TE buffer at −20°C for further reamplification, cloning, restriction fragment length polymorphism (RFLP) typing and sequencing. DNA in TE buffer was reamplified with primers NS31/ AM1 as before (no nonfat milk added) with a slightly modified thermocycling programme (3 min at 95°C followed by 40 cycles of 45 s at 94°C, 45 s at 60°C, 45 s at 72°C, and a final extension for 10 min at 72°C), and the fragment of expected length (550 bp) purified from low-temperature gelling agarose gel by a modified glassmilk purification method using silica instead of glassmilk (Boyle & Lew, 1995). Purified PCR products, showing a single DGGE band, were ligated into pGEM-T Easy vector and cloned in Escherichia coli JM109 (both Promega, Madison, WI, USA) according to the manufacturers instructions. Four to six putative positive clones were screened for sequence differences by NS31/AM1 amplification and restriction analyses (AluI, HinfI and RsaI). Representative clones of all the different RFLP types detected from roots of single plants were sequenced. Inserts of clones to be sequenced were reanalysed by DGGE under the conditions described. Following this quality check, reconfirmed sequence containing plasmids were isolated using the Wizard Plus Minipreps DNA Purification System (Promega). Sequencing was carried out by cycle sequencing of both strands of the insert with primers T7 and SP6 (A. I. Virtanen Institute, University of Kuopio, Kuopio, Finland, and Institute of Biotechnology, University of Helsinki, Helsinki, Finland). Sequences are lodged in the EMBL database under the accession numbers AJ418855–AJ418900 and AJ496040–AJ496119. Phylogenetic analysis Forward and reverse strands were assembled with the software  (Gene Codes Corporation, Ann Arbor, MI, USA). Sequences were aligned manually using Se-Al

585

586 Research

Sequence Alignment Editor 2.0 (http://evolve.zoo.ox.ac.uk/ software/Se-Al/main.html), taking SSU rRNA secondary structure information into account (Wuyts et al., 2002). Similarity comparisons were performed with the  2.0 algorithm (Altschul et al., 1997). Sequences were screened for possible chimeric origin using the Chimera Check 2.7 algorithm of the Ribosomal Database Project II (RDP-II) internet site (http://rdp.cme.msu.edu). In addition, sequences of 58 glomalean and two outgroup taxa (Endogone pisiformis Link, Mortierella polycephala Coem.) were acquired from GenBank/EMBL databases. The sequence alignment is deposited in the EMBL database (accession number ALIGN_000585). The aligned data matrix included 529 characters. 25 ambiguously aligned nucleotide positions were excluded from further analysis. Phylogenetic analyses were performed with * version 4.0b10 for Macintosh (Swofford, 2002) as follows (1) Maximum parsimony analyses: (a) heuristic search option, random addition of sequences, 100 replicates, tree bisection-reconnection (TBR) swapping, MulTrees on, all characters unordered and of equal weight, gaps treated as missing characters, no more than 50 trees saved in each replicate; (b) heuristic search option, TBR swapping of shortest trees found in previous analysis, all characters unordered and of equal weight, gaps treated as missing characters. A total of 45 100 trees were found that were of the same length as the shortest tree in the first analysis. Not all trees were found because of computer memory constraints. (2) Parsimony bootstrap values were computed over 100 replicates, no more than 500 trees saved in each replicate because of computer-time constraints. (3) Distance analysis: neighbour joining analysis with Kimura 2-parameter substitution model. Arbuscular mycorrhizal fungal community analysis The AM fungal communities in P. patens and P. pratensis root samples were described on the basis of presence/absence of

fungal sequence groups identified in the phylogenetic analysis. Compositional analysis of the fungal communities was performed using multivariate cluster analysis implemented in - ver. 4.0 for Windows (McCune & Mefford, 1999). Similarities in root colonizing AM fungal community composition among samples were identified following application of Ward’s linkage method with Euclidean distance measure.

Results Polymerase chain reaction–DGGE analysis of AM fungal root colonization The DNA from spores of reference AM fungal strains and from plant root samples from pot experiments and field sites, amplified with the primers NS31-GC/AM1, gave PCR products of the expected size (c. 590 bp). No amplification was observed from roots and leaves of pot experiment seedlings of nonmycorrhizal treatments, confirming that the plant DNA was not amplified with the above primer pair (data not shown). Subsequent DGGE analysis was conducted to reveal the presence of similar-length fungal SSU rDNA fragments with divergent sequence homology in mycorrhizal roots. Representative gels yielding banding patterns in the denaturant gradient range of 25–28% are presented in Fig. 2. The DGGE analysis of the spores of known AM fungal species showed that AM fungal genera, but not always species, possessed differential SSU sequence mobility under the electrophoretic conditions used. Among the five species from three genera used as controls, G. mosseae and G. geosporum could not be distinguished on the basis of DGGE banding patterns (Fig. 2a, lanes 2, 3). From each root system of study plants, one to four SSU fragments with distinct mobility in the DGGE were detected. All different AM fungal SSU fragments revealed as detectable bands in the DGGE, 86 in total, were excised from the

Fig. 2 Denaturing gradient gel electrophoresis of small subunit (SSU) rDNA fragments from five glomalean isolates and mycorrhizal roots of Pulsatilla spp. arbuscular mycorrhizas (AM) fungal sequence groups identified (cf. Fig. 3) are indicated. (a) Spores and field plants from six sites (cf. Fig. 1). Lanes: 1, Scutellospora castanea BEG1; 2, Glomus geosporum BEG11; 3, G. mosseae BEG84; 4, G. intraradices IMA6; 5, Acaulospora laevis BEG13; 6, Pulsatilla patens (from site F1); 7, P. pratensis (F1); 8, P. patens (B1); 9, P. pratensis (B1); 10, P. patens (B2); 11, P. patens (F2); 12, P. pratensis (G2). (b) Seedlings from the summer experiment, two seedlings from each treatment. Lanes: 1–2, P. patens (on G1 soil); 3 – 4, P. pratensis (on G1 soil); 5– 6, P. patens (on F1 soil); 7–8, P. pratensis (on F1 soil).


Research Table 2 Percentage of denaturing gradient gel electrophoresis (DGGE) bands yielding multiple restriction fragment length polymorphism (RFLP) types and sequence groups identified after cloning the excised re-amplified bands Number of types / DGGE band

RFLP types Sequence groups

1

2

3

67.82 81.61

20.69 17.24

11.49 1.15

gels, reamplified, cloned, and clones screened for sequence differences by AluI, HinfI and RsaI RFLP typing. The DNA in the majority of bands was easily reamplified, except in the case of two low intensity bands (Fig. 2b, lane 1). Therefore, some potentially rare sequences were not detected. One representative clone of all unique RFLP types in a given root system was sequenced. Several excised DGGE bands yielded clones with different RFLP patterns (Table 2, and see later). A total of 128 clones were sequenced. Two clones produced sequences of poor quality and were excluded from further analysis. Phylogenetic analysis of root AMF sequences Maximum parsimony (MP) and neighbour joining (NJ) analyses of 126 glomalean SSU sequences obtained from plant root samples enabled the recognition of 19 sequence groups as separate clades (Fig. 3). Pairwise sequence similarities within the clades ranged from 97 to 100%. In several cases, identical sequences were detected from different plant individuals. As MP and NJ analyses generated trees with essentially the same topology, only the NJ tree is presented (Fig. 3). The SSU sequences from spores of four reference-species sequenced in this study cluster according to their taxonomic identity with the database sequences. An exception to this is Glomus intraradices isolate IMA6, that groups with Glomus sp. UY1227 (bootstrap support 94%) outside the clade containing G. intraradices, G. fasciculatum and G. vesiculiferum. This indicates a need to revise the taxonomical identity of the isolate. Root AM fungal sequence groups covered three families of Glomeromycota, the Glomaceae (Glomus group A; Schwarzott et al., 2001), Gigasporaceae and Acaulosporaceae (14 putative Glomus spp., two Acaulospora spp., two Scutellospora spp., and a Gigaspora sp.). None of the root sequences position outside Glomeromycota in the phylogenetic analysis. Database sequences representing all taxonomic groupings of Glomeromycota are presented in Fig. 3. Four sequence groups identified in this study showed high similarity to respective sequences of known glomalean species or isolates: MO-G2 to Glomus sp. UY1225 (bootstrap value 73%), MO-G7 to G. hoi (99%), MO-G11 to G. mosseae, although with low support (57%), MO-G10 to G. caledonium


BEG15 (62%, Fig. 3). Six other groups appear to be distantly related to a glomalean species or to a species group but specieslevel affinity remained unresolved: MO-G13 (93%), MO-A1 (100%), MO-A2 (96%), MO-S2 (64%), MO-Gi1 (89%), MO-S1 (< 50%). In both highly supported root Acaulospora sequence groups, and in the MO-S1, the closest reference species appear as sister groups in the MP analysis. The genera Scutellospora and Gigaspora were not resolved in the MP analysis. However, MO-Gi1 SSU sequences show the highest similarity to G. decipiens. Four groups were related to rootderived, but taxonomically unknown, sequences in databases: MO-G6 (bootstrap value 97%; related to Glo7), MO-G1 (77%; Glo21), MO-G4 (81%; Glo18), and MO-G12 (86%; Glo13). There were two further sequence groups that received strong support in the phylogenetic analysis, and appear as novel taxa within the genus Glomus: the MO-G14 (94%) and MO-G8 (94%). Three sequence groups were not resolved in MP analysis but are recognized in the NJ tree: MO-G3, MO-G5, MOG9. The MO-G3 appeared to consist of several clusters with low support in the NJ analysis; one included highly similar root-derived Glo8 sequence, another G. intraradices and G. fasciculatum sequences. This suggests the presence of multiple taxa in this group (Fig. 3). The MO-G5 was not resolved by MP within Glomus group A (Schwarzott et al., 2001). Sequences in the group are, however, united by common RFLP patterns (see below), and cluster together in NJ; rootderived Glo2 sequence has an internal position in the cluster. The MO-G9 remains unresolved within genus Glomus (< 50%). Until additional related sequences become available, that enable a better clarification of identity, we propose its classification as a separate sequence group. Sequence variation within resolved DGGE bands Of the excised DGGE bands, 32.18% contained more than one RFLP type and 18.39% of bands yielded multiple sequence groups (Table 2, and see next section). For example, a band yielding a MO-S1 sequence also contained a MO-S2 sequence that generated a differing RFLP pattern. The two other sequences in the latter group and two sequences within the MO-Gi1 group (Fig. 3b) were also detected simultaneously from single excised DGGE bands. Distinct sequences of the MO-G5 group were commonly detected in single bands (see following section). Sequence groups MO-G3, MO-G13 and MO-G14 showed similar mobility in DGGE, as well as sequence groups MO-G1, MO-G11 and MO-G10. In both cases, the groups appear to be phylogenetically related to each other (Fig. 3). In the sequence alignment, the most variable region of the PCR product was found to lie at 80–300 bp from the 5′ end of the product (excluding primers), corresponding to the variable region at 600–820 bp of the SSU rRNA gene. With few insertions and deletions, mutations were complementary in stem regions, but not so in loop regions, showing that the

587

588 Research

Fig. 3 Neighbour joining tree inferred from nSSU rDNA sequences of all identified arbuscular mycorrhizal (AM) fungi (Glomeromycota taxa) in Pulsatilla spp. roots and corresponding sequences of known and unknown taxa (from field samples), using Endogone pisiformis and Mortierella polycephala as relevant outgroup species. The sequence groups (MO-G2, etc.) identify distinct clusters of sequences with similarity > 97%. Branch lengths correspond to the expected nucleotide substitutions per site. Parsimony bootstrap frequencies > 50% (100 replicates) are shown. Source colour coding: blue, field plants; red and green, experiment pot plants grown on forest F1 and grassland G1 soil inoculum, respectively. Individual sequences codes are identified as follows: (locality) (harvest) (host species). Locality: as in Fig. 1. Harvest: field, field plants; 9 wk, seedlings from summer experiment; 5 wk and 14 wk, seedlings from spring experiment. Host species: R, rare P. patens; C, common P. pratensis. *, isolate sequenced in this study. a–e identify different RFLP types in groups MO-G5 and MO-G6.


Research

Fig. 3 Continued

sequence variation was consistent with the SSU secondary structure. No clearly chimeric sequences were detected. Restriction fragment length polymorphism type analysis In general, RFLP types corresponded well with the phylogenetic analysis. However, three sequence groups were identified that were not recognized by RFLP type data. The groups MOG13, MO-G14 and MO-G6 shared the same RFLP type(s) with MO-G3, MO-G2, and MO-G5, respectively. All three ‘unexpected’ groups received strong support in the phylogenetic analysis (see above, Fig. 3). Several different RFLP types were encountered within specific sequence groups. Four sequences in the clade MO-A1, originating from a single seedling of the common P. pratensis, showed different AluI/HinfI/RsaI restriction patterns, and appear to be relatively distant from each other (97% minimum pairwise similarity). The clade Glomus MO-G5 unites sequences with five different RFLP patterns, of which two were common and were frequently encountered simultaneously in


an excised DGGE band (see Table 2), while the other three occurred once (Fig. 3). The MO-G6 clade, although sharing two of the RFLP types with the above MO-G5, was, however, well distinguished in the phylogenetic analysis (Fig. 3). The AM fungal community structure in roots The composition of AM fungal communities in Pulsatilla roots varied in relation to the origin of the soil inoculum (forest vs grassland) and plant origin (experimental vs field) shown by multivariate analysis of AM fungal sequence groupings presence / absence in any one root sample (Fig. 4). No indication of compositional differences of AM fungal communities between the two Pulsatilla species was identified in the analysis. When pruning the dendrogram at 50% information retained, six clusters can be recognized (Fig. 4). Most field plants appear in a well-defined (a long stem) cluster, except four individuals scattered among other clusters. The pot experiment plants clearly group on the basis of soil origin, three clusters containing grassland soil samples, and two clusters forest soil samples with very few exceptions.

589

590 Research

Fig. 4 Grouping of root arbuscular mycorrhizal (AM) fungal communities in relation to soil inoculum treatment of experimental plants; field plants from six sites are included. Hierarchical cluster analysis by Ward’s linkage method with Euclidean distance was used. Community similarities were calculated based on fungal sequence group presence/absence within root samples. Sample groups defined at 50% information remaining are indicated (‘/’ marks). G1, dry meadow within fragmented agricultural landscape, F1, extensive boreal pine forest area, B1, open sandy area bordered by boreal pine forest, G2, dry meadow, F2, boreal pine forest, B2, roadside bordered by boreal pine forest. Source colour coding: blue, field plants; red and green, experiment pot plants grown on forest F1 and grassland G1 soil inoculum, respectively.

Sequence group MO-G3 was exclusively characteristic of the grassland G1 experimental plants, occurring in 45.5% of (successfully amplified) samples (Table 1). Sequence groups MO-G2 and MO-G5 characterized forest F1 samples (in 58.3% and 62.5% of samples), occurring in grassland samples at lower frequencies. These three sequence groups represent a kind of ‘core group’, since they occurred frequently in both Pulsatilla species in both (summer and spring) pot experiments while the remaining sequence groups occurred at lower frequencies. The sequence group MO-G1 was characteristic for field plant roots, being detected in 75% of field plant samples. Up to six fungal sequence groups were identified in any given Pulsatilla root sample. A total of 14 and 10 sequence groups, respectively, were detected at grassland G1 and forest F1 sites (data sum of pot experiment and field plants). From pot-experiment seedlings of P. patens and P. pratensis, 12 and 15 AM fungal sequence groups were identified, respectively. Six sequence groups were found in the roots of field plants (Table 1).

Discussion Diversity of Pulsatilla root-colonizing AM fungi in Northern boreal soils Here, we show for the first time, in a study of native Estonian Pulsatilla species, that diverse AM fungal communities are

present not only in the AM plant-rich grassland fragments (14 SSU sequence groups) but also in the coniferous forest ecosystem of the Northern boreal zone (10 sequence groups). The detection of relatively high diversity of root-colonizing AM fungi in dry Scots pine forest soils with a predominant ericaceous understorey vegetation provides further evidence for the global distribution and symbiotic activity of these fungi. Similar molecular surveys of boreal forests have mainly focused on ecto- and ericoid mycorrhizal fungal communities (Vrålstad et al., 2000; Kõljalg et al., 2002). However, it is clear that all three major mycorrhizal associations are represented in these woody species-rich habitats. Comparisons with other recent AM fungal surveys provide additional support for underlying AM fungal species richness in the habitats investigated: 13 sequence types have been identified from a temperate broad-leaved forest site, 24 from a temperate grassland, 18–22 from tropical rain forest sites and two to seven from nearby temperate arable fields (Helgason et al., 1998, 1999, 2002; Daniell et al., 2001; Husband et al., 2002a,b; Vandenkoornhuyse et al., 2002b). Probably the greatest known AM fungal richness at a single site has been detected in an old field, from where 37 taxa were isolated as a result of extensive trap-culturing (Bever et al., 2001). Of particular note was the identification of only four out of a total of 19 sequence groups from P. patens and P. pratensis roots that contained sequences of cultured spore producing AM fungi (Fig. 3, G. caledonium, G. hoi, G. mosseae, Glomus sp. UY 1225). This was also a common finding in many of the


Research

other above cited AM fungal molecular diversity studies. A further six SSU groups contained GenBank /EMBL accessioned sequences that have been only found in plant roots and not as spores in the other related surveys. The two dominant sequence groups in experimental seedlings grown in the Scots pine forest soil inoculum, MO-G2 and MO-G5 corresponded to Glo3a (Glomus sp. UY1225) and Glo2, that seem to represent ‘generalist’ fungi since they have been detected in all natural undisturbed ecosystems studied to date, for which comparable data exists (i.e. tropical rain forest, temperate forest, temperate grassland; Helgason et al., 1998, 1999, 2002; Husband et al., 2002a,b; Vandenkoornhuyse et al., 2002b), and not in the arable fields (Daniell et al., 2001). Glomus sp. UY1225 shows a relatively broad host-range with no obvious signs of specificity (Helgason et al., 2002). The MO-G3, dominant in the grassland soil (corresponds to Glo8), has also been detected in tropical and temperate forest and roots of temperate arable field plants (Helgason et al., 1998, 2002; Daniell et al., 2001; Husband et al., 2002a,b). A total of nine novel AM fungal sequence groups from Estonian grassland and boreal forest sites are described in this study. Preliminary attempts to identify respective fungal spores from the soils that yield these distinctive SSU sequences have not been successful, indicating the need for a more intensive survey or alternative spore enrichment culturing. However, taken together, the AM fungal sequence data from this and other related studies cited strongly point to the importance of vegetative fungal propagation via repeated root colonization and nonspore sources of infective propagules (e.g. hyphal, vesicle or infected root fragments) in native ecosystems. Further, all DNA based investigations suggest that the actual diversity of AM fungi in the field is far higher than that currently represented in culture collections as sporulating AM fungi (Helgason et al., 1998, 1999, 2002; Daniell et al., 2001; Kjøller & Rosendahl, 2001; Bidartondo et al., 2002; Husband et al., 2002a,b; Russell et al., 2002; Vandenkoornhuyse et al., 2002b; this study). Molecular detection and identification of AM fungi in Pulsatilla spp. roots In order to provide a comprehensive description of the composition of AM fungal communities associated with the mycorrhizal roots of Pulsatilla spp. seedlings and adult plants, we included a PCR–DGGE based sequence separation/ identification step before cloning and sequencing. The DGGE technique has been frequently used in soil bacterial and fungal community identification through similar rDNA sequence profiling (Björklöf et al., 2003), but has only recently been applied to AM fungal sequences (Kowalchuk et al., 2002). The approach was found to be efficient for the separation of AMF-like sequences down to genus and possibly species level. Nucleotide sequencing of individual DGGE bands and phylogenetic comparisons with sequences of known AM


fungal taxa in databases (Schüßler, 1999; Clapp et al., 2002b) confirmed the inability of PCR-DGGE to differentiate between some close taxa (Kowalchuk et al., 2002). Our results indicate that DGGE banding patterns provide an underestimate of total diversity (Table 2) if used directly for estimation of AM fungal community structure in, for example, large-scale ecological studies. Solutions, such as the use of other, more variable DNA regions to improve the resolution of the method were discussed by Kowalchuk et al. (2002). The presently described sequence groups, identified in the phylogenetic analysis, are partly dependent on the limited sample sizes. Further subdivision or ‘lumping’ of current sequence groups should be expected as more AM fungal SSU sequence data is produced, for example, in the group MO-G3, that includes G. intraradices and G. fasciculatum. Refinement of previous groupings has recently been carried out in other studies (Husband et al., 2002b; Vandenkoornhuyse et al., 2002b). Some level of the diversity could still remain hidden owing to inability of the AM1 primer to amplify the basal groups of Glomeromycota (Daniell et al., 2001). The diversity assessment of AM fungi in planta is further complicated by the fact that distinct AM fungal species boundaries, at the level of SSU sequence homology, may not exist (Schwarzott et al., 2001) and that divergent sequences can occur within individual spores (Clapp et al., 1999; Schüßler et al., 2001). However, until a better understanding of both the genetic organization of AM fungi and their taxonomy is reached, the sequence grouping concept remains the most valid system for delimiting taxa recovered from the field (Clapp et al., 2002a). AM fungal community distribution: habitat vs host Earlier investigations have shown that variation in root colonizing AM fungal communities occurs between study sites and between plant species within the same site, but also where a single plant species is considered in contrasting habitats (Helgason et al., 1998, 1999, 2002; Husband et al., 2002a,b; Vandenkoornhuyse et al., 2002). Our analysis highlighted the site-related differences in the AM fungal root community composition of seedlings of two taxonomically closely related Pulsatilla species. These were grown in soil inocula from a boreal forest site where both target host species coexist, and from a grassland site where only the common species, P. pratensis, is found. The former habitat was characterized by two (MO-G2 and MO-G5) and the latter by a single (MO-G3) abundant AM fungus species in seedlings roots. The fungal groups characteristic to the forest site, however, were also detected at lower frequencies in plants grown on grassland soil inoculum, and are known from other localities world-wide (see above). The two Pulsatilla species growing on the same soil inoculum showed negligible differences in their root AM fungal community composition. Our initial hypothesis that a rare plant species is specifically or

591

592 Research

preferentially associated with AM fungus /fungi was thus not supported. However, the data suggest that under noncompetitive gap conditions, local AM fungal communities are symbiotically compatible with regard to root colonization of both host species. Our initial studies show that the two Pulsatilla species can exhibit differential seed germination, plant establishment and growth and N and P status in these soil inocula (M. Moora et al. unpubl. data). Naturally growing plants from six sites hosted AM fungal communities of lower diversity than pot-experiment plants, a large majority of field plants being colonized by a single sequence group, MO-G1, which was rarely detected in experimental plants. However, owing to limited sampling of the endangered plant species P. patens, for obvious reasons, sample size differences (field vs experimental plants) possibly influenced the fungal community structure detected. Further, the experimental conditions contrasted with the rather stable growth environment of the field plants, with experimental plants in recently disturbed (mixed) soil treatments which could have promoted root colonization by more ruderal AM fungi. There is an increasing body of evidence suggesting that the distribution and abundance of natural plant species is influenced by the composition of local microbial communities. In this respect, it is important to obtain more information from natural ecosystems. Our study showed the occurrence of different dominant AM fungi at sites representing contrasting but typical boreal habitats. The four most abundant AM fungal sequence groups of this study include an undescribed Glomus species and three Glomus sequence groups detected in several other plant species and ecosystems. Even though increasing number of studies on native AM fungi are being published, the data could not always be compared because of the different methodological approaches used. Standardization of methodology, both with regard to ecological parameters (e.g. plant species selection and growth phase) and molecular identification (e.g. primer choice, sequence isolation and cloning via DGGE or direct cloning) is clearly called for that will enable more meaningful conclusions to be drawn in the different plant communities studied. Together with functional studies to uncover the underlying causes of nonrandom AM fungal-host distribution, accumulation of (comparable) quantitative data on native AM fungi present in different ecosystems and growth sites is clearly needed, to move closer to AM fungal ‘geobotany’ (Renker et al., 2004), or identification of particular phyto-sociological units (Read, 2002).

Acknowledgements We thank Professors Manuela Giovannetti (University of Pisa, Italy), Vivienne Gianinazzi-Pearson (UMR INRA, Dijon, France) and Søren Rosendahl (University of Copenhagen,

Denmark) for kindly providing AM fungal spore cultures. The Maj and Tor Nessling Foundation and CIND, Finland and Estonian Science Foundation (grant no. 4579) financed the study; E. Toomiste assisted in running of the summer experiment. Official permission to study rare plants was obtained from Ministry of Environment of the Estonian Republic, no. 21-5/1998/T6/1456. We thank two anonymous referees for their comments on the manuscript.

References Altschul SF, Madden TL, Schaffer AA, Zhang J, Zhang Z, Miller W, Lipman DJ. 1997. Gapped BLAST and PSI-BLAST: a new generation of protein database search programs. Nucleic Acids Research 25: 3389–3402. Bever JD. 2003. Soil community feedbacks and the coexistence of competitors: conceptual frameworks and empirical tests. New Phytologist 157: 465–473. Bever JD, Westover K, Antonovics J. 1997. Incorporating the soil community into plant population dynamics: the utility of the feedback approach. Journal of Ecology 85: 561–573. Bever JD, Schultz PA, Pringle A, Morton JB. 2001. Arbuscular mycorrhizal fungi: more diverse than meets the eye, and the ecological tale why. Bioscience 51: 923–931. Bidartondo MI, Redecker D, Hijri I, Wiemken A, Bruns TD, Domínguez L, Sérsic A, Leake J, Read D. 2002. Epiparasitic plants specialized on arbuscular mycorrhizal fungi. Nature 419: 389–392. Björklöf K, Sen R, Jørgensen KS. 2003. Maintenance and impacts of an inoculated mer/luc-tagged Pseudomonas fluorescens on microbial communities in birch rhizospheres developed on humus and peat. Microbial Ecology 45: 39–52. De Boer SH, Ward LJ, Li X, Chitteranjan S. 1995. Attenuation of PCR inhibition in the presence of plant compounds by addition of BLOTTO. Nucleic Acids Research 23: 2567–2568. Boyle JS, Lew AM. 1995. An inexpensive alternative to glassmilk for DNA purification. Trends in Genetics 11: 8. Brummitt RK, Powell CE, eds. 1992. Authors of plant names. Kew, UK: Royal Botanic Gardens. Clapp JP, Young JPW, Merryweather JW, Fitter AH. 1995. Diversity of fungal symbionts in arbuscular mycorrhizas from a natural community. New Phytologist 130: 259–265. Clapp JP, Young JPW, Fitter AH. 1999. Ribosomal small subunit sequence variation within spores of an arbuscular mycorrhizal fungus, Scutellospora sp. Molecular Ecology 8: 915–921. Clapp JP, Helgason T, Daniell TJ, Young JPW. 2002a. Genetic studies of the structure and diversity of arbuscular mycorrhizal fungal communities. In: van der Heijden MGA, Sanders IR, eds. Mycorrhizal ecology. Berlin, Germany: Springer-Verlag, 201–224. Clapp JP, Rodriguez A, Dodd JC. 2002b. Glomalean rRNA gene diversity – all that glistens is not necessarily glomalean? Mycorrhiza 12: 269–270. Dahlberg A. 2001. Community ecology of ectomycorrhizal fungi: an advancing interdisciplinary field. New Phytologist 150: 555–562. Daniell TJ, Husband R, Fitter AH, Young JPW. 2001. Molecular diversity of arbuscular mycorrhizal fungi colonising arable crops. FEMS Microbiology Ecology 36: 203–209. Edwards SG, Fitter AH, Young JPW. 1997. Quantification of an arbuscular mycorrhizal fungus, Glomus mosseae, within plant roots by competitive polymerase chain reaction. Mycological Research 101: 1440–1444. Eom A-H, Hartnett DC, Wilson GWT. 2000. Host plant species effects on arbuscular mycorrhizal fungal communities in tallgrass prairie. Oecologia 122: 435–444.


Research van der Heijden MGA, Boller T, Wiemken A, Sanders IR. 1998. Different arbuscular mycorrhizal fungal species are potential determinants of plant community structure. Ecology 79: 2082 – 2091. Heinonsalo J, Jørgensen K, Sen R. 2001. Microcosm-based analyses of Scots pine seedling growth, ectomycorrhizal fungal community structure and bacterial carbon utilization profiles in boreal forest humus and underlying alluvial mineral horizons. FEMS Microbiology Ecology 36: 73 –84. Helgason T, Daniell TJ, Husband R, Fitter AH, Young JPW. 1998. Ploughing up the wood-wide web? Nature 394: 431. Helgason T, Fitter AH, Young JPW. 1999. Molecular diversity of arbuscular mycorrhizal fungi colonising Hyacinthoides non-scripta (bluebell) in a seminatural woodland. Molecular Ecology 8: 659 – 666. Helgason T, Merryweather JW, Denison J, Wilson P, Young JPW, Fitter AH. 2002. Selectivity and functional diversity in arbuscular mycorrhizas of co-occurring fungi and plants from a temperate deciduous woodland. Journal of Ecology 90: 371 – 384. Husband R, Herre EA, Turner SL, Galler R, Young JPW. 2002a. Molecular diversity of arbuscular mycorrhizal fungi and patterns of host association over time and space in a tropical forest. Molecular Ecology 11: 2669–2678. Husband R, Herre EA, Young JPW. 2002b. Temporal variation in the arbuscular mycorrhizal communities colonising seedlings in a tropical forest. FEMS Microbiology Ecology 42: 131 – 136. Johnson NC. 1993. Can fertilization of soil select less mutualistic mycorrhizae? Ecological Applications 3: 749 – 757. Kjøller R, Rosendahl S. 2001. Molecular detection of glomalean (arbuscular mycorrhizal) fungi determined as distinct Glomus specific DNA sequences from roots of field grown peas. Mycological Research 105: 1027–1032. Klironomos JN. 2002. Feedback with soil biota contributes to plant rarity and invasiveness in communities. Nature 417: 67 – 70. Kõljalg U, Tammi H, Timonen S, Agerer R, Sen R. 2002. ITS rDNA sequence-based phylogenetic analysis of Tomentellopsis species from boreal and temperate forests and the identification of pink-type ectomycorrhizas. Mycological Progress 1: 81 – 92. Kowalchuk GA, de Souza FA, van Veen JA. 2002. Community analysis of arbuscular mycorrhizal fungi associated with Ammophila arenaria. Dutch coastal sand dunes. Molecular Ecology 11: 571 – 581. McCune B, Mefford MJ. 1999. PC-ORD. Multivariate Analysis of Ecological Data, Version 4. Gleneden Beach, OR, USA: MjM Software Design. McGonigle TP, Fitter AH. 1990. Ecological specificity of vesicular– arbuscular mycorrhizal associations. Mycological Research 94: 120–122. Merryweather J, Fitter A. 1998. The arbuscular mycorrhizal fungi of Hyacinthoides non-scripta. II. Seasonal and spatial patterns of fungal populations. New Phytologist 138: 131 – 142. Muyzer G, De Waal EC, Uitterlinden AG. 1993. Profiling of complex microbial populations by denaturing gradient gel electrophoresis analysis of polymerase chain reaction-amplified genes coding for 16S rRNA. Applied and Environmental Microbiology 59: 695 – 700. Newsham KK, Fitter AH, Watkinson AR. 1995. Arbuscular mycorrhiza protect an annual grass from root pathogenic fungi in the field. Journal of Ecology 83: 991 – 1000. Pilt I, Kukk Ü. 2002. Pulsatilla patens and Pulsatilla pratensis (Ranunculaceae) in Estonia: distribution and ecology. Proceedings of the Estonian Academy of Sciences. Biology. Ecology 51: 242 – 256. Read DJ. 2002. An ecological point of view on arbuscular mycorrhiza research. In: Gianinazzi S, Schüepp H, Barea JM, Haselwandter K, eds.


Mycorrhizal Technology in Agriculture. Basel, Switzerland: Birkhäuser-Verlag, 129–136. Renker C, Zobel M, Öpik M, Allen MF, Allen EB, Vosátka M, Rydlová J, Buscot F. 2004. Structure, dynamics and restoration of plant communities: does arbuscular mycorrhiza matter? In: Temperton V, Nuttle T, Hobbs R, Halle S, eds. Assembly rules and restoration ecology – bridging the gap between theory and practice. Washington DC, USA: Island Press, (In press.) Rosendahl S, Sen R. 1992. Isozyme analysis of mycorrhizal fungi and their mycorrhizas. Methods in Microbiology 24: 167–194. Russell AJ, Bidartondo MI, Butterfield BG. 2002. The root nodules of the Podocarpaceae harbour arbuscular mycorrhizal fungi. New Phytologist 156: 283–295. Sambrook J, Fritsch FE, Maniatis T. 1989. Molecular cloning: a laboratory manual, 2nd edn. Cold Spring Harbor, NY, USA: Cold Spring Harbor Laboratory Press. Sanders IR. 1993. Temporal infectivity and specificity of vesicular-arbuscular mycorrhizas in co-existing grassland species. Oecologia 93: 349–355. Schüßler A. 1999. Glomales SSU rDNA gene diversity. New Phytologist 144: 205–207. Schüßler A, Schwarzott D, Walker C. 2001. A new fungal phylum, the Glomeromycota: phylogeny and evolution. Mycological Research 105: 1413–1421. Schwarzott D, Walker C, Schüßler A. 2001. Glomus, the largest genus of the arbuscular mycorrhizal fungi (Glomales), is nonmonophyletic. Molecular Phylogenetics and Evolution 21: 190–197. Sen R. 2003. Root–microbe–soil interactions and plant production. New Phytologist 157: 391–394. Simon L, Lalonde M, Bruns TD. 1992. Specific amplification of 18S fungal ribosomal genes from VA endomycorrhizal fungi colonising roots. Applied and Environmental Microbiology 58: 291–295. Smith SE, Read DJ. 1997. Mycorrhizal symbiosis, 2nd edn. San Diego, CA, USA: Academic Press. Swofford DL. 2002. PAUP*. phylogenetic analysis using parsimony (*and other methods), Version 4. Sunderland., MA, USA: Sinauer Associates. Timonen S, Tammi H, Sen R. 1997. Characterisation of the host genotype and fungal diversity in Scots pine ectomycorrhiza from natural humus microcosms using isozyme and PCR–RFLP analyses. New Phytologist 135: 313–323. Uotila P. 1996. Decline of Anemone patens (Ranunculaceae) in Finland. Acta Universitatis Upsaliensis, Symbolae Botanicae Upsalienses 31: 205– 210. Vandenkoornhuyse P, Baldauf SL, Leyval C, Straczek J, Young JPW. 2002a. Extensive fungal diversity in plant roots. Science 295: 2051. Vandenkoornhuyse P, Husband R, Daniell TJ, Watson IJ, Duck JM, Fitter AH, Young JPW. 2002b. Arbuscular mycorrhizal community composition associated with two plant species in a grassland ecosystem. Molecular Ecology 11: 1555–1564. Vrålstad T, Fossheim T, Schumacher T. 2000. Piceirhiza bicolorata – the ectomycorrhizal expression of the Hymenoscyphus ericae aggregate. New Phytologist 145: 549–563. Watkinson AR. 1998. The role of the soil community in plant population dynamics. Trends in Ecology and Evolution 13: 171–172. Wuyts J, van de Peer Y, Winkelmans T, De Wachter R. 2002. The European database on small subunit ribosomal RNA. Nucleic Acids Research 30: 183–185. Zobel M, Moora M, Haukioja E. 1997. Plant coexistence in the interactive environment: arbuscular mycorrhiza should not be out of mind. Oikos 78: 202–208.

593