Underground primary succession of ... - Wiley Online Library

Research

Underground primary succession of ectomycorrhizal fungi in a volcanic desert on Mount Fuji Blackwell Publishing Ltd.

Kazuhide Nara, Hironobu Nakaya, Bingyun Wu, Zhihua Zhou and Taizo Hogetsu Asian Natural Environmental Science Center, The University of Tokyo, NishiTokyo, Tokyo 188–0002, Japan

Summary Author for correspondence: Kazuhide Nara Tel: + 81 424 655601 Fax: + 81 424 655601 Email: [email protected] Received: 22 February 2003 Accepted: 13 May 2003 doi: 10.1046/j.1469-8137.2003.00844.x

• Ectomycorrhizal (ECM) fungi are indispensable symbionts for the normal growth of many tree species. Here, we report the underground primary succession of ECM fungi in a volcanic desert on Mt. Fuji, Japan. • We identified all the underground fungal constituents by comparing the fragment lengths of the internal transcribed spacer (ITS) regions in nuclear r-DNA with those of sporocarps, considering intraspecific variation of each species at the research site. ITS sequences were also used for identification. • In total, 21 ECM fungi associated with Salix reinii were identified. Species recorded as sporocarps dominated the underground ECM community. The sere of underground ECM fungi was initiated by one or two of three first-stage fungi, and additional species were recruited with host growth, especially in the soil that developed within a vegetation patch. The species richness of ECM fungi increased significantly with host growth. • The underground ECM community associated with unhealthy hosts differed from that of normally growing hosts. The underground ECM communities and their successional patterns might influence plant growth and plant communities during early primary succession. Key words: ectomycorrhizal fungi, early primary succession, ITS terminal-RFLP, ITS sequence, intraspecific variation, species richness, diversity, community structure. © New Phytologist (2003) 159: 743–756

Introduction A wide variety of ectomycorrhizal (ECM) fungi colonize woody plant roots in many forest ecosystems. ECM fungi can promote the growth of host plants by facilitating nutrient and water uptake, and the normal growth of many tree species depends on ECM fungi in nature (Smith & Read, 1997). Therefore, the ECM fungi community might have a great impact on a plant community. The ECM fungal communities in many forest stands have been studied using sporocarps because many ECM fungi form sporocarps. The ECM sporocarp species have repeatedly been shown to change with the growth of a tree or the development of forest stands (Mason et al., 1983). Therefore, several researchers have proposed a model of ECM fungi succession, which includes the initial appearance of ‘early stage fungi’, and the subsequent appearance of ‘late-stage fungi’ (Last et al., 1984). Unfortunately, most ECM sporocarp

© New Phytologist (2003) 159: 743–756 www.newphytologist.com

succession studies have been conducted in secondary successional sites, where ECM colonization probably occurred from a spore bank or from ECM root tips that were established before disturbance (Baar et al., 1999; Taylor & Bruns, 1999; Grogan et al., 2000). Since there is no initial inoculum in the early stages of primary succession, the ECM fungal community and its succession patterns in primary successional sites might be quite different from those of secondary successional sites. Unfortunately, we know very little about the ECM community and its succession during early primary succession. On the south-east side of Mt. Fuji, the vegetation succession is still in the early stages after the Hoei Eruption in 1707, and vegetation is patchily distributed, forming vegetation islands in a sea of volcanic desert. At this site, we recorded 11 450 sporocarps from 23 ECM species in a 2-yr period, and 22 of the 23 species were associated with willow shrubs of various sizes that differed in age or growth stage (Nara et al., 2003). These sporocarps show a clear successional pattern

743

744 Research

with the host growth stage. First, Inocybe lacera, Laccaria amethystina and Laccaria laccata appear. They are joined by Laccaria murina and Scleroderma bovista. Finally, species of Hebeloma, Cortinarius and Russula are recruited as willow shrubs grow. This report is the first quantitative study of ECM sporocarps and their succession during early primary succession; however, the underground ECM fungal communities and their succession characteristics at this site remain to be studied. Several studies have already targeted underground ECM communities during early primary succession. Helm et al. (1996) studied underground ECM communities associated with several woody species along the successional gradient on deglaciated land. They showed that the ECM morphotype composition differed among successional stages for Populus balsamifera, the most common plant species in all successional stages. A volcanic eruption, which is a more catastrophic activity than deglaciation, creates a vast barren desert for primary succession. Allen et al. (1992) noted that several tree species had formed ectomycorrhizas several years after the last eruption. Yang et al. (1998) studied ECM morphotypes of Larix kaempferi, which are replacing the initial colonizing Salix species on a volcano in Japan. Although L. kaempferi was not the first host for ECM colonization, they showed that morphotype composition was affected by soil properties, such as litter accumulation, pumice particle size, and ash thickness. All these studies used morphotype information. Careful examination of morphotypes can be used to classify ECM root tips to some extent (Agerer, 1987–97). However, ECM classification is markedly affected by soil environment, the developmental stage of the ectomycorrhizas, and also by the investigator’s skill. A morphotype may also include several fungal species. Therefore, the mycobionts in most morphotypedependent studies remain unknown species or groups. Various molecular identification methods have been used to identify the ECM fungal species colonizing roots. Restriction fragment length polymorphism (RFLP) analysis, using the internal transcribed spacer (ITS) region in nuclear r-DNA, has been used the most widely (Horton & Bruns, 2001). Recently, ITS terminal RFLP analysis, using DNA sequencers, has been developed as an improved identification method (Dickie et al., 2002; Zhou & Hogetsu, 2002). Since many fungal species show intraspecific variation in the ITS regions (Kårén et al., 1997), we must examine the intraspecific variation in all the species at each research site to correctly identify underground mycobionts by using ITS regions. Unfortunately, this approach has rarely been applied to study the underground community of ECM fungi. This study investigated the succession of underground ECM fungi in a volcanic desert on Mt. Fuji. Underground ECM fungi were identified using an improved molecular method that considered intra- and interspecies variation of ECM fungi at the study site. We demonstrated the underground primary succession with respect to the development of the host vegetation. The similarity between the underground

ECM community and the above-ground sporocarp community during early primary succession was also discussed.

Materials and Methods Research site The research quadrat (100 × 550 m) was the same as that studied in our previous paper (Nara et al., 2003), and was located at altitudes of 1500–1600 m above sea level on the south-east slope of Mt. Fuji, Japan. The vegetation on this slope was completely destroyed by the Hoei Eruption in 1707, and is now recovering. The vegetation is distributed patchily on the scoria substrate; coverage is now around 5%, and consists mainly of perennial herbs (Adachi et al., 1996). There are three ECM species in the quadrat, S. reinii, Betula ermanii and L. kaempferi, and at the time of the study they covered 502 m2 (< 1% of the area), 0.77m2, and 0.85 m2 in the quadrat, respectively (Nara et al., 2003). Because of the exclusive dominance of S. reinii, we only investigated underground ECM fungal species found with this host species. The coverage of individual S. reinii is a good index of developmental stage and is generally related to the age after first colonization at this site (Lian et al., 2003). The coverage in the quadrat ranged from 0.016 to 154 m2. The growth parameters of individuals, such as leaf biomass, nutrient concentration, and photosynthetic activity, varied greatly in some cases (Nara et al., 2003). Other details of our research site are described in Nara et al. (2003). Sporocarps used as reference patterns for molecular identification To accurately identify the underground ECM fungal species at our research site by molecular analysis using ITS regions, we needed to know the intra- and interspecific variation in the ITS regions of ECM fungi at the site. To comprehend the intraspecific variation within the research quadrat, we used sporocarps of each species from different vegetation patches, selecting from collections made in another study (Nara et al., 2003). Because some species produced sporocarps in only one or a few patches, only limited numbers of sporocarps were examined for these species. We also used the sporocarps of 16 nonmycorrhizal fungi found in the study area, including species of Agaricus, Camarophyllus, Coprinus, Cystoderma, Entoloma, Galerina, Hygrocybe, Lycoperdon, Marasmius, Melanoleuca, Mycena and Omphalina. Each sporocarp was desiccated under vacuum using a freeze drier (FDU540, EYELA, Tokyo, Japan) and kept at room temperature in a sealed plastic bag with silica gel. A small piece (approx. 1 mm3) of each dried sporocarp was placed in a 2.0-ml tube containing a zirconia ball (5 mm in diameter) for DNA extraction. In total, DNA was extracted from 255 sporocarps of 39 species.

www.newphytologist.com © New Phytologist (2003) 159: 743–756

Research

Fig. 1 Morphotypes of ectomycorrhizas formed on Salix reinii during early primary succession. NM indicates nonmycorrhizal root tips. The 10 other ectomycorrhizal morphotypes were classified using morphological characters, including surface texture, color, emanating hyphae, and rhizomorph.

ECM root tips for molecular analysis In early November 2001, three soil samples (10 × 10 × 10 cm) were collected randomly from the periphery of each of four small (< 0.5 m2), four middle-sized (2–10 m2), and four large (> 45 m2) hosts. To compare the underground ECM communities in different positions within a patch, we also collected three soil samples randomly from the outside bare ground (> 1 m outside the area of host coverage), and three from the inside ground (> 1 m inside the edge of host coverage), of each large host. We also collected three soil samples randomly from each of four unhealthy hosts in the middle-sized class; these showed less growth, lower nutrient concentrations, and reduced photosynthetic activity, as described previously (Nara et al., 2003). All the roots contained in each soil sample were carefully washed. Roots of nontarget plants were excluded. The root tips were excised from the main roots as each tip was separated from the others. They were spread on a transparent plastic box under which a 5-mm mesh was placed. We selected up to 200 individual root tips from randomized grids on the mesh. The tips were observed under a dissecting microscope to determine the percentage of ECM root tips, and each ECM tip was classified into one of 10 morphotypes (Fig. 1), based on their surface color, texture, emanating hyphae, and rhizomorphs according to Agerer et al. (1987–97). We randomly picked 10 replicate ECM root tips from each morphotype in each soil


sample. Individual root tips were placed in 2.0-ml tubes containing a zirconia ball for DNA extraction after desiccation. When a morphotype contained fewer than 10 ECM root tips in each soil sample, we analyzed all the root tips as replicates. In total, 14 400 root tips from 72 soil samples were examined under a dissecting microscope, and 1242 ECM root tips were used for the molecular identification. DNA preparation and ITS fragment analysis Each sample was pulverized in a 2.0-ml tube containing a zirconia ball using a homogenizer (FastPrep, Funakoshi Co., Tokyo, Japan) for 20 s. We added 350 µl CTAB solution (2% cetyltrimethylammonium bromide, 100 m Tris-HCl (pH 8.0), 20 m EDTA (pH 8.0), 1.4  NaCl, 0.5% βmercaptoethanol) to the tube and homogenized it again for 20 s. After incubation in a block heater at 65°C for 1 h, 350 µl chloroform: isoamyl alcohol mixture (24 : 1) was added to the tube. After the tube had been vortexed and centrifuged (20 000 g, 7 min, room temperature), the supernatant was transferred to another 1.5-ml tube. The DNA was precipitated by adding an equal volume of isopropyl alcohol and keeping the tube at −30°C for 15 min. After centrifugation (3300 g, 10 min, 4°C), the DNA pellet was washed with ethanol (80%) and dried. The DNA pellet was dissolved in 100 µl TE buffer (10 m Tris-HCl (pH 8.0), 1 m EDTA), and stored at −30°C until use.

745

746 Research

PCR amplification was conducted using AmpliTaq Gold (Applied Biosystems, CA, USA) following the manufacturer’s instructions. For the ITS fragment length analysis (ITS3–4), template DNA was amplified by PCR using primers ITS3 and ITS4 (White et al., 1990; Gardes & Bruns, 1993) (Ta = 51°C); primer ITS4 was labelled with Texas Red. The PCR products were diluted 15 times with TE buffer and used for fragment size analysis. For a second ITS fragment length analysis (ITS1F-HinfI), template DNA was amplified by PCR using primers ITS1F and ITS4 (Ta = 51°C); primer ITS1F was labelled with Texas Red. The PCR products (2 µl) were digested with 8 µl HinfI solution (1061 A, Takara Shuzo Co., Shiga, Japan) at 37°C for 8 h, in accordance with the manufacturer’s instructions. The restriction fragments were diluted twice with TE buffer and used for fragment length analysis. Both of the diluted solutions used for fragment length analysis were electrophoresed in a DNA sequencer (SQ-5500E, Hitachi Electronics Engineering Co., Tokyo, Japan) after denaturation (95°C, 5 min) with the loading dye (Zhou & Hogetsu, 2002). DNA size standards (RPN2446, Amersham Pharmacia Biotech UK Ltd, Buckinghamshire, UK) were loaded every twelfth lane on the gel. The length of each fragment was estimated from the size standards using FRAGLYS 3.0 software (Hitachi Electronics Engineering Co.). Identification of underground ECM fungi Identification of ECM fungal species was based on careful morphological analysis of sporocarps, as described in Nara et al. (2003), and the nomenclature in that paper is also used in this study. To identify the underground ECM fungal species, the ITS fragments were compared with those of sporocarps. If the lengths of ITS3−4 and ITS1F-HinfI from a single ECM root tip were within the range of intraspecific variation of a given sporocarp species, we concluded that the fungal constituent on the ECM root tip belonged to that species. The ITS sequences of unmatched ITS fragment types of ECM roots that differed from all sporocarp species were compared with the sequences of known species in the DDBJ/ EMBL/GenBank database using the FASTA homology test. Morphological characters of ECM tips were used as additional information to identify the unmatched ITS types following Agerer et al. (1987–97). ITS sequence analysis Several root tips from each of the unmatched ECM types were used for sequencing. The PCR products amplified using primers ITS1 and ITS4 were subcloned with pT7Blue Perfectly Blunt Cloning Kits (Novagen Inc., WI, USA) following the manufacturer’s instructions. Plasmid DNA was extracted from transformed cells suspended in 50 µl of sterile

water in a 1.5-ml tube that was incubated in boiling water for 5 min. The supernatant was PCR amplified with primers Texas Red M13F and M13R (RPN 2337 and RPN 2338, Amersham International plc., Buckinghamshire, England). After confirming fragment insertion on agarose gels, we sequenced the insert using Thermo Sequenase Pre-mixed Cycle Sequencing Kits (RPN 2444, Amersham International plc.) using primers Texas Red M13F and T7 (US70328, Amersham International plc.) following the manufacturer’s instructions. The sequences obtained, including the complete ITS regions, were registered in DDBJ. Quantification of the underground ECM community The ratio of each species in 10 replicates used for the molecular analysis from each morphotype in each soil sample was multiplied by the number of root tips included in the morphotype. The obtained numbers of root tips colonized with the same species were summed within the soil sample. The relative abundance of the fungus in a soil sample was defined as the percentage of that fungus in the total ECM root tips from the soil sample. The relative abundance in three replicates from a host was averaged and used as the relative abundance in a host. Similarly, the relative abundances of 12 replicate soil samples from four hosts in a category were averaged for the relative abundance in a category. The species richness of underground ECM fungi was evaluated using the number of fungal species identified within each soil sample, host, and category. Statistics All the statistical analyses were conducted with SPSS 11.5 for Windows. One-way  followed by Tukey’s HSD test was used to compare the percentage of ECM colonization among six sampling categories with P < 0.05 as the level of significance. Using the same method, we also tested the difference in ECM species richness, among the three host size classes (small, middle, and large) or the three spatial positions (outside, periphery, and inside) from large hosts. We created a table in which host size classes and fungal taxa were set in the rows and columns, respectively, and filled each cell with the relative abundance for that category. Another table, with the three spatial positions (outside, periphery, and inside) in the rows and fungal taxa in the columns, was also filled with the relative abundance for that category. We also created a table in which health (normally growing and unhealthy) was set in the rows and fungal taxa were set in the columns. For each of the three tables, the exact Pearson chisquare test was used to test the independence of the row and column classification, in which the exact P-value was estimated within the 99% confidence interval based on a crude Monte Carlo sample of 10 000 tables from the reference set (Mehta & Patel, 1996).


Research

Table 1 Intra- and interspecific variation in the lengths of two ITS fragments for ectomycorrhizal fungi that appeared during early primary succession on Mt. Fuji Species

Genets

ITS3 – 4 (bp)

ITS1F-HinfI (bp)

Sporocarps Boletus cf. rubellus Krombh.a Boletus pulverulentus Opat. Cortinarius alboviolaceus (Pers. Fr.) Fr. Cortinarius decipiens (Pers. Fr.) Fr.a Hebeloma leucosarx Ortona Hebeloma mesophaeum (Pers.) Quél.a Hebeloma pusillum Lange Inocybe acuta Boud. Inocybe calospora Quél. Inocybe dulcamara (Pers.) Kumm.a Inocybe lacera (Fr.) Kumm.a Inocybe fastigiata (Schaeff.) Quél. Inocybe sp. 1 Inocybe sp. 2a Laccaria amethystina Cookea Laccaria laccata (Scop. Fr.) Berk. & Br.a Laccaria murina Imaia Russula norvegica Reid Russula pectinatoides Peck Russula sororia (Fr.) Romella Scleroderma bovista Fr.a

8 1 3 18 14 21 4 1 5 6 20 1 6 12 20 19 22 10 5 8 20

671–673 453 384 367–372 402 –403* 400 –403* 396 –400* 394 375 –376* 402 –403 395 –396 370 368 386 399 –404* 393 –394 393 448 415 413 431–436*

269 –273 368 399 321–323 397 401–403 397–399 161 368 424 384 –386 395 –397* 159 400 385 –386 389 –391 386 –387 351–353 372–373 373 –374 285 –287*

332–334 380 –381 354 386 407–410 410 –411 407–408 343 –345 381

Not amplified 349 –351 368 327 360 –365 356 –358 180 –182 177–178 182

Species only found underground Cenococcum geophilum (> 99%) Laccaria sp. (> 90%) Pezizales sp. (> 90%) Sebacina sp. (> 85%) Thelephoraceae sp. (> 95%) Tomentella sp. 1 (> 95%) Tomentella sp. 2 (> 90%) Unidentified D morphotype sp. 1 Unidentified I morphotype sp. 1

Acc. No.

AB089815 AB096868 AB096869 AB096870 AB089959 AB089960 AB096871 AB089817 AB096872

The genets column contains the number of examined sporocarps that were collected from different vegetation patches, belonging to different genets. Column ITS3 – 4 gives the lengths of the PCR products amplified by primers ITS3 and ITS4. Column ITS1F-HinfI lists the terminal (ITS1F side) restriction fragment lengths of the PCR products amplified by primers ITS1F and ITS4. Sequences, including the complete ITS regions of the species found only underground, are available in the DDBJ/EMBL/GenBank database and the Accession Numbers are listed. Their similarities with known species in close matches are shown in parentheses following the species name. aSpecies found in both the sporocarp and underground ectomycorrhizal communities. *Two bands that differ in size by a few base pairs were detected from individual sporocarps in some cases.

Results Intra- and interspecific variation in ITS fragment lengths of sporocarps Of the ECM fungal species associated with Salix reinii, only Cantharellus cibarius was not amplified by PCR. All the other ECM fungal species were amplified, and single bands were typically detected from individual sporocarps in the ITS3 – 4 and ITS1F-HinfI fragment length analyses (Table 1). However, many individuals of six species, including Hebeloma mesophaeum and Scleroderma bovista, showed double bands in the ITS3 – 4 fragment length analysis that differed from each other by a few base pairs (Table 1). In the ITS1F-HinfI fragment


length analysis, Inocybe fastigiata and S. bovista showed this kind of double bands. The double bands were always reproducible in replicate experiments using the same samples, and did not interfere with the separation of species. The estimated lengths of the two ITS fragments were relative to the size standards, and were sometimes inconsistent with the sequence data by a few base pairs. However, the estimated figures were highly reproducible in replicate experiments using the same samples. The lengths of ITS3–4 and ITS1F-HinfI ranged from 367 to 673 and from 159 to 424 base pairs, respectively. Most of the investigated species showed intraspecific variation in the fragment length, and most of this variation was restricted to a few base pairs (Table 1).

747

748 Research

Fig. 2 DNA sequencer gel electrophoresis of ectomycorrhizal ITS-PCR-products amplified by primers ITS3 and ITS4. M, molecular size standard; lanes 1–6, L morphotype (Laccaria laccata); lanes 7–11, B morphotype (Tomentella sp. 1); lanes 12–16, Sb morphotype (Scleroderma bovista); lanes 17–22, L morphotype (L. laccata in lanes 17, 18, 20–22, and Laccaria amethystina in lane 19).

The interspecific variation was greater than the intraspecific variation in most cases. However, one of the two fragment length analyses alone was insufficient to separate all the species (e.g. Hebeloma spp. and Laccaria amethystina were inseparable by ITS3 –4 length alone). By combining the ITS3–4 and ITS1F-HinfI lengths, all the ECM sporocarps were separated into species (Table 1). Separation was still possible when the 16 nonmycorrhizal fungus species were included (data not shown). Identification of underground ECM fungi The microscopic examination of 14 400 root tips collected from 72 soil samples found ectomycorrhizas on 59.2% of them. A comparison of the ITS fragment lengths of sporocarps and ECM root tips showed that the H, I, L, and Sb morphotypes consisted mostly of Hebeloma spp., Inocybe lacera, Laccaria spp., and S. bovista, respectively (Fig. 2). In some soil samples, however, these morphotypes also contained ECM root tips that did not match any species recorded as sporocarps. Moreover, several species of Laccaria and Hebeloma were included in morphotypes L and H, respectively, from many soil samples (Fig. 2). The L morphotype also contained I. lacera in some soil samples. Therefore, the constituent

fungal species and their ratios within a morphotype varied greatly between soil samples. Most ECM root tips in other morphotypes, such as B, Cg, D, and Y, did not match any species recorded as sporocarps. These unmatched ECM root tips were clearly separated into nine groups by the molecular analyses, some of which showed intragroup variation in fragment length of several base pairs, as seen in some of the species recorded as sporocarps (Table 1). One group was identified as Cenococcum geophilum due to its high ITS sequence similarity (> 99%) with C. geophilum isolates registered in the database, in addition to the morphological characters of its ectomycorrhizas. Another six groups showed high similarity (> 85%) with known fungi and were regarded as related species within the genus of their closest match (Table 1). The remaining two groups had no close match, and we considered them unknown species: an unidentified I morphotype species 1 (UN-I1) and an unidentified D morphotype species 1 (UN-D1). In total, 21 species were identified in the underground ECM community, including 12 out of the 23 fungal species recorded as sporocarps at this site. To evaluate the proportion of the detected species to the actually existing species in this site, we created a figure in which examined ECM root tips were randomly arranged on the x-axis and the cumulative number of fungal species detected were ploted (Fig. 3). In all


Research

Fig. 3 Number of detected fungal species in relation to the number of examined ectomycorrhizal (ECM) root tips in each of the six sampling categories during early primary succession on Mt. Fuji. Root tips were sampled from the periphery positions of small (< 0.5 m2 in coverage area, square), middle-sized (2–10 m2, triangles), and large (> 45 m2, white circles) host Salix reinii. We also sampled root tips from the inside positions (black circle) and outside positions (gray circle) of the large hosts. There were unhealthy hosts in the middlesized host class, and we collected the samples from the periphery positions of the unhealthy hosts (inverted triangle). Each sampling category contained 12 sampling positions from four individual hosts, and 200 root tips were sampled from each position. Examined ECM root tips in each category were randomly arranged on the X-axis. The Y-axis indicates the cumulative number of fungal species detected.

the sampling categories, the number of fungal species detected leveled off with increasing numbers of examined ECM root tips. In total, 5, 8, and 13 species were detected in the periphery of small, middle-sized, and large hosts, respectively (Fig. 3). In the outside and inside positions relative to large hosts, we detected 7 and 16 species, respectively, and 13 species were found in the periphery positions for unhealthy hosts (Fig. 3). In total, 88% of the ECM root tips were colonized by the species whose sporocarps were recorded in the quadrat. The percentages were high under the small (99%) and middlesized (98%) hosts, and outside the large hosts (99%), and were lower under unhealthy hosts (75%), in the periphery (82%), and inside the large hosts (80%). Underground succession of ECM fungi The underground ECM communities associated with small hosts (coverage < 0.5 m2) were simple, and consisted almost exclusively of three of the species recorded as sporocarps. One of I. lacera, Laccaria laccata and L. amethystina (Fig. 4a) was dominant in each host. These three species were defined as the first-stage fungi in underground ECM communities during early primary succession on Mt. Fuji.


For the middle-sized hosts (2–10 m2), the first-stage fungi were still the major fungal constituents (Fig. 4b). In addition, S. bovista and Laccaria murina were associated with all of these hosts, and seven of 12 soil samples contained these two species (Fig. 4b), which were defined as the second-stage fungi in the underground ECM sere. With further growth of the host vegetation, many additional species appeared in the underground ECM fungal community (Fig. 4e). The first- and second-stage fungi were still major components of the communities associated with large hosts (> 45 m2). In addition, UN-D1 was observed in the communities of all the large hosts, and seven of 12 soil samples contained this species as one of the major constituents. Inocybe sp. 2, C. geophilum, and species of Cortinarius and Hebeloma also appeared, but were quite rare. The species composition of the underground ECM fungi for the three host size classes differed significantly by the exact Pearson chi-square test (P < 0.001) because of the successional recruitment of new ECM fungi with increasing host size. The number of species of ECM fungi in a soil sample was significantly higher from middle-sized (2.6 ± 0.3) and large hosts (3.3 ± 0.4) than from small hosts (1.4 ± 0.1) (Table 2, P < 0.05). The number of species of ECM fungi in a host also increased significantly from the small hosts (2.0 ± 0.4) to the middle-sized (5.5 ± 0.3) and large hosts (7.8 ± 0.8) (Table 2, P < 0.05). Underground spatial distribution of ECM fungi associated with large hosts We compared the ECM fungal communities between outside (Fig. 4d) and inside (Fig. 4f ) positions for four large hosts. S. bovista was contained in nine of 12 soil samples from outside positions, and was the most frequent species in outside positions. Four soil samples contained this species as the only fungal constituent. The relative abundance of S. bovista in a soil sample was 46.6 ± 12.3% from outside positions, and was significantly higher than the abundance from inside positions (0.0 ± 0.0%) by the t-test (P < 0.001). L. murina was contained in five soil samples from the outside (Fig. 4d), and no soil samples from the inside (Fig. 4f ). The relative abundance of this fungus in a soil sample was also significantly higher outside (24.5 ± 9.9%) than inside (0.0 ± 0.0%) by the t-test (P < 0.021). By contrast, Cortinarius decipiens and Hebeloma species were found in five soil samples from inside and were never found from outside. Although the difference in the relative abundance of Hebeloma spp. in a soil sample between the inside and outside positions was not significant (P = 0.092 > 0.05), the abundance of C. decipiens in a soil sample was significantly higher inside (9.0 ± 3.6%) than outside (0.0 ± 0.0%) by the t-test (P < 0.019). Species of Thelephoraceae, including Tomentella spp., were also more frequent inside (Fig. 4f ) than outside (Fig. 4d). Eleven of 12 soil samples from the inside contained some Thelephoraceae species, and their total abundance in soil samples was

749

Fig. 4 The underground ectomycorrhizal (ECM) fungus community in relation to the developmental stages of Salix reinii, the growth activity of the host, and the spatial position within a vegetation patch. The figures show the fungal composition at the periphery of (a) small (b) middle-sized, and (c) unhealthy middle-sized hosts, and (d) outside (e) on the periphery, and (f) inside large hosts. The host coverage area in each vegetation patch is shown after the patch number. The four hosts investigated in each category are illustrated in different colours in each figure: red, blue, yellow, and green. The figures on the bars give the number of soil samples containing each fungus out of the 12 soil samples in each sampling category. Abbreviations of the ECM fungal species on the X-axes are Ll, Laccaria laccata; Il, Inocybe lacera; La, Laccaria amethystina; Lm, Laccaria murina; Sb, Scleroderma bovista; Id, Inocybe dulcamara; I2, Inocybe sp. 2; Cd, Cortinarius decipiens; Hl, Hebeloma leucosarx; Hm, Hebeloma mesophaeum; Rs, Russula sororia; Br, Boletus cf. rubellus; L, Laccaria sp.; Cg, Cenococcum geophilum; P, Pezizales sp.; S, Sebacina sp.; T1, Tomentella sp. 1; T2, Tomentella sp. 2; T, Thelephoraceae sp.; UI, Unidentified I morphotype sp. 1; UD, Unidentified D morphotype sp. 1. The Y-axes indicate the relative ECM abundance for a host. The bars on orange, pale blue, and white backgrounds indicate first-, second-, and late-stage fungi, respectively, in the sere of ECM fungal succession. The bars on gray backgrounds indicate species whose sporocarps have not been confirmed in weekly sporocarp surveys conducted over two years.

750 Research


Research

Table 2 Ectomycorrhizal (ECM) colonization and the number of ECM species associated with different sizes of the host Salix reinii, different growth activities of the host, and different spatial positions within the large hosts during early primary succession Number of ECM fungal species Category* (host, position)

Percentage of ECM colonization

Per soil sample

Per host

Per category

Small, periphery Middle-sized, periphery Large, periphery Large, outside Large, inside Unhealthy, periphery

72.5 ± 6.2a 57.8 ± 6.7ab 62.5 ± 3.9ab 52.4 ± 6.2ab 67.1 ± 4.8ab 42.4 ± 7.3b

1.4 ± 0.1a 2.6 ± 0.3b 3.3 ± 0.4bA 1.9 ± 0.3B. 3.7 ± 0.5A. 2.7 ± 0.3ns

2.0 ± 0.4a 5.5 ± 0.3b 7.8 ± 0.8c† 3.8 ± 1.0† 7.8 ± 1.3† 5.5 ± 1.2ns

5 8 13 7 16 13

Except for the last column, the figures show the mean ± S.E.M. Figures followed by different letters in the column for the percentage of ECM colonization indicate a significant difference by Tukey’s HSD test (P < 0.05). Figures followed by different lower-case letters in the columns for number of ECM fungal species per soil sample and per host, indicate a significant difference among the three host size classes (small, middle, large), and figures followed by different capital letters indicate a significant difference among the three spatial positions (outside, periphery, inside) for large hosts by Tukey’s HSD test (P < 0.05). *We collected three soil samples from the periphery of each of four small hosts (< 0.5 m2 in coverage area), four middle-sized hosts (2–10 m2), and four large hosts (> 45 m2). For each large host, three soil samples from inside positions and three from outside positions were collected. There were unhealthy hosts in the middle-sized host class, and we collected three soil samples from the periphery of each of the four unhealthy hosts. †The number of ECM fungal species per host was significantly different among the three spatial positions by ANOVA (P = 0.034), but was not separable using Tukey’s HSD test. nsFigures for ECM fungal species from unhealthy middle-sized hosts, per soil sample and per host, were not significantly different from the figures from normally growing middle-sized hosts by the t-test (P > 0.05).

20.5 ± 7.3%, which was significantly higher than outside (1.7 ± 1.2%) by the t-test (P < 0.018). As a whole, the ECM species composition differed significantly among the three spatial positions, outside, periphery, and inside, by the exact Pearson chi-square test (P < 0.001). The number of species in a soil sample from the outside (1.9 ± 0.3) was significantly smaller than from inside (3.7 ± 0.5) or on the periphery (3.3 ± 0.4) (Table 2, P < 0.05). The number of species in a host differed significantly inside (7.8 ± 1.3), on the periphery (7.8 ± 0.8), and outside (3.8 ± 1.0) by  (Table 2, P < 0.034). Underground ECM fungi associated with unhealthy hosts Of the 13 ECM fungal species associated with unhealthy middle-sized hosts, L. laccata and L. murina were frequent species, and were found in eight and seven out of 12 soil samples, respectively, and these two fungi were the major components based on relative abundance (Fig. 4c). The total of the relative abundances of C. geophilum, Sebacina sp., and all the species of Thelephoraceae in an unhealthy host was 31.4 ± 9.1%, and was significantly higher than 1.7 ± 0.9% in a normally growing similar-sized host by the t-test (P < 0.024). By contrast, I. lacera and S. bovista were not found in any soil sample from unhealthy hosts (Fig. 4c), while they were found in four and seven out of 12 soil samples from normally growing hosts, respectively (Fig. 4b). The relative abundances of I. lacera and S. bovista for unhealthy hosts were both 0.0 ± 0.0%, and were significantly different from the abundance of I. lacera (25.0 ± 11.2%, P < 0.036) and S.


bovista (19.7 ± 8.6%, P < 0.033) for normally growing similarsized hosts. The number of fungal species in a soil sample (2.7 ± 0.3) or host (5.5 ± 1.2) for unhealthy hosts was not significantly different from that for normally growing similarsized hosts (Table 2). Community structure of ECM fungi in different host size classes or different spatial positions The rank-abundance relationships of ECM fungi among the three host size classes are shown in Fig. 5(a). Associated with small hosts the most abundant fungal constituent prominently dominated the underground ECM community. With increasing host size, the number of species increased and the dominance of abundant species tended to be reduced. Consequently, the structure of the underground community changed with host growth, even when the peripheral positions were compared. In addition, the underground ECM fungal community structures also differed substantially among the different positions of large hosts. The ECM fungal community in outside positions was dominated by a few abundant species, and shifted to a low-dominant structure that included more species in a rather evenly distributed abundance along the direction toward the inside position (Fig. 5b). Species composition in the sporocarp and underground ECM communities associated with individual hosts The constituent species and their abundance ranks in underground ECM communities from individual hosts are

751

752 Research

shown in Table 3. To compare the underground ECM fungal community with the sporocarp community in each patch, we also showed the sporocarp species and their abundance ranks, in which the data from a prior study were used (Nara et al., 2003). The top-ranked species in the underground ECM community from every small host was the same as the topranked species in the sporocarp community. In other host size classes, the top-ranked species in the sporocarp community in each patch were included in the top five ranks of its underground community in all cases, and the top-ranked species in the underground community were included in the top five ranks in the sporocarp community in all cases except one.

Discussion Methodology

Fig. 5 Comparison of the structure of the underground ectomycorrhizal community (a) for different size classes of the host Salix reinii, and (b) for different spatial positions in large hosts in a volcanic desert on Mt. Fuji. (a) ECM communities in the periphery positions of small (< 0.5 m2 in coverage area), middle-sized (2–10 m2), and large (> 45 m2) hosts were compared. (b) ECM communities in the inside positions (> 1 m inside the edge of host coverage), periphery, and outside positions (> 1 m outside the area of host coverage) of the large hosts were compared. After ECM relative abundances were ranked in individual hosts, the abundances of the same rank were averaged among four hosts within a sampling category. The averaged abundance of each rank (Y-axes, logarithmically shown) was plotted against its rank (X-axes, in order of the rank) in each sampling category.

Sampling strategy greatly affects the results of underground ECM community studies (Taylor, 2002). A simple sampling method using a small soil core without subsampling is usually used for underground ECM community studies (Horton & Bruns, 2001). The number of root tips in a core usually varies between different stands or different treatments. Taylor (2002) suggested that this variation could cause misunderstanding of the species richness and community structure of ECM fungi, especially when the sampling effort was insufficient. Our sampling method using larger soil cubes (10 × 10 × 10 cm) and subsequent subsampling of an even number of root tips (200 tips) should reduce the influence of the variation in the number of root tips per sample, and would be suitable to correctly perceive the ECM community in heterogeneous habitats in our early successional site. Moreover, the cumulative number of ECM species vs the number of sampled ECM root tips leveled off in all sampling categories (Fig. 3). This indicates that we sampled sufficient numbers of ECM root tips to detect most of the species present in each sampling category. To our knowledge, such sufficiency has never been demonstrated in any study. ITS polymorphism has been used extensively to identify underground ECM fungi, because of the high variation among species. PCR-RFLP of the ITS regions has been most widely used (Gardes & Bruns, 1996; Dahlberg et al., 1997; Kårén & Nylund, 1997; Pritsch et al., 1997; Jonsson et al., 1999a,b; Horton & Bruns, 2001). Recently, ITS terminalRFLP has been used to differentiate bacterial communities (Liu et al., 1997; Moeseneder et al., 1999), and has been used to identify fungal species in ECM root tips (Zhou & Hogetsu, 2002) or ECM hyphae in soil (Dickie et al., 2002). Since this method includes electrophoresis in a DNA sequencer, which can detect a one-base-pair length difference, its resolution is far higher than the agarose gels usually used for RFLP analysis. Our ITS fragment length analysis is fundamentally identical to ITS terminal-RFLP analysis. In their ITS terminal-RFLP


Research

Table 3 Species composition in the sporocarp and underground ectomycorrhizal (ECM) communities associated with individual hosts, Salix reinii, in an early successional volcanic desert Sporocarp abundance rank*

ECM abundance rank†

Patch No.

Host size (m2)

1

Small hosts 8 125 41 127

0.03 0.15 0.21 0.47

Ll Ll Il La

Middle-sized hosts 131 61 120 142

2.83 3.87 4.31 6.58

Sb Sb Sb Il

Ll Ll Ll Sb

Hp Hm La Lm

Il Lm Il La

Ll Ll Ll Ll

La La Sb La

Hm Hm Hm Hm

Lm Sb La Cd

Large hosts 89 83 90 139

48.45 50.23 59.84 153.73

2

3

4

5

1

2

3

Ll Ll Il La

La T1 Lm

Ll

Lm La Lm I2

Il La Lm Il

Lm Sb La Lm

Sb Il Il Sb

La Il La Lm

Ll Ll Lm T2

La

4

5

Sb Ll Il Sb

L Lm Sb Ll

T1 T1 Ll La

Sb Sb Ud Sb

Lm La Sb Il

Ud Lm Ll Ll

Abbreviations of the ECM fungal species are Cd, Cortinarius decipiens; Hm, Hebeloma mesophaeum; Hp, Hebeloma pusillum; Il, Inocybe lacera; I2, Inocybe sp. 2; L, Laccaria sp.; La, Laccaria amethystina; Ll, Laccaria laccata; Lm, Laccaria murina; Sb, Scleroderma bovista; T1, Tomentella sp. 1; Ud, Unidentified D morphotype sp. 1. *The data for sporocarps were from our prior study (Nara et al., 2003). †The ECM abundance rank was based on the relative abundance of ECM root tips in three soil samples collected from each of the small and middle-sized hosts, and in 12 soil samples collected from each large host.

analysis, Zhou and Hogetsu (2002) included several ECM root tips in a tube for DNA extraction, resulting in several bands. Although this method is effective for identifying existing fungal species quickly, one band can represent one or several root tips. To accurately calculate the ECM root tip abundance of each fungus, it is desirable that one band always represents one root tip. Therefore, we extracted DNA from single ECM root tips, resulting in one clear band per sample in our fragment length analyses (Fig. 2). Single band analysis is also effective for reducing scoring errors due to detecting obscure bands and for discarding contaminated samples that were often included in ECM samples from developed forests. In addition to these technical improvements, we also improved the accuracy of identifying species of underground ECM fungi by evaluating intra- and interspecific variation. Kårén et al. (1997) found intraspecific variation in the RFLP patterns of a considerable proportion of the species they examined. This intraspecific variation might cause an overestimate of the underground ECM fungal species, if the RFLP analysis were applied without considering the intraspecific variation at each research site. To grasp the intraspecific variation of a species at a site, one must examine sufficient sporocarps from genetically different units (genets) at that site. Since host vegetation patches in our research quadrat are physically isolated from each other like islands in a sea, ECM fungal mycelia in a patch cannot extend to other patches vegetatively. Therefore, the sporocarps of a species from different patches would belong to different genets. Using intersimple


sequence repeat markers and microsatellite markers, all the S. bovista sporocarps used in our study belonged to different genets (Nakaya et al., pers. comm.). Using the maximum number of genets of all sporocarp species recorded over two years, we initially tested several sets of primers and restriction enzymes to determine those appropriate for our research site, and ITS3–4 and ITS1F-HinfI were the best combination for separating all the species. This careful consideration of intra- and interspecific variation in the selection of primers and restriction enzyme combination further improved the accuracy of the species identification of underground ECM communities at our study site. Underground ECM fungi compared with sporocarps during early primary succession In all, we identified 21 species in the underground ECM community from 14 400 root tips in 72 soil samples. In Norway spruce (30 yr old), Bishop pine (35 yr old), Ponderosa pine (100 yr old), and Red fir (350–400 yr old) forests, 21, > 20, > 50, and 80 ECM fungal taxa were identified from 2,352, 2,418, 942, and 664 cm2 of ground surface actually excavated in the forests, respectively (Horton & Bruns, 2001). In these forest stands, the number of fungal species would likely increase with further sampling effort (Taylor, 2002). We identified 21 species from 7200 cm2 (0.14% total host coverage in the quadrat) of ground surface actually excavated at our early succession site. Furthermore,

753

754 Research

the number of species leveled off with the number of root tips examined (Fig. 3), indicating that most of the species at the site were detected. The number of fungal species in early successional sites should be small compared with developed forests. The underground ECM fungal community at our research site was dominated by the species whose sporocarps were recorded at this site, which was responsible for 88% in total. Since there is initially no inoculum in soil at early primary successional sites, spore dispersal from outside areas is prerequisite for ECM formation (Allen et al., 1992; Cázares & Trappe, 1994). After the initial colonization, the species that readily form sporocarps might better disperse their spores to surrounding areas to widen their distribution. The underground ECM community corresponded closely with the sporocarp community in species composition. We recorded 23 ECM fungal species in sporocarp surveys over two years (Nara et al., 2003), and all the abundant sporocarp species, including I. lacera, L. amethystina, L. laccata, L. murina and S. bovista, were also found in the underground community in this study (Fig. 4, Table 3). Ten rare sporocarp species were not detected in the underground community, probably because they are distributed sporadically underground, as were their sporocarps above-ground (Nara et al., 2003). Most of the species that were found only underground at our research site were nonmushroom fungi that do not form conspicuous mushrooms (Table 1). This indicates that most of the fungi that are able to form sporocarps formed sporocarps. In fact, sporocarp production at this site was surprisingly large due to the favorable environmental conditions for fruiting (Nara et al., 2003). The correspondence between the sporocarp and underground ECM communities may also be explained by the favorable conditions for fruiting. The constituent species in individual associations were similar between the sporocarp and ECM communities, but the ECM abundance ranks did not correspond with the sporocarp ranks, except for small hosts (Table 3). This inconsistency in abundance may be due to the different fruiting abilities of ECM fungi. Moreover, comparing the abundance of sporocarps and underground ECM tips might be inappropriate due to different sampling methods; sporocarps were collected from the entire quadrat, while ECM root tips were collected from a limited number of sampling points associated with several defined categories. However, the most abundant sporocarp species associated with a host were always ranked in the five most abundant species in that host’s underground ECM community, and conversely the most abundant species underground usually ranked in the five most abundant species in its sporocarp community (Table 3). The correspondence between the sporocarp and ECM communities has never been demonstrated during early primary succession, and there is little such data for forest stands (Gardes & Bruns, 1996; Dahlberg et al., 1997; Kårén & Nylund, 1997; Pritsch et al., 1997; Jonsson et al., 1999a,b; Horton & Bruns, 2001).

In forest stands, nonmushroom species, like C. geophilum and Thelephoraceae, often dominate the underground ECM community, and cause a discrepancy between the sporocarp and underground ECM communities (Gardes & Bruns, 1996; Dahlberg et al., 1997; Peter et al., 2001). C. geophilum, Sebacina sp. and Thelephoraceae also appeared at our research site, and they increased in relative abundance with host growth, especially inside large hosts (Fig. 4). The total colonization percentage of these fungi for a category, including other nonmushroom fungi, in the ECM community was low in small (1%) and middle-sized (2%) hosts, and outside large hosts (1%), and was high in unhealthy hosts (25%), on the periphery (18%), and inside large hosts (20%). The correspondence between the sporocarp and underground ECM communities was lower in the latter three categories than in the former three categories, although the occupation of the latter three categories by nonmushroom fungi was still low compared with the data for forest stands (Gardes & Bruns, 1996; Dahlberg et al., 1997; Peter et al., 2001). The correspondence between the sporocarp and ECM communities may be high in the earlier stages of primary succession, and may decrease with vegetation development accompanied by the colonization of nonmushroom fungi. Underground primary succession of ECM fungi The succession of ECM fungi in each vegetation patch was initiated by one or two of the three first-stage fungi: I. lacera, L. laccata, and L. amethystina. With host growth, first-stage fungi recruited each other and were joined by second-stage fungi L. murina and S. bovista. Species of Cortinarius, Hebeloma, and Thelephoraceae were late-stage fungi that were recruited with further host growth. The late-stage fungi were frequent and abundant inside large patches where the soil was relatively well developed at our research site. Except for Thelephoraceae, this sere of underground ECM fungi is quite similar to the sere of sporocarp succession with respect to species, order, and spatial distribution (Nara et al., 2003). Although the disappearance of earlier colonists is usually reported in secondary successional sites based on sporocarp surveys (Last et al., 1984), no fungal species disappeared in the sere of ECM succession in our results. However, the relative abundance of some fungal species decreased with host development in peripheral positions. For example, the relative abundance of L. laccata was higher under small hosts (Fig. 4a) than under middle-sized (Fig. 4b) and large hosts (Fig. 4e), reflecting the reduction in its ectomycorrhizas in soil samples with host growth. This suggests that L. laccata could disappear from the periphery with further host growth. However, the peripheral area itself increased considerably with host growth, so L. laccata was still abundant in peripheral positions overall. Furthermore, the relative abundance of L. laccata was still high inside and outside large hosts. Overall, the number of root tips of a species associated with each host should increase


Research

monotonically with the host size class. Interestingly, such a monotonic increase was also demonstrated in the sporocarp production of every major species at this site (Nara et al., 2003). Due to the successional recruitment of new ECM fungi with host growth, the number of species of ECM fungi associated with a host increased significantly from small to large hosts (Table 2). Although the same number of soil samples was collected from every host size class, the distance between soil samples was short in small hosts and long in large hosts. Therefore, the reduced number of species for small hosts might be partly due to the continuity of the same ECM community throughout the different samples. However, the number of species in a soil sample also increased significantly from small to large hosts (Table 2). This indicates that the species richness increased with host growth even within a small volume of soil, apart from the diversification due to the overall increment in host size. An important factor for species richness was seen in the conspicuous richness inside large hosts, which was not found in the bare ground outside the same large hosts (Table 2). Inside a vegetation patch, litter accumulates, increasing the total soil nitrogen, which also changes from inorganic to organic form (Tateno & Hirose, 1987). Litter type and soil quality strongly influence the ECM community (Conn & Dighton, 2000). During early primary succession, soil development might be necessary for the colonization of many fungal species, especially late-stage fungi, and might be a prerequisite for the increase in ECM species richness. The community structure of plant species is often analyzed using rank-abundance plots (Tokeshi, 1993). The same method has been used in studies of ECM fungal communities and is effective for evaluating the structure of fungal communities (Visser, 1995; Jonsson et al., 1999a; Horton & Bruns, 2001). Visser (1995) demonstrated that the ECM morphotype community in 6-yr-old-jack pine stands following wild fire is dominated by several morphotypes and its rankabundance pattern is a geometric series type. Although the communities in 41-, 65-, 122-yr-old stands of the same species in Visser’s study consisted of a larger number of morphotypes in a more evenly abundant structure, or a log-normal type, the differences among the latter three communities were unclear. In our study, a geometric series type community clearly shifted to the log-normal structure with host growth (Fig. 5a), and along the soil development gradient from outside to inside positions in large hosts (Fig. 5b). Similar shifts were also shown in plant community development after a disturbance from earlier stages where dominance was pronounced to later stages where relative abundances were more equitable. (Tokeshi, 1993). Therefore, our results indicate that the ECM fungal community structure develops with host growth, and develops more inside large hosts than outside. The ECM communities differed significantly between normally growing hosts and unhealthy hosts by the exact


Pearson chi-square test (P < 0.001). The unhealthy hosts had extremely low leaf biomasses and low photosynthetic abilities (Nara et al., 2003). Therefore, a limited assimilate supply from the host may inhibit the colonization of some fungi, such as I. lacera and S. bovista, and may support the colonization of nonconspicuous sporocarp species, like Sebacina, Thelephoraceae, and C. geophilum. Conversely, colonization by different ECM fungi may determine the growth and survival of S. reinii on Mt. Fuji. In this study, we demonstrated the underground primary succession pattern of ECM fungi in a volcanic desert on Mt. Fuji in terms of fungal sere, species richness, relative abundance, and community structure. We also compared the underground succession pattern with the sporocarp succession pattern. Most results indicate that these successional patterns were closely related to host growth, which accompanied soil development. Moreover, the ECM community of unhealthy hosts deviated from the normal ECM succession pattern. Therefore, the primary succession of ECM fungi might have greater impacts on host growth, survival, and vegetation succession than ever thought.

Acknowledgements This work was supported in part by a grant from PROBRAIN and Grants-in-Aid from the Ministry of Education, Culture, Sports, Science and Technology of Japan.

References Adachi N, Terashima I, Takahashi M. 1996. Central die-back of monoclonal stands of Reynoutria japonica in an early stage of primary succession on Mount Fuji. Annals of Botany 77: 477–486. Agerer R. 1987–97. Colour atlas of ectomycorrhizae. Schwäbisch Gmünd, Germany: Einhorn-Verlag. Allen MF, Crisafulli C, Friese CF, Jeakins SL. 1992. Re-formation of mycorrhizal symbioses on Mount St Helens, 1980–90: Interactions of rodents and mycorrhizal fungi. Mycological Research 96: 447–453. Baar J, Horton TR, Kretzer AM, Bruns TD. 1999. Mycorrhizal colonization of Pinus muricata from resistant propagules after a standreplacing wildfire. New Phytologist 143: 409–418. Cázares E, Trappe JM. 1994. Spore dispersal of ectomycorrhizal fungi on a glacier forefront by mammal mycophagy. Mycologia 86: 507–510. Conn C, Dighton J. 2000. Litter quality influences on decomposition, ectomycorrhizal community structure and mycorrhizal root surface acid phosphatase activity. Soil Biology and Biochemistry 32: 489–496. Dahlberg A, Jonsson L, Nylund JE. 1997. Species diversity and distribution of biomass above and below ground among ectomycorrhizal fungi in an old-growth Norway spruce forest in south Sweden. Canadian Journal of Botany 75: 1323–1335. Dickie IA, Xu B, Koide RT. 2002. Vertical niche differentiation of ectomycorrhizal hyphae in soil as shown by T-RFLP analysis. New Phytologist 156: 527–535. Gardes M, Bruns TD. 1993. ITS primers with enhanced specificity for basidiomycetes – application to the identification of mycorrhizae and rusts. Molecular Ecology 2: 113–118. Gardes M, Bruns TD. 1996. Community structure of ectomycorrhizal fungi in a Pinus muricata forest: Above- and below-ground views. Canadian Journal of Botany 74: 1572–1583.

755

756 Research Grogan P, Baar J, Bruns TD. 2000. Below-ground ectomycorrhizal community structure in a recently burned bishop pine forest. Journal of Ecology 88: 1051–1062. Helm DJ, Allen EB, Trappe JM. 1996. Mycorrhizal chronosequence near Exit Glacier, Alaska. Canadian Journal of Botany 74: 1496 –1506. Horton TR, Bruns TD. 2001. The molecular revolution in ectomycorrhizal ecology: Peeking into the black-box. Molecular Ecology 10: 1855–1871. Jonsson L, Dahlberg A, Nilsson MC, Kårén O, Zackrisson O. 1999a. Continuity of ectomycorrhizal fungi in self-regenerating boreal Pinus sylvestris forests studied by comparing mycobiont diversity on seedlings and mature trees. New Phytologist 142: 151–162. Jonsson L, Dahlberg A, Nilsson MC, Zackrisson O, Kårén O. 1999b. Ectomycorhizal fungal communities in late-successional Swedish boreal forests, and the composition following wildfire. Molecular Ecology 8: 205–215. Kårén O, Högberg N, Dahlberg A, Jonsson L, Nylund JE. 1997. Inter- and intraspecific variation in the ITS region of rDNA of ectomycorrhizal fungi in Fennoscandia as detected by endonuclease analysis. New Phytologist 136: 313–325. Kårén O, Nylund JE. 1997. Effects of ammonium sulphate on the community structure and biomass of ectomycorrhizal fungi in a Norway spruce stand in southwestern Sweden. Canadian Journal of Botany 75: 1628–1642. Last FT, Mason PA, Ingleby K, Fleming LV. 1984. Succession of fruitbodies of sheathing mycorrhizal fungi associated with Betula pendula. Forest Ecology and Management 9: 229 – 234. Lian C, Oishi R, Miyashita N, Nara K, Nakaya H, Wu B, Zhou Z, Hogetsu T. 2003. Genetic structure and reproduction dynamics of Salix reinii during primary succession on Mount Fuji, as revealed by nuclear and chloroplast microsatellite analysis. Molecular Ecology 12: 609 – 618. Liu WT, Marsh TL, Cheng H, Forney LJ. 1997. Characterization of microbial diversity by determining terminal restriction fragment length polymorphism of genes encoding 16S rRNA. Applied and Environmental Microbiology 63: 4516 – 4522. Mason PA, Wilson J, Last FT, Walker C. 1983. The concept of succession in relation to the spread of sheathing mycorrhizal fungi on inoculated tree seedlings growing in unsterile soils. Plant and Soil 71: 247– 256. Mehta CR, Patel NR. 1996. SPSS Exact Tests 7.0 for Windows. Chicago, IL, USA: SPSS Inc.

Moeseneder MM, Arrieta JM, Muyzer G, Winter C, Herndl G. 1999. Optimization of terminal-restriction fragment length polymorphism analysis for complex marine bacterioplankton communities and comparison with denaturing gradient gel electrophoresis. Applied and Environmental Microbiology 65: 3518–3525. Nara K, Nakaya H, Hogetsu T. 2003. Ectomycorrhizal sporocarp succession and production during early primary succession on Mount Fuji. New Phytologist 158: 193–206. Peter M, Ayer F, Egli S. 2001. Nitrogen addition in a Norway spruce stand altered macromycete sporocarp production and below-ground ectomycorrhizal species composition. New Phytologist 149: 311–325. Pritsch K, Boyle H, Munch JC, Buscot F. 1997. Characterization and identification of black alder ectomycorrhizas by PCR /RFLP analyses of the rDNA internal transcribed spacer (ITS). New Phytologist 137: 357– 369. Smith SE, Read DJ. 1997. Mycorrhizal symbiosis, 2nd edn. London, UK: Academic Press. Tateno M, Hirose T. 1987. Nitrification and nitrogen accumulation in the early stages of primary succession on Mt. Fuji. Ecological Research 2: 113–120. Taylor AFS. 2002. Fungal diversity in ectomycorrhizal communities: Sampling effort and species detection. Plant and Soil 244: 19–28. Taylor DL, Bruns TD. 1999. Community structure of ectomycorrhizal fungi in a Pinus muricata forest: Minimal overlap between the mature forest and resistant propagule communities. Molecular Ecology 8: 1837–1850. Tokeshi M. 1993. Species abundance patterns and community structure. Advances in Ecological Research 24: 111–186. Visser S. 1995. Ectomycorrhizal fungal succession in jack pine stands following wildfire. New Phytologist 129: 389–401. White TJ, Bruns TD, Lee S, Taylor J. 1990. Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis MA, Gelfand DH, Sninsky JJ, White TJ, eds. PCR protocols: a guide to methods and applications. New York, USA: Academic Press, 315–322. Yang G, Cha Joo Y, Shibuya M, Yajima T, Takahashi K. 1998. The occurrence and diversity of ectomycorrhizas of Larix kaempferi seedlings on a volcanic mountain in Japan. Mycological Research 102: 1503–1508. Zhou Z, Hogetsu T. 2002. Subterranean community structure of ectomycorrhizal fungi under Suillus grevillei sporocarps in a Larix kaempferi forest. New Phytologist 154: 529–539.

About New Phytologist • New Phytologist is owned by a non-profit-making charitable trust dedicated to the promotion of plant science. Regular papers, Letters, Research reviews, Rapid reports and Methods papers are encouraged. Complete information is available at www.newphytologist.org • All the following are free – essential colour costs, 25 offprints as well as a PDF (i.e. an electronic version) for each article, online summaries and ToC alerts (go to the website and click on 'Journal online') • You can take out a personal subscription to the journal for a fraction of the institutional price. Rates start at £86 in Europe/$145 in the USA & Canada for the online edition (go to the website and click on 'Subscribe') • If you have any questions, do get in touch with Central Office ([email protected]; tel +44 1524 592918) or, for a local contact in North America, the USA Office ([email protected]; tel 865 576 5261)
