Mycorrhiza DOI 10.1007/s00572-013-0535-6
ORIGINAL PAPER
Fine-scale distribution of ectomycorrhizal fungi colonizing Tsuga diversifolia seedlings growing on rocks in a subalpine Abies veitchii forest Naohiro Yoshida & Joung A Son & Norihisa Matsushita & Kojiro Iwamoto & Taizo Hogetsu
Received: 12 July 2013 / Accepted: 13 October 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Numerous species of ectomycorrhizal (ECM) fungi coexist under the forest floor. To explore the mechanisms of coexistence, we investigated the fine-scale distribution of ECM fungal species colonizing root tips in the root system of Tsuga diversifolia seedlings in a subalpine forest. ECM root tips of three seedlings growing on the flat top surface of rocks were sampled after recording their positions in the root system. After the root tips were grouped by terminalrestriction fragment length polymorphism (T-RFLP) analysis of ITS rDNA, the fungal species representing each T-RFLP group were identified using DNA sequencing. Based on the fungal species identification, the distribution of root tips colonized by each ECM fungus was mapped. Significant clustering of root tips was estimated for each fungal species by comparing actual and randomly simulated distributions. In total, the three seedlings were colonized by 40 ECM fungal species. The composition of colonizing fungal species was quite different among the seedlings. Twelve of the 15 major ECM fungal species clustered significantly within a few centimeters. Some clusters overlapped or intermingled, while others were unique. Areas with high fungal species diversity were also identified in the root system. In this report, the mechanisms underlying generation of these ECM root tip clusters in the root system are discussed. N. Yoshida (*) : J. A. Son : N. Matsushita : T. Hogetsu Graduate school of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan e-mail: [email protected] J. A. Son Division of Forest Insect Pests and Diseases, Korea Forest Research Institute, Seoul 130-712, South Korea K. Iwamoto Tama Forest Science Garden, Forestry and Forest Products Research Institute, 1833-81 Todori, Hachioji, Tokyo 193-0843, Japan
Keywords Ectomycorrhizal (ECM) fungal community . Clusters of ECM root tips . Extraradical mycelium . Interspecific interaction . Competition . Priority effect
Introduction Numerous ectomycorrhizal (ECM) fungi coexist under the forest floor and have symbiotic associations with the root systems of woody plants (Smith and Read 2008). Many analyses of subterranean community structures of ECM fungi have been performed based on morphological species identification from sporocarps (Dahlberg 2001; Horton and Bruns 2001; Richard et al. 2004) or ECM root tips (Yamada and Katsuya 1996), as well as molecular analyses of root tips (Zhou and Hogetsu 2002; Rosling et al. 2003; Lian et al. 2006; Pickles et al. 2010, 2012). To date, these studies demonstrated that many ECM fungal species coexist and are patchily distributed under the forest floor on various scales. Lilleskov et al. (2004) drew standardized variograms from the data obtained in several previous reports and found that most of the dominant taxa showed patchiness at a scale of less than 3 m. Autocorrelation analysis of ECM fungi in root tips sampled from 441 points with 1 m spacing between each other in a 20×20 m study area showed that several ECM fungal species were distributed in clusters on a meters scale (Pickles et al. 2010, 2012). Individual genets are also found patchily distributed among sporocarps (Dahlberg and Stenlid 1994; Zhou et al. 1999, 2000; Bergemann and Miller 2002) or ECM root tips (Zhou et al. 2001) of several ECM fungal species at centimeters and meters scales. On smaller scales, the fact that two soil cores separated by only several centimeters contained significantly different sets of ECM fungal species also indicated that each ECM fungal species coexisting under the forest floor occupied only a small area (Horton and Bruns 2001; Tedersoo et al. 2003; Izzo et al.
Mycorrhiza
2005). Tedersoo et al. (2003) found that the abundance of root tips and ECM fungi were highly variable on a 5-cm scale in a mixed boreal forest, and that most species were replaced on a 50-cm scale. Genney et al. (2006) investigated ECM community structures in pine forest soil and identified seven ECM fungal species in 400 soil cubes of 2×2×2 cm. Zhou and Hogetsu (2002) investigated the distribution of ECM fungal T-RFLP types under Suillus grevillei sporocarps in the Larix forest floor. They identified fungal T-RFLP types in ECM root tips sampled from a vertical soil section at a 10cm lattice interval and demonstrated that more than 30 ECM fungal taxa were distributed mosaically in a 1-m2 area. Such patchy and mosaic distribution of ECM fungal species could reflect the spreading processes of ECM colonization and the competitive interactions between ECM fungal species as pointed out by Kennedy (2010). Two types of spreading processes for ECM colonization can generally occur, namely infection of root tips by extraradical hyphae emanating from pre-existing ECM root tips or by hyphae derived from spores or remaining hyphal fragments derived from degraded ECM mycelia. The former produces a clustered distribution of root tips colonized by the same fungal species, while the latter can produce a mixture of root tips colonized by different fungal species with high species diversity. These spreading processes of an ECM fungus may be reflected in the extent of clustering of its mycelium and ECM root tips. In addition, competitiveness of the interaction between mycelia of different ECM fungi may be reflected in exclusiveness of ECM root tip distribution at the boundary coexisting ECM species. Spreading of ECM colonization and competitive interaction between different ECM fungal species may be based on the hyphal level phenomena, hyphal infection to fine root tips, and competitive interaction between hyphae of different ECM fungal species. If the patchy and mosaic distributions of ECM fungal species on meters and centimeters scales reflect the hyphal spreading and competitive interactions in the subterranean ECM fungal community, we could hypothesize that patchy and mosaic distributions on the root-tip scale also occur in the rhizosphere. Finer scale investigations on the presence of the patchy and mosaic distributions of ECM fungal species in the rhizosphere would prove if the above hypothesis is correct or not and also provide novel information on the mechanism for the patchy and mosaic distributions. In this study, we investigated the in situ distribution of each ECM fungal species coexisting under the forest floor at a roottip level to prove the above hypothesis. To examine the in situ distribution of ECM fungal species at a root-tip level, the positions of root tips in the root system should be determined more accurately. Thus, we examined the fine-scale distribution of ECM fungi colonizing Tsuga diversifolia seedlings regenerating on the flat top surface of rock in a subalpine Abies veitchii forest. Since the root systems of these
T. diversifolia seedlings were relatively flat within a layer of litters and peat mosses on the rock, the root-tip positions could be determined.
Materials and methods Research site The research site was located on the south slope of Mts. Yatsugatake, Japan (138°24′E, 35°60′N about 1,800 m asl). The soil type was wet podzol, and the forest floor was evenly covered by peat mosses. The forest was affected by selective logging 80–90 years ago and regenerated naturally. A . veitchii were dominant in the tree and sub-tree layers, and T. diversifolia and Betula ermanii trees were mixed. In the shrub and herb layers, the dominant woody plant was T. diversifolia, while A. veitchii seedlings were rare. Sampling seedlings and collecting root tips In August 2010, three T. diversifolia seedlings (approximately 10 years old) named T1, T2, and T3 were collected from three rocks, the top surfaces of which were relatively flat, together with the substrate holding the root system to conserve its in situ architecture (Table 1). The distances between T1 and T2, T1 and T3, and T2 and T3 were 61, 78, and 20 m, respectively. To obtain root tips, the root system of each seedling was fixed on a wooden board through which 5-cm nails were driven in a 5-cm lattice forming a flower arrangement; the substrate was then carefully removed from the root system to expose roots with the least positional displacement. After recording the root and root-tip positions by taking photographs of the whole root system, we collected all root tips under a dissecting microscope. Root tips were transferred individually into 2.0-mL tubes containing seven 2-mm zirconium balls and dried for DNA extraction. PCR amplification and T-RFLP analysis Fungal species in all collected root tips were identified individually by molecular analysis. DNA was extracted from Table 1 Number of root tips and ECM fungal colonization rate of each seedling Sample number
Number of root tips
Number of ECM root tips
Colonization rate
T1 T2 T3
203 214 754
175 162 534
86.2 75.7 70.8
Mycorrhiza
each root tip using a modified CTAB method (Lian et al. 2001). Briefly, the sample was pulverized for 2 min using a Micro Smash™ MS-100 (Tomy Seiko, Tokyo, Japan), after which 500 μL of 2× CTAB solution [2 % CTAB, 0.1 M Tris– HCl (pH 8.0), 20 mM ethylenediaminetetraacetic acid (EDTA; pH 8.0), 1.4 M NaCl, and 0.125 % dithiothreitol] were added and the sample pulverized again for 1 min. After incubation at 65 °C for 1 h, DNA was isolated by chloroform– isoamyl alcohol (24:1) extraction, isopropanol precipitation, and 80 % ethanol washing. The isolated DNA was dissolved in 50 μL Tris–EDTA (TE) buffer [10 mM Tris–HCl (pH 8.0) and 1 mM EDTA (pH 8.0)] and stored at −30 °C. DNA from the fungus colonizing in each root tip was labeled with a fluorescent pigment as follows. The DNA template was amplified by PCR with the fluorescent primers ITS1F (Gardes and Bruns 1993) labeled with IRD-700 and ITS4 (White et al. 1990) labeled with IRD-800. PCR was performed in a 5-μL reaction mixture containing 1× NH4 reaction buffer, 0.05 units Biotaq DNA polymerase, 1.5 mM MgCl2, 0.2 mM dNTP mixture (Bioline, London, UK), 0.02 μM labeled primers, 0.2 μM non-labeled primers, and 0.5 μL DNA template. The PCR reaction was performed under the following conditions: initial denaturation at 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 54 °C for 30 s and 72 °C for 90 s, and a final extension at 72 °C for 10 min. Two types of restricted DNA fragments were produced from each PCR product by mixing 20 μL of each PCR product with 50 μL of 1× NE buffer 4 containing 1.5 units of Hae III or 1× NE buffer 1 containing 1.5 units of Hpy CH4IV (New England BioLabs, Beverly, USA) and incubating at 37 °C for 2 h. Both restricted fragments were electrophoresed on 6 % Long Ranger sequencing gels with a LI-COR 4300 DNA sequencer (LI-COR Biosciences, Lincoln, NE, USA). Terminal restriction fragment lengths (T-RFLs) were calculated using the 1D Image Analysis Software (Kodak, New York, NY, USA). We obtained four T-RFLs from one PCR product and grouped root tips into TRFLP groups according to a combination of four T-RFLs. Sequencing One of the DNA samples from each T-RFLP group was selected randomly and amplified by PCR using the primers ITS1F and ITS4 for sequence analysis. PCR was performed in 10 μL of 1× NH4 reaction buffer containing 0.05 units Biotaq, 2.5 mM MgCl2, 0.4 mM dNTP mixture, 0.2 μM of each primer, and 2.0 μL of DNA template under the following conditions: initial denaturation at 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 54 °C for 30 s and 72 °C for 60 s, and a final extension at 72 °C for 10 min. The PCR product was subcloned into pT7Blue T-Vector (Novagen, Madison, WI, USA) using Takara DNA Ligation Kit Ver. 2.1 and Escherichia coli JM 109 Competent Cells
(Takara Bio, Shiga, Japan) according to the manufacturer’s instructions. E. coli subclones were proliferated and boiled in 30 μL of sterile water for 10 min at 96 °C. DNA fragments were amplified from the supernatant by PCR using primers M13R and U19 and sequenced. Sequencing was performed using conventional methods. All sequences were deposited in the DNA Data Bank of Japan (DDBJ; Table 2). Data analysis Sequences representing each T-RFLP group were compared with sequences registered in the UNITE (Kõljalg et al. 2005) and GenBank databases, and the most homologous sequence was identified by BLAST. When registered species matched the representing sequence with more than 97 % homology, we considered the species with the highest homology to represent the T-RFLP group. All sequences with less than 97 % homology to registered sequences in the databases were aligned with several sequences from the databases, and a phylogenetic tree was constructed based on maximumlikelihood estimation using MEGA, version 5.10 (Tamura et al. 2011). The species, genus, or family of these sequences was determined based on database sequences belonging to the same clade in the phylogenetic tree. To analyze the fine-scale distribution of ECM fungi, maps of roots and root tips were drawn by tracing photos of the root system using Adobe Illustrator CS 5 (Adobe systems, San Jose, CA, USA). The X and Y coordinates of all root tips were obtained from the root system maps and used for simulation analysis. Since roots were crowded and entangled at the foot of the seedling stem in the T1 root system (encircled by a pink solid line in Fig. 1a), positions of ECM root tips could not be determined accurately. Therefore, we excluded the data of fungal species colonizing this area from the clustering simulation analysis. Clustering of ECM root tips colonized by the same fungal species in each rhizosphere was analyzed by a program developed with the R (http://www.r-project.org/) as given in Appendix 1 of Shiraishi et al. (2011). The program hypothesized a mother population that contained root tips colonized by each ECM fungal species at the same ratio as in the actual rhizophere for each rhizosphere. To generate a randomly simulated distribution of ECM root tips in the rhizosphere, a root tip was randomly chosen from the mother population and fitted to each ECM root-tip position in the actual rhizosphere. Following such a process, 100,000 simulated distributions were generated. The program drew a circle area (50 mm in radius) or a ring area between two concentric circles of given successive radii (at a 50-mm interval) around every ECM root tip in a root-tip distribution and calculated the cluster index for every ECM fungal species. The cluster index was defined as (1)/(2)–(3)/ (4), where (1) is the total root-tip number of the given fungal
2 23 14
0
0 0
17 0 0 0
5 0 0 0 1 0 0 0 3 2 0 0 0 30 46 9
Ascomycota sp.
Helotiales sp. 1 Helotiales sp. 2
Basidiomycota Amphinema sp. Clavulina sp. 1 Clavulina sp. 2 Clavulinaceae sp.
Cortinarius sp. 1 Cortinarius sp. 2 Cortinarius sp. 3 Cortinarius sp. 4 Cortinarius sp. 5 Cortinarius sp. 6 Cortinarius sp. 7 Inocybe sp. Laccaria sp. 1 Laccaria sp. 2 Lactarius sp. Paxillus sp. Piloderma sp. 1 Piloderma sp. 2 Russula sp. 1 Russula sp. 2
0 3 6 1 0 0 0 0 0 0 58 0 4 52 0 0
0 0 0 0
0 0
0
9 9 0
0 0 0 0 0 5 12 1 0 0 0 2 7 269 0 0
0 2 2 5
26 76
6
17 4 0
280 160 453 444 231 439 443 89 730 718 487 133 97 97 492 174
92 263 508 452
109 163
75
161 161 154
438 191 194 183 442 192 195 529 730 718 246 157 501 203 208 87
467 129 54 134
158 158
434
88 88 159
606 632 647 627 609 631 638 618 730 718 334 157 180 540 219 207
104 237 718 432
410 616
112
367 169 607
112 632 647 627 113 631 638 618 730 718 110 150 124 140 129 139
511 493 718 507
533 616
403
175 175 607
ITS4
ITS1F
ITS1F
ITS4
HpyCH4IV
Hae III
T3
T1
T2
Labeled fragment lengths (bp)a
Number of ECM root tips
Ascomycota Cenococcum geophilum Cenococcum geophilum Meliniomyces sp.
Fungal species
Table 2 ECM fungi detected in all root tips
AB828017 AB828018 AB828019 AB828020 AB828021 AB828022 AB828023 AB828024 AB828025 AB828026 AB828027 AB828028 AB828029 AB828030 AB828031 AB828032
AB828013 AB828014 AB828015 AB828016
AB828011 AB828012
AB828010
AB828007 AB828008 AB828009
Access no.b
1,279 1,108 979 1,174 1,318 1,094 1,063 662 1,368 1,189 1,187 1,384 686 932 1,332 1,023
531 926 1,199 484
1,332 1,110
934
1,840 1,842 1,195
Score
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
e-151 0.0 0.0 e-135
0.0 0.0
0.0
0.0 0.0 0.0
e value
UDB002218 UDB015918 UDB000068 UDB002440 UDB002210 UDB002161 UDB001453 UDB000766 UDB001466 UDB017611 UDB003330 UDB001205 UDB001744 UDB001750 EU597082 UDB000894
UDB001719 UDB001121 EU669209 DQ497944
FJ475724 AB211249
FJ152543
DQ179119 JX145390 GU998787
Access no.
Closest BLAST matchc
663/669 (99 565/567 (99 597/628 (95 605/608 (99 680/685 (99 587/596 (98 606/625 (96 401/419 (95 694/696 (99 668/687 (97 721/760 (94 721/728 (99 600/668 (89 580/610 (95 682/684 (99 675/718 (94
397/436 (91 626/672 (93 646/654 (98 386/424 (91
%) %) %) %) %) %) %) %) %) %) %) %) %) %) %) %)
%) %) %) %)
841/892 (94 %) 604/616 (98 %)
502/507 (99 %)
980/996 (98 %) 1006/1029 (97 %) 606/607 (99 %)
Identities
Cortinarius anomalus Cortinarius casimiri Cortinarius sertipes Cortinarius sp. Cortinarius croceoconus Cortinarius diasemospermus Cortinarius rubrovioleipes Inocybe petiginosa Laccaria laccata Laccaria laccata Lactarius fulvissimus Paxillus involutus Piloderma lanatum Piloderma sphaerosporum Russula crassotunicata Russula livescens
Amphinema sp. Clavulina cristata Clavulina castaneipes uncultured Clavulinaceae
Cenococcum geophilum Cenococcum geophilum uncultured Meliniomyces |Meliniomyces variabilis| uncultured Pezizomycotina| Archaeorhizomyces| uncultured Sordariomycetes uncultured ectomycorrhizal fungus
Species
Mycorrhiza
3 2 0 0 0 0 1 2 11 0 0 21 0 57
7 735 110 223 221 223 610 221 221 609 143 139 487 480 480
746
All sequences were deposited in DDBJ and their access nos. are listed
735 151 481 179 169 96 96 483 96 97 98 96 98 98
746 202 659 111 613 645 614 202 613 206 733 235 643 111 514
190 187 659 61 61 61 61 503 61 61 733 120 643 62 120
176 AB828034 AB828035 AB828036 AB828037 AB828038 AB828039 AB828040 AB828041 AB828042 AB828043 AB828044 AB828045 AB828046 AB828047
AB828033
Access no.b
1,294 773 1,302 912 1,243 1,235 1,189 886 1,076 1,340 1,037 1,144 442 971
1,265
Score
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 e-124 0.0
0.0
e value
UDB016009 UDB000773 UDB016649 UDB016649 UDB016370 UDB016369 UDB003302 UDB016493 UDB016369 UDB011652 UDB002469 UDB002468 UDB002468 UDB002468
UDB015068
Access no.
Closest BLAST matchc
678/685 (98 585/636 (91 678/685 (98 633/687 (92 655/663 (98 655/663 (98 624/632 (98 603/651 (92 636/663 (95 713/724 (98 631/656 (96 595/599 (99 281/299 (93 575/598 (96
%) %) %) %) %) %) %) %) %) %) %) %) %) %)
656/662 (99 %)
Identities
Russula ochroleuca Sebacina sp. Thelephora albomarginata Thelephora albomarginata Tomentella lapida Tomentella terrestris Tomentella stuposa Tomentella coerulea Tomentella terrestris Tricholoma virgatum Tylospora asterophora Tylospora fibrillosa Tylospora fibrillosa Tylospora fibrillosa
Russula xerampelina
Species
ITS sequence data were compared with those of known fungal species in the UNITE database (Kõljalg et al. 2005). ITS sequences were compared with sequence data from the GenBank database when the UNITE database did not show similar sequences. Scores, e values, access nos., identities, and species names of the closest matches in BLAST searches are shown
c
b
0 0 0 0 0 3 0 0 0 0 0 2 0 0
15
Each figures represent the band sizes by T-RFLP
0 0 5 3 20 0 0 0 0 3 1 0 3 0
a
0
Russula sp. 4 Sebacina sp. Thelephoraceae sp. 1 Thelephoraceae sp. 2 Tomentella sp. 1 Tomentella sp. 2 Tomentella sp. 3 Tomentella sp. 4 Tomentella sp. 5 Tricholoma sp. Tylospora sp. 1 Tylospora sp. 2 Tylospora sp. 3 Tylospora sp. 4
ITS4
ITS1F
ITS1F
ITS4
HpyCH4IV
Hae III
T3
T1
T2
Labeled fragment lengths (bp)a
Number of ECM root tips
Russula sp. 3
Fungal species
Table 2 (continued)
Mycorrhiza
Mycorrhiza
Fig. 1 Spatial distribution of ECM on T. diversifolia root tips. Root tips were colored according to molecular homology and phylogenetic analyses of the T-RFLP types. Scale bars represent 5 cm., × T. diversifolia seedlings, black lines roots, circles root tips. a, b, and c correspond to the tree
numbers T1, T2, and T3, respectively, in Table 1. The area encircled by a pink solid line in a was excluded from the clustering simulation analysis since positions of ECM root tips could not be determined accurately. Blue dotted lines represented areas with high species diversity
Mycorrhiza
species within all circle or ring areas around root tips of the given fungal species, (2) is total ECM root-tip number within all circle or ring areas around root tips of the given fungal species, (3) is total root-tip number within all circle or ring areas around root tips of the other fungal species, and (4) is total ECM root-tip number within all circle or ring areas around root tips of the other fungal species. Clustering of root tips colonized by each ECM fungal species was estimated by comparing the clustering index of each randomly simulated distribution with that of the actual one. When the cluster index of the actual distribution was larger and smaller than that of a simulated distribution, one point was scored to the actual and simulated distributions, respectively. When the indices of both distributions were equal, 0.5 was scored to both distributions. When (2) or (4) in the actual distribution or (2) in the simulated one was zero, the comparison was considered invalid and skipped. When the actual distribution won more than 95 % of total scores (95,000 points) for an ECM fungal species in a radius range (i.e., distance range from the centered root tips), root tips colonized by the ECM fungal species were considered significantly clustered in that range. This estimation of clustering was performed on root tips of every ECM fungal species in each rhizosphere for every radius range.
Results ECM fungal diversity We obtained 1,171 root tips, 871 of which were colonized by fungi. The fungal colonization rates were 86.2, 75.7, and 70.8 % for seedlings T1, T2, and T3, respectively (Table 1). Root tips were grouped into 41T-RFLP types represented by 40 fungal species: two T-RFLP types were identified as Cenococcum geophilum. Piloderma sp. 2 was the most dominant species, accounting for 40.3 % of the total ECM root tips. Helotiales sp. 2 and C. geophilum were the second and third most abundant species, accounting for 8.7 % and 7.3 %, respectively. Thirty-five out of 40 species were detected in only one seedling. The fungal community compositions and dominant species differed among the three root systems. On seedling T1, 16 ECM fungi were present; among them, Russula sp. 1 was the most dominant species, followed by Piloderma sp. 2, C . geophilum, and Tomentella sp. 1. These four species accounted for 26.3, 17.1, 14.3, and 11.4 %, respectively, of the total ECM on root tips. Root tips from seedling T2 were colonized by 10 ECM fungi, the dominant species being Lactarius sp., Piloderma sp. 2, C. geophilum, and Russula sp. 3 at 35.8, 32.1, 11.1, and 9.3 %, respectively. Seedling T3 was the largest seedling of the three and was colonized by 21 ECM fungi. A total of 50.4 % of the root tips were colonized by Piloderma sp. 2.
Distribution of fungal species in T. diversifolia root systems Simulation analysis showed that the major fungal species in the T1 root system, Russula sp. 1, Piloderma sp. 2, and C. geophilum significantly (p
ORIGINAL PAPER
Fine-scale distribution of ectomycorrhizal fungi colonizing Tsuga diversifolia seedlings growing on rocks in a subalpine Abies veitchii forest Naohiro Yoshida & Joung A Son & Norihisa Matsushita & Kojiro Iwamoto & Taizo Hogetsu
Received: 12 July 2013 / Accepted: 13 October 2013 # Springer-Verlag Berlin Heidelberg 2013
Abstract Numerous species of ectomycorrhizal (ECM) fungi coexist under the forest floor. To explore the mechanisms of coexistence, we investigated the fine-scale distribution of ECM fungal species colonizing root tips in the root system of Tsuga diversifolia seedlings in a subalpine forest. ECM root tips of three seedlings growing on the flat top surface of rocks were sampled after recording their positions in the root system. After the root tips were grouped by terminalrestriction fragment length polymorphism (T-RFLP) analysis of ITS rDNA, the fungal species representing each T-RFLP group were identified using DNA sequencing. Based on the fungal species identification, the distribution of root tips colonized by each ECM fungus was mapped. Significant clustering of root tips was estimated for each fungal species by comparing actual and randomly simulated distributions. In total, the three seedlings were colonized by 40 ECM fungal species. The composition of colonizing fungal species was quite different among the seedlings. Twelve of the 15 major ECM fungal species clustered significantly within a few centimeters. Some clusters overlapped or intermingled, while others were unique. Areas with high fungal species diversity were also identified in the root system. In this report, the mechanisms underlying generation of these ECM root tip clusters in the root system are discussed. N. Yoshida (*) : J. A. Son : N. Matsushita : T. Hogetsu Graduate school of Agricultural and Life Sciences, The University of Tokyo, 1-1-1 Yayoi, Bunkyo-ku, Tokyo 113-8657, Japan e-mail: [email protected] J. A. Son Division of Forest Insect Pests and Diseases, Korea Forest Research Institute, Seoul 130-712, South Korea K. Iwamoto Tama Forest Science Garden, Forestry and Forest Products Research Institute, 1833-81 Todori, Hachioji, Tokyo 193-0843, Japan
Keywords Ectomycorrhizal (ECM) fungal community . Clusters of ECM root tips . Extraradical mycelium . Interspecific interaction . Competition . Priority effect
Introduction Numerous ectomycorrhizal (ECM) fungi coexist under the forest floor and have symbiotic associations with the root systems of woody plants (Smith and Read 2008). Many analyses of subterranean community structures of ECM fungi have been performed based on morphological species identification from sporocarps (Dahlberg 2001; Horton and Bruns 2001; Richard et al. 2004) or ECM root tips (Yamada and Katsuya 1996), as well as molecular analyses of root tips (Zhou and Hogetsu 2002; Rosling et al. 2003; Lian et al. 2006; Pickles et al. 2010, 2012). To date, these studies demonstrated that many ECM fungal species coexist and are patchily distributed under the forest floor on various scales. Lilleskov et al. (2004) drew standardized variograms from the data obtained in several previous reports and found that most of the dominant taxa showed patchiness at a scale of less than 3 m. Autocorrelation analysis of ECM fungi in root tips sampled from 441 points with 1 m spacing between each other in a 20×20 m study area showed that several ECM fungal species were distributed in clusters on a meters scale (Pickles et al. 2010, 2012). Individual genets are also found patchily distributed among sporocarps (Dahlberg and Stenlid 1994; Zhou et al. 1999, 2000; Bergemann and Miller 2002) or ECM root tips (Zhou et al. 2001) of several ECM fungal species at centimeters and meters scales. On smaller scales, the fact that two soil cores separated by only several centimeters contained significantly different sets of ECM fungal species also indicated that each ECM fungal species coexisting under the forest floor occupied only a small area (Horton and Bruns 2001; Tedersoo et al. 2003; Izzo et al.
Mycorrhiza
2005). Tedersoo et al. (2003) found that the abundance of root tips and ECM fungi were highly variable on a 5-cm scale in a mixed boreal forest, and that most species were replaced on a 50-cm scale. Genney et al. (2006) investigated ECM community structures in pine forest soil and identified seven ECM fungal species in 400 soil cubes of 2×2×2 cm. Zhou and Hogetsu (2002) investigated the distribution of ECM fungal T-RFLP types under Suillus grevillei sporocarps in the Larix forest floor. They identified fungal T-RFLP types in ECM root tips sampled from a vertical soil section at a 10cm lattice interval and demonstrated that more than 30 ECM fungal taxa were distributed mosaically in a 1-m2 area. Such patchy and mosaic distribution of ECM fungal species could reflect the spreading processes of ECM colonization and the competitive interactions between ECM fungal species as pointed out by Kennedy (2010). Two types of spreading processes for ECM colonization can generally occur, namely infection of root tips by extraradical hyphae emanating from pre-existing ECM root tips or by hyphae derived from spores or remaining hyphal fragments derived from degraded ECM mycelia. The former produces a clustered distribution of root tips colonized by the same fungal species, while the latter can produce a mixture of root tips colonized by different fungal species with high species diversity. These spreading processes of an ECM fungus may be reflected in the extent of clustering of its mycelium and ECM root tips. In addition, competitiveness of the interaction between mycelia of different ECM fungi may be reflected in exclusiveness of ECM root tip distribution at the boundary coexisting ECM species. Spreading of ECM colonization and competitive interaction between different ECM fungal species may be based on the hyphal level phenomena, hyphal infection to fine root tips, and competitive interaction between hyphae of different ECM fungal species. If the patchy and mosaic distributions of ECM fungal species on meters and centimeters scales reflect the hyphal spreading and competitive interactions in the subterranean ECM fungal community, we could hypothesize that patchy and mosaic distributions on the root-tip scale also occur in the rhizosphere. Finer scale investigations on the presence of the patchy and mosaic distributions of ECM fungal species in the rhizosphere would prove if the above hypothesis is correct or not and also provide novel information on the mechanism for the patchy and mosaic distributions. In this study, we investigated the in situ distribution of each ECM fungal species coexisting under the forest floor at a roottip level to prove the above hypothesis. To examine the in situ distribution of ECM fungal species at a root-tip level, the positions of root tips in the root system should be determined more accurately. Thus, we examined the fine-scale distribution of ECM fungi colonizing Tsuga diversifolia seedlings regenerating on the flat top surface of rock in a subalpine Abies veitchii forest. Since the root systems of these
T. diversifolia seedlings were relatively flat within a layer of litters and peat mosses on the rock, the root-tip positions could be determined.
Materials and methods Research site The research site was located on the south slope of Mts. Yatsugatake, Japan (138°24′E, 35°60′N about 1,800 m asl). The soil type was wet podzol, and the forest floor was evenly covered by peat mosses. The forest was affected by selective logging 80–90 years ago and regenerated naturally. A . veitchii were dominant in the tree and sub-tree layers, and T. diversifolia and Betula ermanii trees were mixed. In the shrub and herb layers, the dominant woody plant was T. diversifolia, while A. veitchii seedlings were rare. Sampling seedlings and collecting root tips In August 2010, three T. diversifolia seedlings (approximately 10 years old) named T1, T2, and T3 were collected from three rocks, the top surfaces of which were relatively flat, together with the substrate holding the root system to conserve its in situ architecture (Table 1). The distances between T1 and T2, T1 and T3, and T2 and T3 were 61, 78, and 20 m, respectively. To obtain root tips, the root system of each seedling was fixed on a wooden board through which 5-cm nails were driven in a 5-cm lattice forming a flower arrangement; the substrate was then carefully removed from the root system to expose roots with the least positional displacement. After recording the root and root-tip positions by taking photographs of the whole root system, we collected all root tips under a dissecting microscope. Root tips were transferred individually into 2.0-mL tubes containing seven 2-mm zirconium balls and dried for DNA extraction. PCR amplification and T-RFLP analysis Fungal species in all collected root tips were identified individually by molecular analysis. DNA was extracted from Table 1 Number of root tips and ECM fungal colonization rate of each seedling Sample number
Number of root tips
Number of ECM root tips
Colonization rate
T1 T2 T3
203 214 754
175 162 534
86.2 75.7 70.8
Mycorrhiza
each root tip using a modified CTAB method (Lian et al. 2001). Briefly, the sample was pulverized for 2 min using a Micro Smash™ MS-100 (Tomy Seiko, Tokyo, Japan), after which 500 μL of 2× CTAB solution [2 % CTAB, 0.1 M Tris– HCl (pH 8.0), 20 mM ethylenediaminetetraacetic acid (EDTA; pH 8.0), 1.4 M NaCl, and 0.125 % dithiothreitol] were added and the sample pulverized again for 1 min. After incubation at 65 °C for 1 h, DNA was isolated by chloroform– isoamyl alcohol (24:1) extraction, isopropanol precipitation, and 80 % ethanol washing. The isolated DNA was dissolved in 50 μL Tris–EDTA (TE) buffer [10 mM Tris–HCl (pH 8.0) and 1 mM EDTA (pH 8.0)] and stored at −30 °C. DNA from the fungus colonizing in each root tip was labeled with a fluorescent pigment as follows. The DNA template was amplified by PCR with the fluorescent primers ITS1F (Gardes and Bruns 1993) labeled with IRD-700 and ITS4 (White et al. 1990) labeled with IRD-800. PCR was performed in a 5-μL reaction mixture containing 1× NH4 reaction buffer, 0.05 units Biotaq DNA polymerase, 1.5 mM MgCl2, 0.2 mM dNTP mixture (Bioline, London, UK), 0.02 μM labeled primers, 0.2 μM non-labeled primers, and 0.5 μL DNA template. The PCR reaction was performed under the following conditions: initial denaturation at 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 54 °C for 30 s and 72 °C for 90 s, and a final extension at 72 °C for 10 min. Two types of restricted DNA fragments were produced from each PCR product by mixing 20 μL of each PCR product with 50 μL of 1× NE buffer 4 containing 1.5 units of Hae III or 1× NE buffer 1 containing 1.5 units of Hpy CH4IV (New England BioLabs, Beverly, USA) and incubating at 37 °C for 2 h. Both restricted fragments were electrophoresed on 6 % Long Ranger sequencing gels with a LI-COR 4300 DNA sequencer (LI-COR Biosciences, Lincoln, NE, USA). Terminal restriction fragment lengths (T-RFLs) were calculated using the 1D Image Analysis Software (Kodak, New York, NY, USA). We obtained four T-RFLs from one PCR product and grouped root tips into TRFLP groups according to a combination of four T-RFLs. Sequencing One of the DNA samples from each T-RFLP group was selected randomly and amplified by PCR using the primers ITS1F and ITS4 for sequence analysis. PCR was performed in 10 μL of 1× NH4 reaction buffer containing 0.05 units Biotaq, 2.5 mM MgCl2, 0.4 mM dNTP mixture, 0.2 μM of each primer, and 2.0 μL of DNA template under the following conditions: initial denaturation at 94 °C for 2 min, 30 cycles of 94 °C for 30 s, 54 °C for 30 s and 72 °C for 60 s, and a final extension at 72 °C for 10 min. The PCR product was subcloned into pT7Blue T-Vector (Novagen, Madison, WI, USA) using Takara DNA Ligation Kit Ver. 2.1 and Escherichia coli JM 109 Competent Cells
(Takara Bio, Shiga, Japan) according to the manufacturer’s instructions. E. coli subclones were proliferated and boiled in 30 μL of sterile water for 10 min at 96 °C. DNA fragments were amplified from the supernatant by PCR using primers M13R and U19 and sequenced. Sequencing was performed using conventional methods. All sequences were deposited in the DNA Data Bank of Japan (DDBJ; Table 2). Data analysis Sequences representing each T-RFLP group were compared with sequences registered in the UNITE (Kõljalg et al. 2005) and GenBank databases, and the most homologous sequence was identified by BLAST. When registered species matched the representing sequence with more than 97 % homology, we considered the species with the highest homology to represent the T-RFLP group. All sequences with less than 97 % homology to registered sequences in the databases were aligned with several sequences from the databases, and a phylogenetic tree was constructed based on maximumlikelihood estimation using MEGA, version 5.10 (Tamura et al. 2011). The species, genus, or family of these sequences was determined based on database sequences belonging to the same clade in the phylogenetic tree. To analyze the fine-scale distribution of ECM fungi, maps of roots and root tips were drawn by tracing photos of the root system using Adobe Illustrator CS 5 (Adobe systems, San Jose, CA, USA). The X and Y coordinates of all root tips were obtained from the root system maps and used for simulation analysis. Since roots were crowded and entangled at the foot of the seedling stem in the T1 root system (encircled by a pink solid line in Fig. 1a), positions of ECM root tips could not be determined accurately. Therefore, we excluded the data of fungal species colonizing this area from the clustering simulation analysis. Clustering of ECM root tips colonized by the same fungal species in each rhizosphere was analyzed by a program developed with the R (http://www.r-project.org/) as given in Appendix 1 of Shiraishi et al. (2011). The program hypothesized a mother population that contained root tips colonized by each ECM fungal species at the same ratio as in the actual rhizophere for each rhizosphere. To generate a randomly simulated distribution of ECM root tips in the rhizosphere, a root tip was randomly chosen from the mother population and fitted to each ECM root-tip position in the actual rhizosphere. Following such a process, 100,000 simulated distributions were generated. The program drew a circle area (50 mm in radius) or a ring area between two concentric circles of given successive radii (at a 50-mm interval) around every ECM root tip in a root-tip distribution and calculated the cluster index for every ECM fungal species. The cluster index was defined as (1)/(2)–(3)/ (4), where (1) is the total root-tip number of the given fungal
2 23 14
0
0 0
17 0 0 0
5 0 0 0 1 0 0 0 3 2 0 0 0 30 46 9
Ascomycota sp.
Helotiales sp. 1 Helotiales sp. 2
Basidiomycota Amphinema sp. Clavulina sp. 1 Clavulina sp. 2 Clavulinaceae sp.
Cortinarius sp. 1 Cortinarius sp. 2 Cortinarius sp. 3 Cortinarius sp. 4 Cortinarius sp. 5 Cortinarius sp. 6 Cortinarius sp. 7 Inocybe sp. Laccaria sp. 1 Laccaria sp. 2 Lactarius sp. Paxillus sp. Piloderma sp. 1 Piloderma sp. 2 Russula sp. 1 Russula sp. 2
0 3 6 1 0 0 0 0 0 0 58 0 4 52 0 0
0 0 0 0
0 0
0
9 9 0
0 0 0 0 0 5 12 1 0 0 0 2 7 269 0 0
0 2 2 5
26 76
6
17 4 0
280 160 453 444 231 439 443 89 730 718 487 133 97 97 492 174
92 263 508 452
109 163
75
161 161 154
438 191 194 183 442 192 195 529 730 718 246 157 501 203 208 87
467 129 54 134
158 158
434
88 88 159
606 632 647 627 609 631 638 618 730 718 334 157 180 540 219 207
104 237 718 432
410 616
112
367 169 607
112 632 647 627 113 631 638 618 730 718 110 150 124 140 129 139
511 493 718 507
533 616
403
175 175 607
ITS4
ITS1F
ITS1F
ITS4
HpyCH4IV
Hae III
T3
T1
T2
Labeled fragment lengths (bp)a
Number of ECM root tips
Ascomycota Cenococcum geophilum Cenococcum geophilum Meliniomyces sp.
Fungal species
Table 2 ECM fungi detected in all root tips
AB828017 AB828018 AB828019 AB828020 AB828021 AB828022 AB828023 AB828024 AB828025 AB828026 AB828027 AB828028 AB828029 AB828030 AB828031 AB828032
AB828013 AB828014 AB828015 AB828016
AB828011 AB828012
AB828010
AB828007 AB828008 AB828009
Access no.b
1,279 1,108 979 1,174 1,318 1,094 1,063 662 1,368 1,189 1,187 1,384 686 932 1,332 1,023
531 926 1,199 484
1,332 1,110
934
1,840 1,842 1,195
Score
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0
e-151 0.0 0.0 e-135
0.0 0.0
0.0
0.0 0.0 0.0
e value
UDB002218 UDB015918 UDB000068 UDB002440 UDB002210 UDB002161 UDB001453 UDB000766 UDB001466 UDB017611 UDB003330 UDB001205 UDB001744 UDB001750 EU597082 UDB000894
UDB001719 UDB001121 EU669209 DQ497944
FJ475724 AB211249
FJ152543
DQ179119 JX145390 GU998787
Access no.
Closest BLAST matchc
663/669 (99 565/567 (99 597/628 (95 605/608 (99 680/685 (99 587/596 (98 606/625 (96 401/419 (95 694/696 (99 668/687 (97 721/760 (94 721/728 (99 600/668 (89 580/610 (95 682/684 (99 675/718 (94
397/436 (91 626/672 (93 646/654 (98 386/424 (91
%) %) %) %) %) %) %) %) %) %) %) %) %) %) %) %)
%) %) %) %)
841/892 (94 %) 604/616 (98 %)
502/507 (99 %)
980/996 (98 %) 1006/1029 (97 %) 606/607 (99 %)
Identities
Cortinarius anomalus Cortinarius casimiri Cortinarius sertipes Cortinarius sp. Cortinarius croceoconus Cortinarius diasemospermus Cortinarius rubrovioleipes Inocybe petiginosa Laccaria laccata Laccaria laccata Lactarius fulvissimus Paxillus involutus Piloderma lanatum Piloderma sphaerosporum Russula crassotunicata Russula livescens
Amphinema sp. Clavulina cristata Clavulina castaneipes uncultured Clavulinaceae
Cenococcum geophilum Cenococcum geophilum uncultured Meliniomyces |Meliniomyces variabilis| uncultured Pezizomycotina| Archaeorhizomyces| uncultured Sordariomycetes uncultured ectomycorrhizal fungus
Species
Mycorrhiza
3 2 0 0 0 0 1 2 11 0 0 21 0 57
7 735 110 223 221 223 610 221 221 609 143 139 487 480 480
746
All sequences were deposited in DDBJ and their access nos. are listed
735 151 481 179 169 96 96 483 96 97 98 96 98 98
746 202 659 111 613 645 614 202 613 206 733 235 643 111 514
190 187 659 61 61 61 61 503 61 61 733 120 643 62 120
176 AB828034 AB828035 AB828036 AB828037 AB828038 AB828039 AB828040 AB828041 AB828042 AB828043 AB828044 AB828045 AB828046 AB828047
AB828033
Access no.b
1,294 773 1,302 912 1,243 1,235 1,189 886 1,076 1,340 1,037 1,144 442 971
1,265
Score
0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 0.0 e-124 0.0
0.0
e value
UDB016009 UDB000773 UDB016649 UDB016649 UDB016370 UDB016369 UDB003302 UDB016493 UDB016369 UDB011652 UDB002469 UDB002468 UDB002468 UDB002468
UDB015068
Access no.
Closest BLAST matchc
678/685 (98 585/636 (91 678/685 (98 633/687 (92 655/663 (98 655/663 (98 624/632 (98 603/651 (92 636/663 (95 713/724 (98 631/656 (96 595/599 (99 281/299 (93 575/598 (96
%) %) %) %) %) %) %) %) %) %) %) %) %) %)
656/662 (99 %)
Identities
Russula ochroleuca Sebacina sp. Thelephora albomarginata Thelephora albomarginata Tomentella lapida Tomentella terrestris Tomentella stuposa Tomentella coerulea Tomentella terrestris Tricholoma virgatum Tylospora asterophora Tylospora fibrillosa Tylospora fibrillosa Tylospora fibrillosa
Russula xerampelina
Species
ITS sequence data were compared with those of known fungal species in the UNITE database (Kõljalg et al. 2005). ITS sequences were compared with sequence data from the GenBank database when the UNITE database did not show similar sequences. Scores, e values, access nos., identities, and species names of the closest matches in BLAST searches are shown
c
b
0 0 0 0 0 3 0 0 0 0 0 2 0 0
15
Each figures represent the band sizes by T-RFLP
0 0 5 3 20 0 0 0 0 3 1 0 3 0
a
0
Russula sp. 4 Sebacina sp. Thelephoraceae sp. 1 Thelephoraceae sp. 2 Tomentella sp. 1 Tomentella sp. 2 Tomentella sp. 3 Tomentella sp. 4 Tomentella sp. 5 Tricholoma sp. Tylospora sp. 1 Tylospora sp. 2 Tylospora sp. 3 Tylospora sp. 4
ITS4
ITS1F
ITS1F
ITS4
HpyCH4IV
Hae III
T3
T1
T2
Labeled fragment lengths (bp)a
Number of ECM root tips
Russula sp. 3
Fungal species
Table 2 (continued)
Mycorrhiza
Mycorrhiza
Fig. 1 Spatial distribution of ECM on T. diversifolia root tips. Root tips were colored according to molecular homology and phylogenetic analyses of the T-RFLP types. Scale bars represent 5 cm., × T. diversifolia seedlings, black lines roots, circles root tips. a, b, and c correspond to the tree
numbers T1, T2, and T3, respectively, in Table 1. The area encircled by a pink solid line in a was excluded from the clustering simulation analysis since positions of ECM root tips could not be determined accurately. Blue dotted lines represented areas with high species diversity
Mycorrhiza
species within all circle or ring areas around root tips of the given fungal species, (2) is total ECM root-tip number within all circle or ring areas around root tips of the given fungal species, (3) is total root-tip number within all circle or ring areas around root tips of the other fungal species, and (4) is total ECM root-tip number within all circle or ring areas around root tips of the other fungal species. Clustering of root tips colonized by each ECM fungal species was estimated by comparing the clustering index of each randomly simulated distribution with that of the actual one. When the cluster index of the actual distribution was larger and smaller than that of a simulated distribution, one point was scored to the actual and simulated distributions, respectively. When the indices of both distributions were equal, 0.5 was scored to both distributions. When (2) or (4) in the actual distribution or (2) in the simulated one was zero, the comparison was considered invalid and skipped. When the actual distribution won more than 95 % of total scores (95,000 points) for an ECM fungal species in a radius range (i.e., distance range from the centered root tips), root tips colonized by the ECM fungal species were considered significantly clustered in that range. This estimation of clustering was performed on root tips of every ECM fungal species in each rhizosphere for every radius range.
Results ECM fungal diversity We obtained 1,171 root tips, 871 of which were colonized by fungi. The fungal colonization rates were 86.2, 75.7, and 70.8 % for seedlings T1, T2, and T3, respectively (Table 1). Root tips were grouped into 41T-RFLP types represented by 40 fungal species: two T-RFLP types were identified as Cenococcum geophilum. Piloderma sp. 2 was the most dominant species, accounting for 40.3 % of the total ECM root tips. Helotiales sp. 2 and C. geophilum were the second and third most abundant species, accounting for 8.7 % and 7.3 %, respectively. Thirty-five out of 40 species were detected in only one seedling. The fungal community compositions and dominant species differed among the three root systems. On seedling T1, 16 ECM fungi were present; among them, Russula sp. 1 was the most dominant species, followed by Piloderma sp. 2, C . geophilum, and Tomentella sp. 1. These four species accounted for 26.3, 17.1, 14.3, and 11.4 %, respectively, of the total ECM on root tips. Root tips from seedling T2 were colonized by 10 ECM fungi, the dominant species being Lactarius sp., Piloderma sp. 2, C. geophilum, and Russula sp. 3 at 35.8, 32.1, 11.1, and 9.3 %, respectively. Seedling T3 was the largest seedling of the three and was colonized by 21 ECM fungi. A total of 50.4 % of the root tips were colonized by Piloderma sp. 2.
Distribution of fungal species in T. diversifolia root systems Simulation analysis showed that the major fungal species in the T1 root system, Russula sp. 1, Piloderma sp. 2, and C. geophilum significantly (p
