Multi?Host Ectomycorrhizal Fungi Are Predominant in a Guinean Tropical

Environmental Microbiology (2010) 12(8), 2219–2232 doi:10.1111/j.1462-2920.2010.02183.x

Multi-host ectomycorrhizal fungi are predominant in a Guinean tropical rainforest and shared between

canopy trees and seedlingsemi_2183 2219..2232

Abdala Gamby Diédhiou,1,2*† Marc-André Selosse,3 lings harboured a similar fungal community. These Antoine Galiana,1 Moussa Diabaté,1,4 findings suggest that there was a potential for the Bernard Dreyfus,1 Amadou Moustapha Bâ,1,5 formation of common mycorrhizal networks in close Sergio Miana de Faria6 and Gilles Béna1 vicinity. However, no significant difference was 1Laboratoire des Symbioses Tropicales et detected for the d13C and d15N values between seed- Méditerranéennes, UMR113 – INRA/AGRO- lings and adults of each ECM plant, and no ECM M/CIRAD/IRD/UM2 – TA10/J, Campus International de species exhibited signatures of mixotrophy. Our Baillarguet, 34398 Montpellier Cedex 5, France. results revealed (i) variation in ECM fungal diversity 2Laboratoire Commun de Microbiologie, according to the seedling versus adult development IRD/UCAD/ISRA, BP 1386 Dakar, Sénégal. stage of trees and (ii) low host specificity of ECM 3Centre d’Ecologie Fonctionnelle et Evolutive (CNRS, fungi, and indicated that multi-host fungi are more UMR 5175), Equipe Interactions Biotiques, 1919 Route abundant than single-host fungi in this forest stand. de Mende, 34293 Montpellier Cedex 5, France. 4Institut de Recherche Agronomique de Guinée, Division Introduction des Cultures Pérennes, Programme Recherche Forestière, BP 1523, Conakry, République de Guinée. Ectomycorrhizal (ECM) symbiosis involves soil fungi and 5Laboratoire de Biologie et Physiologie Végétales, tree roots. It provides mineral nutrients, water and protec- Faculté des Sciences Exactes et Naturelles, Université tion against pathogens to the plant which, as a reward, des Antilles et de la Guyane, BP 592, 97159 provides carbon to its fungal partner (Smith and Read, Pointe-à-Pitre, Guadeloupe, France. 2008). Each individual tree associates with several ECM 6Embrapa Agrobiologia km 47, antiga estrada Rio-São fungal species, while fungal species display variable Paulo, Seropédica 23851-970 Brazil. levels of specificity, ranging from highly specific species to generalists (= multi-host; Selosse et al., 2006; Smith Summary et al., 2009). Multi-host ECM fungi are often abundant, at The diversity of ectomycorrhizal (ECM) fungi on adult least in Holarctic and Mediterranean ECM communities trees and seedlings of five species, Anthonotha fra- (e.g. Kennedy et al., 2003; Dickie et al., 2004; Richard grans, Anthonotha macrophylla, Cryptosepalum tet- et al., 2005), but sometimes display host preferences as raphyllum, Paramacrolobium coeruleum and Uapaca shown in mixed Japanese forests (Ishida et al., 2007), esculenta, was determined in a tropical rain forest of and Tasmanian sclerophyllous forest (Tedersoo et al., Guinea. Ectomycorrhizae were sampled within a 2008). Some fungal taxa are more specific, e.g. suilloids, surface area of 1600 m2, and fungal taxa were identi- which are almost entirely restricted to Pinaceae (Bruns fied by sequencing the rDNA Internal Transcribed et al., 2002), and associates of Alnus (Tedersoo et al., Spacer region. Thirty-nine ECM fungal taxa were 2009) in Holarctic regions. determined, of which 19 multi-hosts, 9 single-hosts Very few studies have addressed the question of spe- and 11 singletons. The multi-host fungi represented cific versus multi-host abilities in tropical forests, where 92% (89% when including the singletons in the analy- ECM associations remain little studied (Alexander et al., sis) of the total abundance. Except for A. fragrans, the 1992; Alexander and Lee, 2005; Tedersoo et al., 2007a). adults of the host species displayed significant differ- In tropical Africa, there are currently few reports of ECM entiation for their fungal communities, but their seed- host preferences, particularly in mixed forests. In this region, ECM trees include Caesalpinioideae, Phyllan- Received 14 May, 2009; accepted 21 December, 2009. *For thaceae as well as some Dipterocarpaceae, Proteaceae correspondence. E-mail [email protected]; Tel. (+221) 33 and Sapotaceae (Högberg and Piearce, 1986; Newbery 849 33 32; Fax (+221) 33 849 33 02. †Present address: Laboratoire Commun de Microbiologie, IRD/UCAD/ISRA, BP 1386 Bel-Air, Dakar, et al., 1988; Torti and Coley, 1999; Ducousso et al., 2004), Senegal. but ECM fungal communities and their links with

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd 2220 A. G. Diédhiou et al. composition of the plant community remain little studied. not yet been used to test for the origin of seedlings’ Fruit body collections from mixed African forests have carbon. already suggested that numerous ECM fungi are multi- In the Southern Guinea rainforests (West Africa), Cae- host (Sanon et al., 1997; Rivière et al., 2007). Thoen and salpinioideae and Phyllanthaceae trees are the most Bâ (1989) suspected host effects on ECM fungi collected abundant native ECM species, growing in mixed patches, under Afzelia africana and Uapaca guineensis in southern which raises the possibility of some fungal sharing. A high Senegal. Nevertheless, fruit body survey cannot predict regeneration occurs under adult trees, strongly suggest- the ECM association patterns at the root level, because ing whether any ‘nurse effect’ occurs, that is: (i) do coex- below-ground fungal diversity and abundance differ from isting trees and seedlings harbour the same fungal those observed above-ground (Richard et al., 2005). communities? and/or (ii) do ECM fungi provide carbohy- Inoculation of seedlings demonstrated that some African drates to seedlings? In this typical primary forest area, we fungal isolates are compatible with multiple tree species first described the fungal diversity on ECM root tips, to (Diédhiou et al., 2005). Hence, in situ studies of ECM assess ECM association patterns among the five domi- diversity at root level are still required to assess the diver- nant host species and between both development stages, sity and specificity (versus host sharing) of the ECM i.e. adults versus seedlings. Second, in order to test fungal community from tropical Africa. whether some carbohydrates were transferred to seed- Some ECM tropical tree species tend to aggregate in lings, we compared the 13C and 15N contents of the latter patches where they dominate together or alone (Newbery with those of adult trees, fruit bodies of ECM fungi and et al., 2004; Alexander and Lee, 2005), surrounded by surrounding AM tree species. trees forming arbuscular mycorrhizae (AM; McGuire, 2007). In the Korup National Park (Cameroon), ECM trees form up to 70% of local patches (Newbery et al., 1997); Results similarly, the ECM Gilbertiodendron dewevrei represents Ectomycorrhizal diversity more than 90% of trees in some stands of the Congo basin, with abundant seedlings (Hart et al., 1989). Such From 12 adult trees (Fig. 1) and 138 surrounding seed- mono- or oligo-dominant stands suggest efficient regen- lings, 150 and 190 root samples were, respectively, col- eration under adults. It has been claimed that adult trees lected (Table 1). From these samples, 362 ECM tips were could provide ECM inocula to seedlings, in the form of obtained, and 293 were successfully amplified and already established (and perhaps even already nour- sequenced (81%, Table 1); only 10 ECM tips exhibited ished) fungi. This likely favours the survival of seedlings in two different Internal Transcribed Spacer (ITS) sequences temperate ecosystems (Nara, 2006) and tropical ones at the same time. The 303 different ITS sequences (Onguene and Kuyper, 2002; McGuire, 2007). However, obtained represented 39 ECM fungal taxa (i.e. diverging no data are available on fungal sharing between adults sequences) from seven fungal families: Boletaceae, Cla- and seedlings in the latter ecosystems. vulinaceae, Russulaceae, Sclerodermataceae, Thelepho- Besides an indirect interaction (with adults paying the raceae, Tricholomataceae and three other undetermined cost of the fungus more than the seedlings), facilitation families (Table 2). can also occur thanks to carbon and nutrients transferred from adults to seedlings by way of mycorrhizal networks, Ectomycorrhizal abundance and distribution interconnecting root systems of the same or different among host species species (Simard et al., 1997). Labelling experiments have shown in both laboratory and field conditions that carbon Among the 39 ECM taxa, 20 (including 11 singletons) and nutrients can be transferred between plants through were found on a single host species and 19 were multi- mycorrhizal networks (Selosse et al., 2006). However, the host: six were shared by the five tree species, two by four, role of the hyphal pathway during carbon movement, and two by three and nine by two tree species (Table 2). Each the physiological or ecological relevance of inter-plant tree species associated with 10–15 multi-host ECM fungal transfers remain controversial (e.g. Fitter, 2001; Wu et al., taxa, 2–3 single-host ECM fungal taxa, and 1–5 single- 2001). Natural abundance of stable isotopes has given tons. To compare the abundance of single-host ECM taxa increased support to such exchanges, at least in oversto- and multi-host ECM taxa, the singletons were excluded rey receiver species in forest ecosystems (Selosse and from the analysis. Therefore, from the considered 292 ITS Roy, 2009). Such plants receiving carbohydrates from sequences, the multi-host ECM taxa represented 92% of their mycorrhizal fungi exhibit 13C and 15N abundances the total ECM abundance versus 8% for the single-host intermediate between mycorrhizal fungi providing some ECM taxa (Fig. 2). The five tree species displayed rela- carbon and autotrophic plants placed in similar conditions tively similar abundances of multi-host ECM taxa (18– (Julou et al., 2005; Tedersoo et al., 2007b) – but this has 22% of total abundance of multi-host ECM). The most

Fig. 1. Map of the study site, with adults (big squares) and (m) seedlings (circles) of the ﬁve ECM tree species, Anthonotha 80 fragans (grey), A. macrophylla (yellow), Cryptosepalum tetraphyllum (red), Paramacrolobium coeruleum (blue), Uapaca O P esculenta (green). The ﬁeld stand is divided into 16 plots (A to P). The square shapes with bars represent the adults from which ECM were sampled.

70 abundant ECM fungal taxon, Russulaceae #16 occurred on the five host species and represented 47% of the overall abundance. A comparison of the frequency of each multi-host fungal M N taxon on the five host species revealed that some multi- 60 host fungal taxa tend to display a preferential occurrence P (Table 2). For instance, Thelephoraceae #8 was found on K L P P P P only Anthonotha fragrans and Anthonotha macrophylla, P P PP and was more frequent (nine occurrences) on A. fragrans P than on A. macrophylla (only one occurrence). Boleta- P ceae #2 was also found on only two host plant species 50 with five occurrences on A. fragrans and one occurrence P on Paramacrolobium coeruleum. At the fungal family level, the factorial correspondence I P J P analysis (FCA) showed that Boletaceae, Clavulinaceae and Thelephoraceae tended to associate preferentially with A. fragrans: these fungal families were associated 40 with positive values of the axis F1 (with high contribution to this axis; 26%, 26% and 25% respectively), while Rus- sulaceae was associated with negative values of this axis (with a 18% contribution) and tended to associate more with the other tree species (Fig. 3). G H 30 Composition of the ECM community among host species and development stages Anthonotha fragrans displayed different fungal communi- E F ties relative to other species, except P. coeruleum for its seedlings. Conversely, whereas the adult trees of the four 20 remaining host species displayed significantly different fungal communities from each other, except the pair P. coeruleum – A. macrophylla, their seedlings had similar C D fungal communities (Table 3). This was also well supported in FCA, where seedlings were more closely patched than adults (Fig. 3). 10 Moreover, similarity values were higher for all seedlings pairs (Sørensen coefficient Ն 0.44) than for adult tree pairs or for seedlings and adult tree pairs, regardless of the host species, except for Cryptosepalum tetraphyllum seedlings and A. macrophylla adults (0.57; Table 3). A B Moreover, adults of Uapaca esculenta, P. coeruleum and 0 A. macrophylla had higher similarity to non-conspecific 01020(m) seedlings than to conspecific seedlings (Table 3). For instance, U. esculenta adults displayed 21% similarity to their own seedlings versus 30 Ϯ 3% (mean Ϯ SE) to non- conspecific seedlings. The opposite trend was observed

Table 1. Number of root samples, ECM tips and ITS sequences obtained from adults (a) and seedlings (s) of the ﬁve host tree species.

A. fragrans A. macrophylla C. tetraphyllum P. cœruleum U. esculenta

asa s a s asas

No. of sampled individuals 1 24 3 37 2 33 3 30 3 24 No. of root samples 30 34 30 37 30 46 30 41 30 32 No. of ECM tips sampled for DNA analysis 34 36 31 39 33 49 32 43 30 35 No. of ECM tips successfully sequenced 32 30 21 27 22 40 26 38 25 32 No. of ITS sequences obtained 32 31 22 28 22 41 26 41 28 32 No. of ECM taxaa 8 10 8 10 6 13 7 15 11 8 No. of ECM taxa (after rarefaction to 22) 7.4 8.7 8.0 8.4 6.0 9.0 6.6 10.1 9.4 6.9 a. Each sequence differing from the other is considered as a different taxon. for C. tetraphyllum (C. tetraphyllum adults were 32% diversity for seedlings, except for U. esculenta. Although similar to their seedlings versus 17 Ϯ 4% to non- significant qualitative differences were observed between conspecific seedlings) while no significant difference was seedling and mature stages of the different trees species observed for A. fragrans (adults were 33% similar to their for fungal diversity (not shown), 12 out of 19 multi-host conspecific seedlings versus 28 Ϯ 3% to non-conspecific fungal taxa occurred on both adults and seedlings. Con- seedlings). When pooling seedlings and adults together, versely, when considering the single-host fungal taxa species of Caesalpinioideae did not share significantly solely or including singletons, adults had a higher level of more ECM fungal taxa among them than with the Phyl- diversity than seedlings (1.62 versus 1.32, respectively, lanthaceae U. esculenta: the average similarity was when considering single-host taxa solely and rarefying at 55 Ϯ 5% for Caesalpinioideae species pairs versus n = 8, and 2.22 versus 1.97, respectively, when admitting 49 Ϯ 3% for Caesalpinioideae and Phyllanthaceae singletons and rarefying at n = 12). Altogether, in accor- species pairs, so that host phylogenetic position poorly dance with the previous results (Table 3), adults tended to structured the ECM community. harbour more single-host ECM fungi than their seedlings. When considering the 19 multi-host fungal taxa exclu- sively and the five tree species combined, the Shannon Ectomycorrhizal composition among the four plots index showed that seedlings had a larger diversity than adults (1.89 versus 1.72, respectively, for both accumula- When combining all host species and stages, the similar- tion curves rarefied at n = 109). Rarefaction analyses ity of the ECM fungal assemblages among the four plots (Fig. S1) suggested that diversity did not greatly vary B, G, L and O was relatively low (Table 4). The inter- between cohorts (Table 1), and also supported a higher plot distances did not correlate with similarity in fungal

Fig. 2. Multi-host ECM fungal taxa (above the zero line of the histogram), single-host ECM fungal taxa and singletons (below the zero line of the histogram) classiﬁed by rank of abundance. Taxa were recovered from seedlings only (white), adults only (grey), or both development stages (black). For clarity, the most abundant ECM taxon, Russulaceae #16 (47% of the total abundance) was omitted.

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 2219–2232 00SceyfrApidMcoilg n lcwl ulsigLtd, Publishing Blackwell and Microbiology Applied for Society 2010 ©

Table 2. List of ECM fungal taxa found on the study stand.

Best BLASTN full-length ITS match

ECM taxa GenBank Accession No. Specimen / Accession no Identity (%) Host species (frequency) Development stage of host

Basidiomycota #1a AM113461 Uncultured soil fungus / DQ420877 93 Am (1), Ct (2) Both stages Basidiomycota #2a AM113462 Fungal endophyte / FJ450050 82 Ue (2) Adults Basidiomycota #3a AM113463 uncultured Basidiomycota / AY969518 97 Af (1), Am (1), Ct (1), Pc (1), Ue (1) Both stages Boletaceae #1 AM113453 Boletus bicolour / GQ166877 80 Af (3), Am (1), Pc (2), Ue (2) Both stages Boletaceae #2 AM113454 Bothia castanella / DQ867114 84 Af (5), Pc (1) Both stages Boletaceae #3 AM113455 Phylloporus rhodoxanthus / DQ533980 79 Af (2) Adults

Clavulinaceae #1 AM113459 Clavulina castaneipes / EU669209 82 Af (4), Ct (1) Seedlings rainforest tropical Guinean a in predominant are fungi ectomycorrhizal Multi-host Clavulinaceae #2 AM113460 Clavulinaceae sp. / AJ534708 81 Af (1) Seedlings Russulaceae #1 AM113427 Russula compacta / EU598172 83 Am (6), Ct (1), Pc (2), Ue (1) Both stages Russulaceae #2 AM113428 Uncultured Russulaceae / DQ777978 84 Af (1), Pc (1) Adults Russulaceae #3 AM113429 Uncultured Russulaceae / DQ777978 82 Ct (3) Adults Russulaceae #4 AM113430 Lactarius pelliculatus / AY606978 95 Am (1), Pc (1) Seedlings Russulaceae #5 AM113431 Lactarius pelliculatus / AY606978 95 Am (1), Ct (1), Pc (2) Both stages Russulaceae #6 AM113432 Lactarius pelliculatus / AY606978 96 Ue (1) Adults Russulaceae #7 AM113433 Uncultured Russula / AY667426 89 Pc (2) Adults Russulaceae #8 AM113434 Lactarius pelliculatus / AY606978 95 Ct (3) Both stages Russulaceae #9 AM113435 Lactarius pelliculatus / AY606978 95 Ue (1) Adults Russulaceae #10 AM113436 Lactarius pelliculatus / AY606978 95 Ue (1) Adults Russulaceae #11 AM113437 Lactarius pelliculatus / AY606978 95 Pc (2) Both stages Russulaceae #12 AM113438 Lactarius pelliculatus / AY606978 95 Ue (1) Adults niomna Microbiology Environmental Russulaceae #13 AM113439 Lactarius pelliculatus / AY606978 95 Pc (2), Ue (2) Both stages Russulaceae #14 AM113440 Lactarius pelliculatus / AY606978 95 Ue (2) Seedlings Russulaceae #15 AM113441 Uncultured fungus / FM999659 85 Af (1), Am (1), Ct (2), Pc (1), Ue (5) Seedlings Russulaceae #16 AM113442 Lactarius pelliculatus / AY606978 96 Af (17), Am (29), Ct (34), Pc (34), Ue (29) Both stages Sclerodermataceae #1 AM113464 Scleroderma sp. / AB099900 93 Am (1), Ct (2), Ue (2) Seedlings Sclerodermataceae #2 AM113465 Scleroderma sp. / AB099900 94 Af (2), Am (1), Ct (3), Pc (1), Ue (5) Both stages Thelephoraceae #1 AM113443 Uncultured Tomentella / EF218826 91 Af (3) Adults Thelephoraceae #2 AM113444 Uncultured Thelephoraceae / DQ273420 89 Am (1), Pc (1) Seedlings Thelephoraceae #3 AM113445 Uncultured ectomycorrhizal fungus / FM993211 92 Ct (5), Pc (1) Both stages Thelephoraceae #4 AM113446 Uncultured Tomentella / GQ240908 92 Ct (1) Seedlings Thelephoraceae #5 AM113447 Uncultured Thelephoraceae / GQ240903 90 Ct (1) Adults

, Thelephoraceae #6 AM113448 Uncultured fungus / GQ205372 88 Af (11), Am (1), Ct (2), Pc (5), Ue (1) Both stages 12 Thelephoraceae #7 AM113449 Uncultured ectomycorrhizal fungus / FM993262 88 Af (1) Seedlings 2219–2232 , Thelephoraceae #8 AM113450 Uncultured fungus / GQ205372 88 Af (9), Am (1) Adults Thelephoraceae #9 AM113451 Uncultured fungus / FJ820581 89 Pc (3) Seedlings Thelephoraceae #10 AM113452 Uncultured fungus / GQ205371 88 Af (2), Am (3), Ct (1), Pc (4), Ue (3) Both stages Tricholomataceae #1 AM113456 Uncultured fungus / FJ820560 97 Ue (1) Adults Tricholomataceae #2 AM113457 Tricholoma atroviolaceum / AY750166 91 Pc (1) Seedlings Tricholomataceae #3 AM113458 Uncultured mycorrhizal fungus / AB454382 94 Am (1) Adults

The closest BLAST match represents the most accurate taxonomic match between our ITS sequences and those in the NCBI database. Host species are Anthonotha fragrans (Af), A. macrophylla (Am), Cryptosepalum tetraphyllum (Ct), Paramacrolobium cœruleum (Pc), and Uapaca esculenta (Ue). a. The fungi corresponding to these sequences were considered ectomycorrhizal since BLAST analysis revealed a lot of sequences ampliﬁed from ectomycorrhizas in GenBank (but an unclear

taxonomic position); anyway, a status of endophytes or contaminant cannot be rigorously ruled out until a ﬁner taxonomic position is obtained. 2223 2224 A. G. Diédhiou et al.

1,5 Fig. 3. First factorial plane projection of ECM Clavulinaceae taxonomical groups (closed circles), adults and seedlings of the ﬁve host tree species (open square). The axes F1 and F2 represent 54.4% and 24% of the total variability respectively. Afa, A. fragans adult; Afs, A. 1 fragans seedlings; Ama, A. macrophylla adults; Ams, A. macrophylla seedlings; Cta, C. tetraphyllum adults; Cts, C. tetraphyllum OtherBasidiomycota seedlings; Pca, P. coeruleum adults; Pcs, P. coeruleum seedlings; Uea, U. esculenta adults; Ues, U. esculenta seedlings. Afs 0,5 Uea Trichlolomataceae

Sclerodermataceae Cts Boletaceae

F2 (24,04 %) Ama Russulaceae 0 Ams Ues Pca Pcs

Cta

Thelephoraceae -0,5 Afa

-1 -1,5 -1 -0,5 0 0,5 1 1,5 2 F1 (54,36 %) composition: e.g. plots B was more similar to O than to G, each plot), and neither did Sclerodermataceae #2 (always although G was closer to B. Similar results were obtained < 4% in total abundance), showing that the detection when considering adults and seedlings separately (not on several plots was not necessarily related to the shown). Some ECM taxa tended to associate with par- abundance. ticular plots, as shown by the c2 independence test Moreover, ECM taxa did not present the same apparent (P = 0.999). For example, Russulaceae #1, Thelephora- extent inside the different plots. For example, Russu- ceae #3 and Thelephoraceae #6 (the most abundant taxa laceae #16 was among the most widespread ECM taxa in in plots O, L and G respectively) were recorded in only the four plots, while Thelephoraceae #10, which covered one or two plot(s), suggesting aggregated distribution for one of the largest apparent areas in plot B, was not found some taxa. Conversely, Russulaceae #16 did not prefer in plot G. The longest distance between the two outermost any particular plot (15–47% of the total abundance in sampled ectomycorrhizas was recorded for Thelephora-

Table 3. Genotypic differentiation tests (upper half of the matrix) among fungal communities isolated from adults and seedlings of the five host species (P-values reported were estimated from the Fisher exact test using the Genepop program, and bold values indicate significant or marginally significant differentiation), and similarity index (lower half) between host species (similarity values calculated as the coefficient of Sørensen from occurrence of ECM fungal taxa on adults and seedlings).

A. fragrans A. macrophylla C. tetraphyllum P. cœruleum U. esculenta

Host species Adult Seedlings Adults Seedlings Adults Seedlings Adults Seedlings Adults Seedlings

A. fragrans Adult < 0.001 < 0.001 < 0.001 < 0.001 Seedlings 0.33 0.019 0.005 0.059 < 0.001 A. macrophylla Adults 0.38 0.22 0.008 0.28 0.03 Seedlings 0.22 0.60 0.33 0.68 0.70 0.52 C. tetraphyllum Adults 0.29 0.13 0.29 0.13 < 0.001 0.002 Seedlings 0.29 0.52 0.57 0.61 0.32 0.12 0.32 P. cœruleum Adults 0.40 0.24 0.40 0.35 0.15 0.40 < 0.001 Seedlings 0.35 0.56 0.26 0.64 0.29 0.43 0.27 0.14 U. esculenta Adults 0.21 0.29 0.21 0.29 0.24 0.25 0.11 0.39 Seedlings 0.25 0.44 0.38 0.67 0.14 0.57 0.40 0.44 0.21

Table 4. ECM taxa similarity among plots B, G, L and O, pooling data seedlings (higher by 1.05‰ in plot L, P = 0.001). Alto- from adults and seedlings. gether, when pooling data of adults on each site, d13C values of ECM and AM legumes did not differ (P > 0.05), Plot B G L O and were higher than those of non-legumes in plot L only B – 0.28 0.31 0.48 (P = 0.001); d15N never significantly differed among these G 20 – 0.21 0.36 13 L 40 10 – 0.53 three categories. No significant differences in d C and O603010–d15N were recorded between seedlings and adult individuals for all ECM tree species on both plots (Fig. 4); these Similarity values calculated as the coefficient of Sørensen are reported in the upper halves of the matrices, and distances (m) differences were not significant for theAM legumes, except among stands are reported in the lower half of the first matrix. for Bussea occidentalis d13C values on plot B that were significantly higher for seedlings (+0.9‰, Fig. 4A). Among ceae #6 in plot G, while it displayed a relatively small non-legumes, the only significant differences between apparent area size in plot O, and was not found in plots B adults and seedlings were for T. longifolia d13C values in and L. On the other hand, there was some overlap in plot B (seedlings values are 3.4‰ lower, Fig. 4A), and spatial distribution of ECM taxa in plots (Fig. S2). Newbouldia laevis in plot L (d13C was 1.4‰ higher and d15N was 2.2‰ higher for seedlings, Fig. 4D, 4E). The fungal fruit bodies differed in d13C compared with all other tree Isotope signature, C/N ratio and N concentration of the samples (d13C = –25.84 Ϯ 1.80‰, mean Ϯ SE; P = 0.001; plant leaves not shown). Their d15N values were largely dispersed Whenever seedlings or adults of the same species were (d15N = 2.17 Ϯ 2.31‰; not shown), and did not differ available from the investigated plots B and L (n = 6 (P > 0.05) from all other tree samples. species), the differences in d13C and d15N values were The C/N ratio values measured did not differ signifi- not significant, except for A. macrophylla d13C values in cantly for species investigated on the two plots, except for

Fig. 4. Values of d13C and d15N (‰) and C/N ratio of plants collected in plot B (A, B, C) and plot L (D, E, F), with indication of signiﬁcant differences between seedlings and adults for each species according to pairwise ANOVA (*P < 0.05; **P < 0.01; ***P < 0.001). Ectomycorrhizal trees: Ama and Ams correspond to Anthonotha macrophylla adults and seedlings respectively; and in the same way: Cta and Cts for Cryptosepalum tetraphyllum; Pca and Pcs, Paramacrolobium coeruleum; Uea and Ues, Uapaca esculenta. Non-ECM legumes: Boa and Bos, Bussea occidentalis; Ega, Entala gigas adults. Non-legumes: Baa and Bas, Bosquiea angolensis; Nla and Nls, Newbouldia laevis; Pba and Pbs, Pentadesma butyracea; Tla and Tls, Trichoscypha longifolia.

B. occidentalis (higher C/N ratio in plot L; P = 0.01; Fig. 4). ences for various fungi suggest that tropical ECM fungi When pooling data of adults on each plot, C/N ratio did not are more specific than temperate ones (Smits, 1994), but differ between AM legumes and non-legumes, or between very few data support this in African tropical forests ECM and AM species (P > 0.05), as well as when consid- (Onguene, 2000; Onguene and Kuyper, 2002). Specificity ering total N content (P > 0.05; not shown). C/N ratios in is thought to be low in tropical ECM Dipterocarpaceae leaves of adults and seedlings never differed for all from Asia (Alexander and Lee, 2005; K. Nara, pers. species, with the exception of lower C/N ratios in seed- comm.). In our study, half of fungal taxa displayed a broad lings for the AM legume B. occidentalis in plot L (Fig. 4F) host-range. In temperate forests, early reports suggested and for the non-legumes B. angolensis and T. longifolia in dominance of multi-host fungi in forest ecosystems (up to plot B (Fig. 4C). 90% of the ECM fungal community: Horton and Bruns, 1998; Cullings et al., 2000), while later reports revealing Discussion more rare fungal species indicated only 15–35% multi- Pattern of an ECM fungal community from host species (Kennedy et al., 2003; Richard et al., 2005; tropical Africa Ishida et al., 2007; Tedersoo et al., 2008). However, multi- host ECM species tend to be the most abundant: here, Up to now, with a few exceptions (Tedersoo et al., 2007a), they represent 89% (92% when excluding singletons from most studies on ECM communities carried out in African the analysis) of the overall ECM colonization. Congru- tropical rainforests have focused on epigeous fruit bodies ently, in a Mediterranean forest, while representing only (e.g. Rivière et al., 2007). We report here one of the first 15% of the fungal taxa, multi-host ECM fungi colonized c. samplings of ECM roots from five host species in a 70% of the roots (Richard et al., 2005). Moreover, we primary rainforest. It showed 39 ECM fungal taxa from cannot exclude that the apparent specificity of some 2 362 ECM tips on 1600 m , a species number to area ratio fungal taxa only reflects their rarity in the restricted sam- a bit lower than for ECM communities in Holarctic regions pling area (Fig. 2): all single-host taxa displayed no more (e.g. Richard et al., 2005; Walker et al., 2005; Ishida et al., than 1% of the total abundance. Apparently single-host 2007). Whether this reflects a trend to less diversity in taxa and singletons may result from insufficient sampling. tropical ECM community, as observed in Asia (K. Nara, However, several multi-host taxa were not abundant, pers. comm.) requires further analysis of other sites. In ranging from 2% to less than 1%, so that the ability to be addition, our low sample size and small sampling area a multi-host is not fully related to abundance. Thus, probably strongly underestimate the diversity. although possibly underestimating the number of multi- The dominance of Russulaceae and Thelephoraceae in host ECM taxa, our study shows the potential for mycor- this community is reminiscent of temperate ECM forests rhizal networks links between tree species: however, only (Horton and Bruns, 2001), in spite of the noticeable intra-specific molecular markers, such as microsatellites absence of Cenococcum geophilum, Cortinariaceae and (Lian et al., 2006), and denser sampling could rigorously Sebacinales, which are widespread or dominant in Hol- reveal the sharing of individual ECM mycelia between tree arctic communities (e.g. Richard et al., 2005; Walker species. et al., 2005; Diédhiou et al., 2009). Similar communities, including Tricholomataceae, Boletaceae and Scleroder- mataceae, have already been reported from the African Multi-stage versus single-stage fungal taxa tropics, in Madagascar (Ducousso et al., 2004), Senegal As far as the host developmental stage is concerned, (Diédhiou et al., 2004), Western Guinea (Rivière et al., three equally frequent kinds of ECM taxa were seen: 11 2007), and the Seychelles (Tedersoo et al., 2007a). Ecto- taxa were isolated solely from seedlings, 14 were isolated mycorrhizal fungal communities from tropical Guyana only from adults and 14 were shared by the two develop- (Moyersoen, 2006) and Asia (Ingleby et al., 1998; Rivière mental stages, including the most common ones (Fig. 2). et al., 2007) tend to harbour different dominant families, In terms of abundance, these multi-stage fungi repre- perhaps as a result of distance. Our study revealed a high sented 79% of the overall ECM colonization. In well- spatial heterogeneity among the four plots analysed, due established forests, the external fungal mycelium in part to the patchy distributions of some ECM fungal supported by adult trees likely contributes to ECM com- taxa, as reported for more northern communities (Richard munity structure on seedlings (Simard et al., 1997; Walker et al., 2005), and also provided information on association et al., 2005), and ECM fungal diversity can be expected to patterns, in relationship with host age and species. be similar among trees of different ages. This was evi- denced in Mediterranean holm oak (Quercus ilex) forests Multi-host versus single-host fungal taxa (Richard et al., 2005) and in temperate coniferous forests Ectomycorrhizal host-specificity is little studied in tropical (Pinus sylvestris, Jonsson et al., 1999; Tsuga hetero- tree species. Outplanting experiments and habitat prefer- phylla, Kranabetter and Friesen, 2002). Accordingly,

Kennedy and colleagues (2003) showed that Lithocarpus old) than in older ones (400 years old; Horton et al., densiflora seedlings shared 17 of 56 ECM fungi (30%) 2005). Three non-exclusive reasons can account for the with overstorey Pseudotsuga menziesii trees, but still the lower specificity of seedlings. First, their smaller roots diversities of ECM fungi on seedlings and adults signifi- systems may limit their access to all fungal partners, cantly differed. Ectomycorrhizal communities can thus especially in such patchy ECM fungal communities where change according to the developmental stage (seedling or possibly specific ECM taxa are rare. Second, a plant may adult) of host species in our forest. Similarly, Lee and be able to select among different fungal partners through- Alexander (1996) observed in Malaysian tropical rainfor- out its life cycle: it can be expected that older trees tend to ests that the ECM community on dipterocarp seedlings recruit more specific fungal species in order to avoid inter- changed within 7 months following germination. Thus, age action with surrounding species (Selosse et al., 2006), and history of hosts may influence the ECM community, a while seedlings tend to use and thus maximize this inter- feature also reported for AM fungal species in the tropics action. Third, changes over a life span could be driven by by Husband and colleagues (2002a,b). Whether the fungal competition: as recruitment of ECM fungi on root occurrence of ECM species preferring (or preferred by) systems proceeds, the increasing competition may lead to seedlings is more pronounced in tropical regions or not progressive elimination of some species, favouring more remains unclear; similarly, the role of this change in buff- specific fungal species, maybe because they are more ering changes of environmental conditions over develop- adapted to a given host. Future research on symbiotic ment (Simard et al., 1997; Husband et al., 2002a,b) interactions between plant and fungi over their life cycle in requires further study. realistic, ECM conditions (e.g. Kennedy and Bruns, 2005) will be required to understand such developmental changes, and their functional implications. Comparisons of the ECM community among host species Possible advantage of lower specificity in seedlings When seedlings and adults were considered together, there was no significant difference in terms of ECM com- Thanks to their lower specificity, seedlings may receive position among host tree species, except for A. fragrans. inocula from adults: curiously, except for C. tetraphyllum Although some host preferences were detected at the and A. fragrans, adults tend to share more ECM fungal level of fungal species or families, the magnitude of host taxa with the non-conspecific seedlings than with their effects is lower than in those observed in previously inves- own seedlings. Adults likely function as ‘nurse trees’ for tigated mixed forests (Ishida et al., 2007; Morris et al., conspecific and non-conspecific seedlings, and therefore 2008; Tedersoo et al., 2008; Smith et al., 2009). Further- promote diversity and coexistence of species in this forest more, we anticipate that the host preference patterns may stand. The selective pressures behind this apparently be related to spatial non-independence of the samples, ‘altruistic’ phenomenon are unclear, as it may be detri- low number and aggregated distribution of the adult trees mental for adults and their conspecific seedlings. and seedlings at the study site. However, the comparison However, it may contribute to the formation of the stands between ECM associates of the various host species where ECM trees dominate, as is often observed for ECM showed contrasting results, depending on the tree age. tropical forests (Hart et al., 1989; Newbery et al., 1997; The seedlings of A. macrophylla, C. tetraphyllum, P. coer- 2004; Alexander and Lee, 2005; McGuire, 2007): uleum and U. esculenta did not significantly differ in ECM perhaps, a global positive feedback of for nourishing ECM fungal associates (Table 3, Fig. 3). This, together with the fungi allows AM species to be outcompeted. presence of multistage fungi and the intermingling of adult Besides a simple inoculum effect, the connection to a roots (particularly dense, our personal observations), sup- ‘prepaid’ mycorrhizal network supported by carbon from ports the potential for mycorrhizal network formation adults may allow understorey seedlings to reduce sym- between seedlings and adults. biosis costs and thus compete better against ECM and The adult trees hosted a more specific fungal commu- AM adults (Molina et al., 1992; Cullings et al., 2000). nity than seedlings (Table 3), linked to a reduction of Indeed, there is good evidence from temperate forests fungal diversity at the expense of the multi-host species. that overstorey trees contribute more to ECM fungal Thus, seedlings are generalist, but tend to specialize nutrition (Högberg et al., 1999). Interestingly, the survival more in single-host ECM fungi when they become older. and establishment rates of seedlings in tropical forests In a Japanese conifer-broadleaf ECM forest, Abies depend on the rate at which they become mycorrhizal homolepis is colonized by more specific ECM fungi many (Alexander et al., 1992; Newbery et al., 2000). Seedlings years after its establishment (Ishida et al., 2007); in mixed in contact with adult ECM trees often have better sur- western hemlock and Douglas-fir stands the two tree vival and mycorrhizal colonization than seedlings iso- species share more ECM fungi in young forests (40 years lated from adults, in tropical (Onguene and Kuyper,

2002) and temperate forests (Dickie et al., 2002). In previously observed. By contrast, d15N values obtained in Guyana’s forest, access to mycorrhizal networks also Entada gigas were almost nil, thus confirming its nodula- increases survival and growth of ECM seedlings tion and N2-fixing ability previously reported (Diabaté (McGuire, 2007). et al., 2005) for about 95% of Mimosoideae species. No Strikingly, only A. fragrans seedlings had a significantly significant difference in d15N was recorded between seed- different community as compared with other species lings and adults of ECM plants, again suggesting that (Table 3). Although it may persist in older stands, A. fra- ECM seedlings did not receive carbon via their fungi. grans is the only species under study that can easily Interestingly, the non-legume AM N. laevis had signifi- establish alone, earlier in the colonisation succession. cantly higher d15N values for seedlings than adults. Differ- Thus, its seedlings are adapted to grow without the assis- ences in total N content between N. laevis seedlings and tance of any mycorrhizal networks. Additionally, due to its adults could explain different d15N, but no such difference large seed reserves, A. fragrans might be less dependent was observed here. Thus, together with a higher d13C, we upon ECM fungi at early stages, as reported for other cannot rule out a C transfer to N. laevis seedlings by AM large-seeded species (Allsopp and Stock, 1992; Zangaro symbionts, although other explanations may account for et al., 2000). these features.

No evidence for nutrient transfer from adult ECM trees Outline to seedlings Our results, with at least a half of multi-host ECM fungi The potential for mycorrhizal networks opened the possi- which are the most abundant ECM types, showed a bility of some nutrient exchanges. As expected, d13C potential for mycorrhizal network formation, perhaps values of fungal fruit bodies were significantly higher than underestimated due to the sampling design. Moreover, those of plants (c. +4‰), as a result of an isotopic frac- seedlings tended to be more generalist and adults more tionation between donor trees and ECM fungi (Högberg specific. Although no carbon transfer to seedlings was et al., 1999; Boström et al., 2008). Thus, seedlings receiv- detected, connecting to pre-existing mycorrhizal networks ing carbon from fungi can be expected to have higher d13C may allow seedlings to spare carbon. Therefore, the exist- values than adult leaves, and closer to ECM fungi, ence of mycorrhizal networks and individual fungal because transfer to receiver plants usually does not entail sharing, as well as their impact on growth and fitness any shift in d13C (Julou et al., 2005; Tedersoo et al., of seedlings, remain to be rigorously assessed in this 2007b). However, the absence of significant difference in ecosystem. d13C between seedlings and adults suggests that both share indistinguishable sources of carbon (Fig. 4), with no carbon transfer to ECM seedlings. This result must be Experimental procedures interpreted with caution, as there are considerable inter- specific variations in d13C signatures among ECM adults Site description (see Fig. 4), and there is a possibility that an undetectable South of Guinea is one of the last regions of West Africa transfer (thus of debatable role) occurs. Isotopic methods retaining a primary tropical rainforest, at elevations ranging are powerful, because they integrate carbon metabolism from 500 to 1750 m. The research was conducted in the over long periods, but they have low resolution for low typical evergreen Ziama rainforest near Sérédou (8°51N, carbon flux – however, here, even a few per cent of the 9°31W, altitude 600 m), characterized by a mean annual whole carbon budget would be relevant, due to heavy rainfall of 3000 mm and a short dry season from January to competition for light. March Temperatures are in general above 24°C with relative humidity up to 80%. Soils are ferralitic, acid and very poor in The mean d15N values did not differ between ECM available phosphorus. The canopy is dominated by C. tetra- legume, AM legume and non-AM legume species, an phyllum (ECM Caesalpinioideae, Fabales), as well as the AM unexpected feature already reported for tropical rainfor- trees Erythrophleum ivorense (Fabaceae) and Heritiera utilis ests (Högberg and Alexander, 1995). Among legume (Malvaceae, Malvales), with rare Piptadeniastrum africanum species, the high d15N values obtained in A. macrophylla, (AM Fabaceae), A. fragrans (ECM Caesalpinioideae) and C. tetraphyllum and P. coeruleum were consistent with Uapaca spp. (ECM Phyllanthaceae, Malpighiales). Canopy trees are at least 30 m tall, with some emergent individuals their non-N2-fixing status considering that they were never found to be nodulated in a study performed in the same reaching 50–60 m in height (Diabaté et al., 2005). We restricted our study to an area dominated by five ECM trees, forest area, as all Caesalpinioideae ranged in the Amher- i.e. four Caesalpinioideae [A. fragrans (Af), A. macrophylla stieae tribe in general (Diabaté et al., 2005). Similarly, the (Am), C. tetraphyllum (Ct), P. coeruleum (Pc)] and the Phyl- high d15N value obtained in B. occidentalis belonging to lanthaceae U. esculenta (Ue), with abundant ECM seedlings the Caesalpinieae tribe confirmed its inability to fix N2 as (Fig. 1).

Sampling of ectomycorrhizae cooled on ice, loaded onto an 8% non-denaturing polyacry- lamide gel and stained with ethidium bromide after migration. Ectomycorrhizae were sampled in July 2003 from the five The SSCP analysis always confirmed the identification 2 ECM host species, in a 1600 m area subdivided into 16 plots obtained by root tracing (data not shown). of 100 m2 each (A to P, Fig. 1), containing 20 adult trees and 332 seedlings (defined as less than 50 cm in height and 1.5 cm in stem diameter) that were individually mapped. For Isotopic analyses the adult trees, whenever possible root tips were sampled around each tree by tracing roots from the base of trunk. Ten Plant leaves were collected from two distant plots (B and L, to 30 root samples were randomly harvested per species, Fig. 1) in March 2006, at 10–30 cm above soil (to avoid from 0 to 35 cm deep (Table 1). The seedling root tips (on a isotope distortion due to CO2 from soil respiration) and in 4–5 cm tap root each) were simultaneously collected at no similar light conditions (because low light supply results in more than 50 cm from each adult root sampling point in order slower photosynthesis, and thus higher discrimination to avoid spatial heterogeneity (Table 1). All root samples were against 13C; Julou et al., 2005). We collected leaves from (i) kept in cool conditions (4°C) until arrival at the laboratory 1–3 ECM legumes (A. macrophylla, C. tetraphyllum and P. coer- days after field collection. They were then gently separated uleum) on plot L and (ii) the same ECM legumes and U. from surrounding soil and rinsed in a sieve (0.5 mm diameter) esculenta on plot B. As controls, AM legume trees (B. occi- with tap water. Ectomycorrhizal tips were observed under a dentalis as non-N2-fixing species on both plots and Entada binocular microscope, carefully separated from roots, and gigas as N2-fixing tree species on plot L) were sampled. We kept for subsequent molecular analyses (Table 1). Ectomyc- also collected leaves from non-legume AM trees, namely orrhizal tips were placed on a cotton layer in 10 ml tubes Pentadesma butyracea (Clusiaceae) and N. laevis (Bignoni- half-filled with silica gel, and stored at room temperature. aceae) on plot L as well as Bosquiea angolensis (Moraceae) and Trichoscypha longifolia (Anacardiaceae) on plot B. Sam- pling was also extended to places surrounding the plots, to Molecular identification of fungi and plants from ECM allow the collection of leaves on independent adults. In addition, fruit bodies of the eight most abundant fruiting taxa of Total DNA was isolated from each ECM tip using a DNAeasy which seven ECM fungi (Amanita griseoflocosa, Amanita sp., Plant mini-kit following the manufacturer’s recommendations Boletus sp., Cantharellus sp., Russula sp.1, Russula sp.2, (Qiagen, Courtaboeuf, France). Identification of ECM fungi Scleroderma sp.) and one non-ECM fungus (Hygrocybe sp.) was performed for all ECM tips by amplifying and sequencing were collected in 2005–2006 to estimate their isotopic con- the nuclear rDNA ITS with the fungus-specific primer pair tents. For each sampled plant species, when available, one ITS1f–ITS4. Whenever several amplification products were leaf collected on six different adults or seedlings were dried at obtained from the same ECM tip, each was extracted from 50°C for 48 h. Samples were handled as in Tedersoo and agarose gel and re-amplified separately. Fungal DNA ampli- colleagues (2007b), and isotope abundances were fication was performed as described in Diédhiou and expressed in d13C and d15N values in parts per thousand colleagues (2004). relative to international standards for measurements of

The ITS region was sequenced using the PCR primers; V-PDB and atmospheric N2: forward and reverse DNA sequences were aligned to δδ13 15 =−()× [] produce a consensus DNA sequence. All sequences were C or N Rsample R standard 1 1000‰ , aligned with the Clustal X program (Thompson et al., 1997). where R is the molar ratio, i.e. 13C/12Cor15N/14N. Alignments were manually optimized with the Genedoc program (Nicholas et al., 1997). Whenever two sequences differed by one base only, amplification and sequencing were Statistical analyses repeated from the samples to confirm the dissimilarity. Indi- vidual ECM tips generating two different ITS sequences were The relative abundance of each ECM taxon (as identified by considered as harbouring two distinct ECM fungi. In order to ITS sequence) was estimated as the percentage of its occur- determine their taxonomical affiliation, sequences were com- rence in each or all tree species. The genotypic differentiation pared with the GenBank database using the BLASTN algorithm test was performed to test putative differences among fungal (Altschul et al., 1990). Ectomycorrhizal taxon names were communities between each host species (or between seed- designated following the BLAST score. The identity values lings and adults), using the Genepop program (Raymond and were often Յ 95% (Table 2), therefore only families were Rousset, 1995). The P-values were estimated from a log- considered here. likelihood-based exact test. The principle is the same as that Confirmation of roots identity for samples from adult trees of the Fisher exact test (Goudet et al., 1996). Each P-value was necessary, due to root intermingling. This was conducted was calculated as the sum of the probabilities of all tables by single strand conformational polymorphism (SSCP) analy- (with the same marginal values as the observed one) with a sis of the trnL intron region of the chloroplast DNA, comparing probability lower than or equal to the observed table. To DNA from ECM and leaves of the five host species. Introns compare species accumulation and richness estimates were amplified with primers trnL-c and trnL-d (Taberlet et al., among the various plant cohorts, rarefaction curves with 95% 1991). Each purified PCR product (1–3 ml) was mixed with confidence intervals were calculated using the EstimateS 7–9 ml of a denaturing buffer (95% formamide, 4% EDTA, software version 8.0.0 (Colwell, 2006). This corrects for dif- 0.05% bromophenol blue and 0.05 xylene cyanol). The ferences in sampling size and allows virtual reduction of the mixture (10 ml) was heated for 3 min at 98°C, subsequently size of all samples to that of the smallest one. The Shannon

© 2010 Society for Applied Microbiology and Blackwell Publishing Ltd, Environmental Microbiology, 12, 2219–2232 2230 A. G. Diédhiou et al. index (Shannon and Weaver, 1949) was also calculated to Mycorrhizas in Ecosystems. Read, D.J., Fitter, D.H., and estimate the fungal diversity sampled from adults and seed- Alexander, I.J. (eds). Wallingford, UK: CAB International, lings, using the EstimateS software. pp. 59–64. Factorial correspondence analysis was performed to visu- Altschul, S.F., Gish, W., Miller, W., Myers, E.W., and Lipman, alize the relationships between ECM fungal communities and D.J. (1990) Basic local alignment search tool. J Mol Biol their hosts using the XLSAT™ software package (version 215: 403–410. 2009, Addinsoft, Paris, France). This involved the construc- Bertout, S., Renaud, F., Swinne, D., Mallie, M., and Bastide, tion of a contingency table (host trees versus fungal taxa), in J.-M. (1999) Genetic multilocus studies of different strains which fungal taxa were represented in terms of their occur- of Cryptococcus neoformans: taxonomy and genetic struc- rences (frequency) on both life stages of the five host ture. J Clin Microbiol 37: 715–720. species. The calculation of variance is based on the c2 test, Boström, B., Comstedt, D., and Ekblad, A. (2008) Can isoto- which measures the extent to which a sample distribution pic fractionation during respiration explain the 13C-enriched deviates from a theoretical distribution and is an integral part sporocarps of ectomycorrhizal and saprotrophic fungi? of the correspondence analysis. Factorial correspondence New Phytol 177: 1012–1019. analysis was presented as a plane projection of the two most Bruns, T.D., Bidartondo, M.I., and Taylor, D.L. (2002) Host informative axes accounting for the ECM community compo- specificity in ectomycorrhizal communities: what do the sition (Bertout et al., 1999). exceptions tell us? Integr Comp Biol 42: 352–359. The comparison of ECM composition among the plots was Colwell, R.K. (2006) Estimates: Statistical Estimation of restricted to four plots (B, G, L and O) that included the Species Richness and Shared Species from Samples, maximum number of individuals and species of ECM hosts Version 8 [WWW document]. URL http://purl.oclc.org/ with the most homogeneous distribution (Fig. 1; each estimates. selected plot was at least 10 m away from the others). Ecto- Cullings, K.W., Vogler, D.R., Parker, V.T., and Finley, S.K. mycorrhizal roots sampled from these four plots were indi- (2000) Ectomycorrhizal specificity patterns in a mixed vidually mapped. The similarity of the ECM composition Pinus contorta and Picea engelmannii forest in Yellowstone among the plots was compared using the Sørensen coeffi- National Park. Appl Environ Microbiol 66: 4988–4991. cient (Krebs, 1999), calculated for all ECM sampled from Diabaté, M., Munive, A., de Faria, S.M., Bâ, A.M., Dreyfus, each plot irrespective of the development stage. To check B., and Galiana, A. (2005) Occurrence of nodulation in whether ECM fungi exhibit some trend for plot preference, we unexplored leguminous trees native to the West African used a c2 test of independence for the number of occurrences tropical rainforest and inoculation response of native of each ECM taxon in each selected plot. species useful in reforestation. New Phytol 166: 231–239. For isotope analysis, total N concentrations, d13C and d15N Dickie, I.A., Koide, R.T., and Steiner, K.C. (2002) Influences values were tested for normality and homogeneity of vari- of established trees on mycorrhizas, nutrition, and growth ances using a Wilk-Shapiro W-test and a Levene test respec- of Quercus rubra seedlings. Ecol Monogr 72: 505–521. tively. One-way analysis of variance (ANOVA) was separately Dickie, I.A., Guza, R.C., Kazewski, S.E., and Reich, P.B. performed for each variable and site combined with the mul- (2004) Shared ectomycorrhizal fungi between a herba- tiple range test to distinguish homogeneous mean groups at ceous perennial (Helinathemum bicknellii) and oak a=0.05. (Quercus). New Phytol 164: 375–382. Diédhiou, A.G., Bâ, A.M., Sylla, S.Nd., Dreyfus, B., Neyra, M., and Ndoye, I. (2004) The early-stage ectomycorrhizal Acknowledgements Thelephoroid fungal sp. is competitive and effective on Afzelia africana Sm. in nursery conditions in Senegal. Myc- We thank Jean Garbaye and Thomas W. Kuyper for their orrhiza 14: 313–322. insightful comments on and discussion of earlier versions of Diédhiou, A.G., Guèye, O., Diabaté, M., Prin, Y., Duponnois, this manuscript, Alexandre Geoffroy for help in isotope analy- R., Dreyfus, B., and Bâ, A.M. (2005) Contrasting sis, and Fabrice Elegbede for help in statistical analyses. We responses to ectomycorrhizal inoculation in seedlings thank the anonymous referees for their valuable comments of six tropical African tree species. Mycorrhiza 16: 11–17. on this paper, and David Marsh for improving the English. Diédhiou, A.G., Dupouey, J.-L., Buée, M., Dambrine, E., Laüt, L., and Garbaye, J. (2009) Response of ectomycor- References rhizal communities to past Roman occupation in an oak forest. Soil Biol Biochem 41: 2206–2213. Alexander, I.J., and Lee, S.S. (2005) Mycorrhizas and eco- Ducousso, M., Béna, G., Bourgeois, C., Buyck, B., system processes in tropical rain forest: implications for Eyssartier, G., Vincelette, M., et al. (2004) The last diversity. In Biotic Interactions in the Tropics: Their Role in common ancestor of Sarcolaenaceae and Asian diptero- the Maintenance of Species Diversity. Burslem, D.F.R.P., carp trees was ectomycorrhizal before the India– Pinard, M.A., and Hartley, S.E. (eds). London, UK: Cam- Madagascar separation, about 88 million years. Mol Ecol bridge University Press, pp. 165–203. 13: 231–236. Alexander, I.J., Ahmad, N., and Lee, S.S. (1992) The role of Fitter, A.H. (2001) Specificity, links and networks in the mycorrhizas in the regeneration of some Malaysian forest control of diversity in plant and microbial communities. In trees. Philos Trans R Soc Lond B 335: 379–388. Ecology: Achievement and Challenge. Press, M.C., Huntly, Allsopp, N., and Stock, W.D. (1992) Mycorrhizas, seed size N.J., and Levin, S. (eds). Oxford, UK: Blackwell Scientific, and seedling establishment in low nutrient environment. In pp. 95–114.

Goudet, J., Raymond, M., De Meeüs, T., and Rousset, F. networks between understorey and canopy trees in a (1996) Testing differentiation in diploid populations. Genet mixed evergreen forest. J Ecol 91: 1071–1080. 144: 933–1940. Kranabetter, J.M., and Friesen, J. (2002) Ectomycorrhizal Hart, T.B., Hart, J.A., and Murphy, P.G. (1989) Mono- community structure on western hemlock seedlings trans- dominant and species-rich forests of the humid tropics: planted from forests into openings. Can J Bot 80: 861–868. causes for their occurrence. Am Nat 133: 613–633. Krebs, C.J. (1999) Ecological Methodology. Menlo Park, CA, Högberg, P., and Alexander, I.J. (1995) Roles of root sym- USA: Benjamin/Cummings. bioses in African woodland and forest: evidence from 15N Lee, S.S., and Alexander, I.J. (1996) The dynamics of ecto- abundance and foliar analysis. J Ecol 83: 217–224. mycorrhizal infection of Shorea leprosula seedlings in Högberg, P., and Piearce, G.D. (1986) Mycorrhizas in Malaysian rain forests. New Phytol 132: 297–305. Zambian trees in relation to host taxonomy, vegetation type Lian, C., Goto, S., and Hogetsu, T. (2006) Microsatellite and successional patterns. J Ecol 74: 775–785. markers for Sachalin fir (Abies sachalinensis Masters). Mol Högberg, P., Plamboeck, A.H., Taylor, A.F.S., and Fransson, Ecol Notes 7: 896–898. P.M.A. (1999) Natural C-13 abundance reveals trophic McGuire, K.L. (2007) Common ectomycorrhizal networks status of fungi and host-origin of carbon in mycorrhizal may maintain monodominace in a tropical rain forest. fungi in mixed forests. PNAS 96: 8534–8539. Ecology 88: 567–574. Horton, T.R., and Bruns, T.D. (1998) Multiple host fungi are Molina, R., Massicotte, H.B., and Trappe, J.M. (1992) Speci- the most frequent and abundant ectomycorrhizal types in a ficity phenomena in mycorrhizal symbioses: community- mixed stand of Douglas fir (Pseudotsuga menziesii (Mirb.) ecological consequences and practical implications. In Franco) and bishop pine (Pinus muricata D. Don). New Mycorrhizal Functioning: An Integrative Plant-Fungal Phytol 139: 331–339. Process. Allen, M.F. (ed.). New York, USA: Chapman and Horton, T.R., and Bruns, T.D. (2001) The molecular revolution Hall, pp. 357–423. in ectomycorrhizal ecology: peeking into the black-box. Mol Morris, M.H., Smith, M.E., Rizzo, D.M., Rejmanek, M., and Ecol 10: 1855–1871. Bledsoe, C. (2008) Contrasting ectomycorrhizal fungal Horton, T.R., Molina, R., and Hood, K. (2005) Douglas-fir communities on the roots of co-occurring oaks (Quercus ectomycorrhizae in 40- and 400-year-old stands: myco- spp.) in a California woodland. New Phytol 178: 167–176. biont availability to late successional western hemlock. Moyersoen, B. (2006) Pakaraimea dipterocarpacea is ecto- Mycorrhiza 15: 393–403. mycorrhizal, indicating an ancient Gondwanaland origin for Husband, R., Herre, E.A., Turner, L., Gallery, R., and Young, the ectomycorrhizal habit in Dipterocarpaceae. New Phytol J.P.W. (2002a) Molecular diversity of arbuscular mycor- 172: 753–762. rhizal fungi and patterns of host association over time and Nara, K. (2006) Ectomycorrhizal networks and seedling space in a tropical forest. Mol Ecol 11: 2669–2678. establishment during early primary succession. New Phytol Husband, R., Herre, E.A., and Young, J.P.W. (2002b) Tem- 169: 169–178. poral variation in the arbuscular mycorrhizal communities Newbery, D.M., Alexander, I.J., Thomas, D.W., and Gartlan, colonising seedlings in a tropical forest. FEMS Microbiol J.S. (1988) Ectomycorrhizal rain-forest legumes and soil Ecol 42: 131–136. phosphorus in Korup National Park, Cameroon. New Ingleby, K., Munro, R.C., Noor, M., Mason, P.A., and Clear- Phytol 109: 433–455. water, M.J. (1998) Ectomycorrhizal populations and growth Newbery, D.M., Alexander, I.J., and Rother, J.A. (1997) of Shorea parvifolia (Dipterocarpaceae) seedlings regen- Phosphorus dynamics in a lowland African rain forest: the erating under three different forest canopies following influence of ectomycorrhizal trees. Ecol Monogr 67: 367– logging. For Ecol Manage 111: 171–179. 409. Ishida, T.A., Nara, K., and Hogetsu, T. (2007) Host effects on Newbery, D.M., Alexander, I.J., and Rother, J.A. (2000) Does ectomycorrhizal fungal communities: insight from eight proximity to conspecific adults influence the establishment host species in mixed conifer–broadleaf forests. New of ectomycorrhizal trees in rain forest? New Phytol 147: Phytol 174: 430–440. 401–409. Jonsson, L., Dahlberg, A., Nilsson, M.C., Karen, O., and Newbery, D.M., van der Burgt, X.M., and Moravie, M.-A. Zackrisson, O. (1999) Continuity of ectomycorrizal fungi in (2004) Structure and inferred dynamics of a large grove of self-regenerating boreal Pinus sylvestris forests studied by Microberlinia bisulcata trees in central African rain forest: comparing mycobiont diversity on seedlings and mature the possible role of periods of multiple disturbance events. trees. New Phytol 142: 151–162. J Trop Ecol 20: 131–143. Julou, T., Burhardt, B., Gebauer, G., Berviller, D., Damesin, Nicholas, K.B., Nicholas, H.B.J., and Deerfield, D.W. (1997) C., and Selosse, M.-A. (2005) Mixotrophy in orchids: GeneDoc: analysis and visualization of genetic variation. insights from a comparative study of green individuals and The European Mol Biology Network Newsletter 4: 14. non-photosynthetic mutants of Cephalanthera damaso- Onguene, N.A. (2000) Diversity and Dynamics of Mycorrhizal nium. New Phytol 166: 639–653. Associations in Tropical Rain Forests with different Distur- Kennedy, P.G., and Bruns, T.D. (2005) Priority effects deter- bance Regimes in South Cameroon. Tropenbos- mine the outcome of ectomycorrhizal competition between Cameroon Series 3. two Rhizopogon species colonizing Pinus muricata seed- Onguene, N.A., and Kuyper, T.W. (2002) Importance of the lings. New Phytol 166: 631–638. ectomycorrhizal network for seedling survival and ectomy- Kennedy, P.G., Izzo, A.D., and Bruns, T.D. (2003) There is corrhiza formation in rain forests of south Cameroon. Myc- high potential for the formation of common mycorrhizal orrhiza 12: 13–17.

Raymond, M., and Rousset, F. (1995) GENEPOP (version ectomycorrhizal fungi in a Tasmanian wet sclerophyll forest 1.2): population genetics software for exact tests and ecu- as revealed by DNA barcoding and taxon-specific primers. menicism. J Hered 86: 248–249. New Phytol 180: 479–490. Richard, F., Millot, S., Gardes, M., and Selosse, M.-A. (2005) Tedersoo, L., Suvi, T., Jairus, T., Ostonen, I., and Põlme, S. Diversity and specificity of ectomycorrhizal fungi retrieved (2009) Revisiting ectomycorrhizal fungi of the genus Alnus: from an old-growth Mediterranean forest dominated by differential host specificity, diversity and determinants of Quercus ilex. New Phytol 166: 1011–1023. the fungal community. New Phytol 182: 727–735. Rivière, T., Diédhiou, A.G., Diabaté, M., Senthilarasu, G., Thoen, D., and Bâ, A.M. (1989) Ectomycorrhizae and putative Natarajan, K., Verbeken, A., et al. (2007) Genetic diversity ectomycorrhizal fungi of Afzelia africana and Uapaca of ectomycorrhizal Basidiomycetes from African and Indian guineensis in Southern Senegal. New Phytol 113: 549–559. tropical rain forests. Mycorrhiza 17: 415–428. Thompson, J.D., Gibson, T.J., Plewniak, F., Jeanmorgin, F., Sanon, K.B., Bâ, A.M., and Dexheimer, J. (1997) Mycorrhizal and Haggins, D.G. (1997) The clustalx windows, interface: status of somefungi fruiting beneath indigenous trees in flexible, strategies for multiple sequence alignment aided Burkina Faso. For Ecol Manage 98: 61–69. by quality analysis tools. Nucleic Acid Res 25: 4876–4882. Selosse, M.-A., and Roy, M. (2009) Green plants that feed on Torti, S.D., and Coley, P.D. (1999) Tropical monodominance: fungi: facts and questions about mixotrophy. Trends Plant a preliminary test of the ectomycorrhizal hypothesis. Bio- Sci 14: 64–70. tropica 31: 220–228. Selosse, M.-A., Richard, F., He, X., and Simard, S.W. (2006) Walker, J.F., Miller, J.R.O., and Horton, J.L. (2005) Hyperdi- Mycorrhizal networks: des liaisons dangereuses? Trends versity of ectomycorrhizal fungus assemblages on oak Ecol Evol 21: 621–628. seedlings in mixed forests in the southern Appalachian Shannon, C.E., and Weaver, W. (1949) The Mathematical Mountains. Mol Ecol 14: 829–838. Theory of Communication. Urbana, IL, USA: University of Wu, B.Y., Nara, K., and Hogetsu, T. (2001) Can C-14-labeled Illinois Press. photosynthetic products move between Pinus densiflora Simard, S.W., Perry, D.A., Smith, J.E., and Molina, R. (1997) seedlings linked by ectomycorrhizal mycelia? New Phytol Effects of soil trenching on occurrence of ectomycorrhizas 149: 137–146. on Pseudotsuga menziesii seedlings grown in mature Zangaro, W., Bononi, V.L.R., and Truffen, S.B. (2000) Myc- forests of Betula papyrifera and Pseudotsuga menziesii. orrhizal dependency, inoculum potentiel and habitat pref- New Phytol 136: 327–340. erence of native woody species in South Brazil. J Trop Ecol Smith, M.E., Douhan, G.W., Fremier, A.K., and Rizzo, D.M. 16: 603–622. (2009) Are true multihost fungi the exception or the rule? Dominant ectomycorrhizal fungi on Pinus sabiniana differ from those on co-occurring Quercus species. New Phytol Supporting information 182: 295–299. Smith, S.E., and Read, D.J. (2008) Mycorrhizal Symbiosis. Additional Supporting Information may be found in the online Ed 3. London, UK: Academic press. version of this article: Smits, W. (1994) Dipterocarpaceae: mycorrhizae and regeneration. Tropenbos Series 9: 1–243. Fig. S1. Species accumulation curves of ectomycorrhizal Taberlet, P., Gielly, L., Pautou, G., and Bouvet, J. (1991) fungi and their 95% confidence intervals (pointed lines with Universal primers for amplification of three non-coding smaller symbols on the right end); (A) samples from adult regions of chloroplast DNA. Plant Mol Biol 17: 1105– trees (B) samples from seedlings. 1109. Fig. S2. Schematic map of spatial distribution of fungal taxa Tedersoo, L., Suvi, T., Beaver, K., and Kõljalg, U. (2007a) found on ectomycorrhizas (ECMs) collected from the five Ectomycorrhizal fungi of the Seychelles: diversity host species, A. fragrans (Af), A. macrophylla (Am), C. tetra- patterns and host shifts from the native Vateriopsis sey- phyllum (Ct), P. coeruleum (Pc) and U. esculenta (Ue) chellarum (Dipterocarpaceae) and Intsia bijuga (Caesal- growing in the four analysed plots (B, G, L and O). The area piniaceae) to the introduced Eucalyptus robusta of each plot was 100 m2. Adult trees are indicated in red, (Myrtaceae), but not Pinus caribea (Pinaceae). New ECMs collected from adult trees are represented with a red Phytol 175: 321–333. border, those collected from seedlings with black border. Tedersoo, L., Pellet, P., Kõljalg, U., and Selosse, M.-A. Fungal taxa are represented by different colour shadings. (2007b) Parallel evolutionary paths to mycoheterotrophy in understorey Ericaceae and Orchidaceae: ecological evi- Please note: Wiley-Blackwell are not responsible for the dence for mixotrophy in Pyroleae. Oecologia 151: 206–217. content or functionality of any supporting materials supplied Tedersoo, L., Jairus, T., Horton, B.M., Abarenkov, K., Suvi, T., by the authors. Any queries (other than missing material) Saar, I., and Kõljalg, U. (2008) Strong host preference of should be directed to the corresponding author for the article.