Ectomycorrhizal (ECM) plants and fungi can be diverse and abundant in certain tropical ecosystems. For example, the primarily paleotropical ECM plant family Dipterocarpaceae is one of the most speciose and ecologically important tree families in Southeast Asia. Pakaraimaea dipterocarpacea is one of two species of dipterocarp known from the Neotropics, and is also the only known member of the monotypic Dipterocarpaceae subfamily Pakaraimoideae. This Guiana Shield endemic is only known from the sandstone highlands of Guyana and Venezuela. Despite its unique phylogenetic position and unusual geographical distribution, the ECM fungal associations of P. dipterocarpacea are understudied throughout the tree’s range. In December 2010 we sampled ECM fungi on roots of P. dipterocarpacea and the co-occurring ECM tree Dicymbe jenmanii (Fabaceae subfamily Caesalpinioideae) in the Upper Mazaruni River Basin of Guyana. Based on ITS rDNA sequencing we documented 52 ECM species from 11 independent fungal lineages. Due to the phylogenetic distance between the two host tree species, we hypothesized that P. dipterocarpacea would harbor unique ECM fungi not found on the roots of D. jenmanii. Although statistical tests suggested that several ECM fungal species did exhibit host preferences for either P. dipterocarpacea or D. jenmanii, most of the ECM fungi were multi-host generalists. We also detected several ECM fungi that have never been found in long-term studies of nearby rainforests dominated by other Dicymbe species. One particular mushroom-forming fungus appears to be unique and may represent a new ECM lineage of Agaricales that is endemic to the Neotropics.
Citation: Smith ME, Henkel TW, Uehling JK, Fremier AK, Clarke HD, Vilgalys R (2013) The Ectomycorrhizal Fungal Community in a Neotropical Forest Dominated by the Endemic Dipterocarp Pakaraimaea dipterocarpacea. PLoS ONE 8(1): e55160. https://doi.org/10.1371/journal.pone.0055160
Editor: Paul V. A. Fine, University of California, Berkeley, United States of America
Received: August 13, 2012; Accepted: December 21, 2012; Published: January 31, 2013
Copyright: © 2013 Smith et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Funding: Funding was provided by the National Science Foundation grants DEB-0918591 to TWH and DEB-3331108 to RV. Participation by MES was also possible in part due to support from University of Florida’s Institute for Food and Agricultural Sciences (IFAS). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors have declared that no competing interests exist.
Ectomycorrhizal (ECM) fungi are a diverse functional group of mutualistic root symbionts that enhance host plant nutrient acquisition, protect against root disease, and mitigate the effects of abiotic stresses , . The ECM symbiosis was historically considered to be restricted to the temperate regions of the world where many forests are dominated by ECM plants. However, evidence has steadily accumulated over the last 50 years that ECM plants and fungi are present in most tropical ecosystems. Tropical ECM plants are most often present at low densities in plant communities dominated by arbuscular mycorrhizal plants, but at specific tropical sites ECM plants can be dominant components of the vegetation , , . The recognition that ECM plants are widely distributed in the tropics has fostered a growing interest in their symbiotic ECM fungi. Because tropical habitats are often challenging to access, there are still major gaps in our understanding of the ecology, biogeography, and host preferences of tropical ECM fungi and plants. Several recent studies have suggested that tropical forests harbor limited ECM fungal diversity with either few or no endemic ECM fungal lineages . In contrast, other tropical studies have detected relatively high ECM fungal diversity and presented evidence that at least some ECM fungal lineages originated from or diversified in the tropics , , .
In the Neotropics, several unrelated plant genera have independently evolved the ability to form ECM symbioses with fungi: Pakaraimaea (Dipterocarpaceae), Quercus (Fagaceae), Coccoloba (Polygonaceae), Aldina (Fabaceae subfamily Papilionoideae), Dicymbe (Fabaceae subfamily Caesalpinioideae), Gnetum (Gnetaceae) and at least three genera in the Nyctaginaeae (Pisonia, Neea, and Guapira) , , , , , . These primarily lowland neotropical ECM plants are highly variable in terms of growth habit and geographic distribution. For example, species of Nyctaginaceae, Coccoloba, and Gnetum are shrubs, small trees, or lianas widely distributed at low densities in many forest types ,  whereas species of Quercus and Dicymbe are canopy trees that tend to dominate stands but have more restricted geographic distributions , . Lowland regional distributions of these ECM plant genera range from pantropical (Gnetum) to neotropical (Coccoloba) to Central American (Quercus) to Guiana Shield-endemic (Dicymbe, Aldina, Pakaraimaea). The ECM fungi associated with most of these plant genera have been characterized to various degrees, but to date the fungal symbionts of Gnetum and Pakaraimaea have been insufficiently studied.
The angiosperm family Dipterocarpaceae is one of the most ecologically and economically important tropical ECM plant lineages , , , . This primarily Old World family contains more than 500 species and many are large, emergent trees that dominate forests in Southeast Asia and to a lesser extent in Africa . The mycorrhizal biology of dipterocarps was recently reviewed by Brearley . The family Dipterocarpaceae was considered restricted to the paleotropics until the latter 20th century when the monotypic genera Pakaraimaea and Pseudomonotes were described from the Guiana Shield region of South America , . Pseudomonotes tropenbosii Londoño, Alvarez & Forero (Dipterocarpaceae subfamily Monotoideae) is known only from southeastern Colombia  and was recently shown to be associated with sporocarps of putatively ECM fungi . Pakaraimaea is represented by one species (P. dipterocarpacea Maguire & P. S. Ashton) with two subspecies (nitidum and dipterocarpacea), and forms dense stands of coppicing trees in savanna-fringing forests in the sandstone uplands of Guyana and Venezuela , . Pakaraimaea dipterocarpacea is the only known member of Dipterocarpaceae subfamily Pakaraimoideae, an intermediate lineage between subfamilies Dipterocarpoideae and Monotoideae , .
Despite the unique phylogenetic position and unusual distribution of P. dipterocarpacea, relatively little is known about the ECM fungal communities associated with this tree. Moyersoen  studied roots of several individuals of P. dipterocarpacea ssp. nitidum in Venezuela and confirmed their association with seven species of ECM fungi. In a follow up study, Moyersoen  identified an additional 31 species of ECM fungi based on collections of ECM roots and sporocarps, but unfortunately molecular data are unavailable for many of these taxa. The ECM fungi associated with the Guyanese P. dipterocarpacea spp. dipterocarpacea have never been studied.
A recent study of ECM fungi in a mixed tropical rainforest in Ecuador found low fungal diversity but strong host preferences . In contrast, we recently documented high ECM fungal diversity but no apparent fungal host preferences on the three co-occurring leguminous host trees Dicymbe corymbosa Spruce ex Benth., Dicymbe altsonii Sandw., and Aldina insignis (Benth.) Endl. in the Pakaraima Mountains of Guyana . Further explorations in this region identified the nearby Pegaima savanna-forest mosaic as a habitat where the leguminous ECM tree, Dicymbe jenmanii Sandw., co-occurs with the dipterocarp P. dipterocarpacea. In portions of the savanna-fringing forest, P. dipterocarpacea grows as large, coppicing trees that dominate the canopy interspersed with medium-sized individuals of D. jenmanii that reach the mid- to upper-canopy. The co-occurrence of these two distantly related, Guiana Shield endemic, ECM-forming tree species in close proximity to our previous study sites provided a unique opportunity to further explore fungal host preferences and beta diversity of ECM fungi in this remote neotropical region.
For this study we sampled the ECM fungal communities of P. dipterocarpacea and D. jenmanii where these host plants co-occur in the Pegaima savanna. We asked the following questions: 1) Do ECM fungi exhibit marked host preferences for either one or the other plant species?, 2) Does P. dipterocarpacea host ECM fungal species not found on leguminous trees of the region? and, 3) Are the dominant ECM fungi in this dipterocarp-dominated forest different from those of the Fabaceae-dominated ECM communities in nearby rainforests?
Study Site and Host Plants
Fieldwork for this study was conducted during December 2010-January 2011 and May-June 2012 in the Upper Mazaruni River Basin in the Pakaraima Mountains of Guyana (Fig. 1). This site is located within a large complex of open savanna communities intermixed with patches of closed-canopy fringing forest on the western side of Mt. Ayanganna, the highest sandstone mountain in Guyana (2041 m). Previous observations indicated that the savanna-fringing forests were dominated by P. dipterocarpacea ssp. dipterocarpacea (hereafter P. dipterocarpacea) with D. jenmanii as a common subdominant, with a general upper canopy height of ∼20 m. We established a base camp at ∼800 m elevation at 5° 26′ 21.3′′ N; 60° 04′ 43.1″ W. This area is ca. 25 km from the rainforest sites on the eastern side of Mt. Ayanganna in the Upper Potaro River Basin where we have conducted multi-year sampling of ECM fungal sporocarps in Dicymbe-dominated forests  and belowground studies of ECM fungi with multiple leguminous host tree species . The Potaro and Pegaima sites are geographically close but vary in annual precipitation (>2400 mm and ∼2000 mm, respectively); the Pegaima site is drier due to its position within the rain shadow of Mt. Ayanganna , . Pegaima soils are highly oligotrophic white sands (entisols) derived from the Roraima Formation sandstone whereas the soils from the specific Potaro rainforest study sites are laterites derived from intrusive igneous rock (oxisols) , . Burn scars on forest edge trees indicated that the Pegaima site experiences periodic anthropogenic or natural fires, whereas fires are extremely rare or absent from the Potaro rainforests , .
Map (left) showing the central Pakaraima Mountain region of western Guyana including the Upper Potaro and Mazaruni River Basins. The tallest mountain in Guyana, (Mt. Ayanganna, 2041 m) is indicated by an X. This study was conducted in Pakaraimaea dipterocarpacea-Dicymbe jenmanii stands at 800 m elevation at the edge of the Pegaima savanna, with location indicated by A. The approximate extent and location of the savanna ecosystem is shown with a black polygon; the rest of the map area is forested. The locations of other study plots in closed-canopy Dicymbe rainforests at ∼800 m elevation along the Potaro river are indicated by B (Henkel et al. ) and C (Smith et al. ). The center photo shows an ecotone of savanna with fringing forest dominated by Pakaraimaea dipterocarpacea, Pegaima savanna, Upper Mazaruni Basin, Guyana. The photo on the right shows an individual tree of Pakaraimaea dipterocarpacea measuring 212 cm diameter at breast height.
We initially examined fertile collections of P. dipterocarpacea to determine which of the two subspecies was present. Several publications have documented the distribution and growth habit of the two subspecies in the seasonally dry areas of the Pakaraima range of Guyana and throughout the Caroni River Basin of Venezuela  and references therein]. Plant identification was based on leaf and petal morphology; P. dipterocarpacea ssp. dipterocarpacea (Guyana) has shorter leaves and glabrous petals whereas ssp. nitidum (Venezuela) has longer leaves and pubescent petals . Fertile plant voucher specimens were photographed in the field and deposited in the BRG (Guyana), US, and HSU herbaria. At the Pegaima site, P. dipterocarpacea dominates the fringing forests, composing >50 percent of the basal area and upper canopy area, with most individuals exhibiting multiple, co-dominant canopy stems and numerous sprout shoots of various sizes (Henkel, unpublished data). Previous reports suggested that P. dipterocarpacea is usually encountered as a small, shrubby tree and that large emergent individuals are rare . However, at Pegaima individual trees of P. dipterocarpacea regularly reach 15–20 m in height and 100–200 cm diameter at breast height (dbh). Consistent with previous observations, the forest floor in stands dominated by P. dipterocarpacea at Pegaima were covered by a 10–20 cm thick litter layer .
Ectomycorrhizal Root and Sporocarp Sampling
Root sampling followed protocols similar to Smith et al.  with minor changes. We identified 20 pairs of P. dipterocarpacea and D. jenmanii in which each tree was >20 cm dbh and where trees in a pair occurred ≤20 m from each other. A total of 40 trees (20 trees per species) were sampled along the edge of the Pegaima savanna in forests dominated by P. dipterocarpacea within ca. 0.3 km of base camp. Distances between sampled tree pairs ranged from 3–20 m (mean = 12.5 m). Four lateral roots from each sampled tree were traced 1–3 m from the base to the fine roots, where roots, soil, and litter were excavated and pooled until ca. 1000 cm3 of material rich in ECM roots was obtained. Roots were harvested only when unequivocally traced back to the sample tree. No molecular tests were used to verify the identity of the plant roots because our previous study showed that we could accurately determine hosts in the field as long as careful root tracing was conducted . Root samples were rinsed in water to remove soil particles and inspected under a dissecting microscope. Eight ECM roots were randomly selected from each tree. In the study by Smith et al.  root morphotyping was used prior to molecular sampling but in this study no morphological sorting was conducted. A total of 320 individual ECM roots were rapidly dried in 8-strip microcentrifuge tubes by placing them overnight in a sealed container with silica gel.
Sporocarps of putative ECM fungi were collected at the Pegaima site in December 2010 and June 2012. These were then compared to a database of sporocarps collected at two sites in Dicymbe-dominated rainforest near the Potaro River during 2000–2010 . These fungi were identified to species/morphospecies but most of the putatively new taxa known only from the Pegaima site have not yet had their ITS rDNA sequenced. For information on site, specimen identification, and herbaria accessions see Henkel et al. , .
Molecular Protocols and Fungal Identification
Molecular protocols for sequencing of ECM fungi from roots and sporocarps followed those of Smith et al. . Briefly, silica gel-dried ECM roots were rinsed in water to remove soil particles and then crushed with forceps in tubes containing 25 µl of extraction buffer from an Extract-N-Amp Plant kit (Sigma-Aldrich, St. Louis, MO, USA). Crushed roots were incubated at 96°C for 10 minutes and then mixed with 25 µl of dilution buffer. Sporocarp DNA was extracted using a CTAB protocol  or the Extract-N-Amp Plant kit (Sigma-Aldrich, St. Louis, MO, USA).
Fungal ITS rDNA was PCR-amplified with forward primer ITS1F in combination with reverse primers ITS4 or ITS4B. When amplification with these primers was unsuccessful, we used reverse primer ITS2 instead , , . PCR protocols followed Smith et al. . Amplicons were visualized on 1.5% agarose gels stained with SYBR Green I (Molecular Probes, Eugene, OR, USA). Amplicons were cleaned with EXO and SAP enzymes . Sequencing was performed with the above primers using Big Dye Sequencing Kit v.3.1 (Applied Biosystems, Foster City, CA, USA). Sequences were edited with Sequencher v.4.1 (Gene Codes Inc., Ann Arbor, MI, USA).
Sporocarps were identified based on a combination of morphological features and rDNA sequences. Taxa detected only on roots were identified to genus and their uniqueness at the species level determined using blastN comparisons against our sporocarp and ECM root sequence database as well as GenBank. Internal transcribed spacer (ITS) sequences were considered to represent the same operational taxonomic unit (OTU), a proxy for species, if they differed by <3% across the ITS region . Taxa were assigned to the phylogenetically defined ECM fungal lineages recognized by Tedersoo et al. . These lineages constitute monophyletic groups of fungi that have independently evolved the ability to form ECM symbioses with plants. ECM sequences that did not fit into these groups, did not match well with known saprotrophic fungi, and were found on multiple healthy ECM roots were putatively considered representatives of new ECM lineages.
Two common measures of diversity, the Shannon-Wiener diversity index (H’) and the Simpson’s diversity index (1-D), were calculated for ECM fungi using PC-ORD 6 . We constructed sampling curves to compare the level of ECM sampling and diversity at the Pegaima savanna and Potaro rainforest sites. PC-ORD 6 was used to generate sampling curves and standard deviations based on 500 bootstrap subsamples. The two sampling curves were then graphed using Microsoft Excel based on the number of roots sampled at each site (8 ECM roots per tree at Pegaima, 20 ECM roots per tree at Potaro). To test for statistically significant differences in frequency of individual ECM fungal species on the two different host plants we used the software package R to perform a Fischer’s exact test for each fungal species that occurred more than one time in the dataset. The Fisher exact test is a significance test used in the analysis of contingency tables with low sample sizes. PC-ORD 6 was also used for preliminary ordination analyses to examine the effects of host plants on ECM fungal communities.
Molecular and Phylogenetic Analysis of the Unique Ectomycorrhizal Fungus Agaricales TH9235
At both the Potaro and Pegaima sites we collected a large, orange-colored, tricholomatoid mushroom, with the stipitate, pileate and gilled macromorphology typical of the basidiomycete order Agaricales (voucher TH9235 from the Potaro site, voucher TH9693 from the Pegaima site). This mushroom produced a white spore print and fruited directly on soil in close proximity to P. dipterocarpacea and D. jenmanii at Pegaima and with Dicymbe spp. at Potaro. The mushroom had white hyphal cords at the base that were usually attached to clusters of white ECM roots. The taxonomic affinities of this fungus have proved particularly difficult to determine based on morphology or ITS rDNA sequences. During the course of this study, we also detected Agaricales TH9235 on ECM roots of Dicymbe jenmanii. Because of the putatively unique position of this fungus within the Agaricales and its potential as a neotropical endemic taxon, we initiated a preliminary exploration of the phylogenetic relationships of Agaricales TH9235 by sequencing portions of three additional regions: 1) 18S rDNA, 2) 28S rDNA, and 3) mtLSU. We followed the same general protocols as outlined above except we used primers ssu1536 and ssu0817 for 18S , LROR and LR5 for 28S  and ML5 and ML6 for mtLSU . Sequence data from these three genes was subjected to blastN analysis.
We also performed a phylogenetic analysis using the 28S rDNA region. This region was selected because sequence data are available for many Agaricales species and it has proved useful for resolving the phylogeny of many mushroom-forming fungi . For this analysis, we included representatives of all Agaricales genera that were similar to Agaricales TH9235 based on blastN analysis as well as several other white-spored ECM fungi from unrelated Agaricales lineages . The 28S rDNA alignment that was used for phylogenetic analysis included 88 species of Agaricales and was 1012 bp long after exclusion of ambiguously aligned DNA regions. To assess the relationships between Agaricales TH9235 and other members of the Agaricales, we performed a Maximum Likelihood analysis using the GTR+I+G model (generalized time reversible+invariant sites+Gamma distribution) with the software package GARLI . Statistical support was assessed by conducting 250 bootstrap replicates.
Ectomycorrhizal Fungal Diversity and Community Structure
A total of 160 ECM roots were randomly selected for analysis from each of the two tree species for a total of 320 ECM tips. Of these, DNA sequences of ECM fungi were successfully produced from 125 ECM tips of D. jenmanii and 130 ECM tips of P. dipterocarpacea (78% and 81% success rates, respectively). A total of 52 ITS species-level operational taxonomic units (OTUs) of ECM fungi were recovered from the roots of P. dipterocarpacea and D. jenmanii (Table 1). Both the Shannon-Wiener Index (H’) diversity value for the total species assemblage and the Simpson’s Diversity Index value of 1-D were moderately high (1.434 and 0.7446, respectively) but slightly lower than the diversity index values found in nearby rainforest sites . Seventeen frequently occurring OTUs were detected three or more times whereas 10 OTUs were found twice and 25 OTUs were detected only once. These ECM taxa represented 11 independent fungal lineages, with the majority occurring in the familiar/boletus,/russula-lactarius,/clavulina, and/tomentella-thelephora lineages (Table 1). Two OTUs (Polyporales ECM32_7 and Agaricales TH9235) did not fall into any of the known ECM lineages outlined by Tedersoo et al. . Polyporales ECM32_7 was highly similar to a polyporoid OTU detected in Guyana on healthy ECM roots with a fungal mantle and Hartig net by Smith et al.  and is therefore considered an ECM fungus. The other unknown OTU found on ECM roots (Agaricales TH9235) matched tricholomatoid mushrooms collected at both the Pegaima and Potaro sites. This species putatively represents a new ECM lineage (see below).
Overall, 40 ECM fungal OTUs were detected on roots of D. jenmanii and 28 OTUs on P. dipterocarpacea. Individual trees of D. jenmanii hosted 2–7 species (mean 5 species per tree) and individual trees of P. dipterocarpacea hosted 2–7 species (mean 4 species per tree). Sixteen ECM OTUs were shared by both tree species. While 24 ECM fungi were unique to D. jenmanii, and 13 to P. dipterocarpacea, the majority of these were detected from a single root tip. Most of the common ECM fungi (13/17) were detected on the roots of both P. dipterocarpacea and D. jenmanii and 6 of the 8 most frequently detected species had a similar number of occurrences on the two host plants (Fig. 2). Four of the 17 common ECM fungi (Cortinarius ECM 34-5 and Tomentella ECM 40-5, Tomentella ECM 34-4 and Tomentella ECM1111) were only found on the roots of D. jenmanii whereas no ECM fungi were restricted to the host plant P. dipterocarpacea. Fisher’s exact tests determined that four species (Russula TH9503, Cortinarius MCA3928, Tomentella ECM1111, and Tomentella ECM40-5) showed a statistically skewed distribution on one of the two host plants. Russula TH9503 was more common on the roots of P. dipterocarpacea but the other three taxa were more common on D. jenmanii. Preliminary ordination analyses did not show any strong patterns of host-specificity at the community level and are thus not discussed further (data not shown).
Each of these common fungal species occurred on three or more individual trees; 20 trees were sampled for each of the host tree species. Species that showed a significantly different distribution on the two host plants (as assessed by Fischer’s Exact test) are indicated by asterisks. Fungal species that have never been found in previous ECM sporocarp or root surveys in nearby rainforest sites are designated by black circles. All other ECM fungal species have been found previously in association with species of Dicymbe and Aldina at other locations. Named fungal species are indicated by a genus and species binomial whereas species with TH or MCA numbers were matched to voucher specimens of undescribed species identified to genus. The ECM numbers correspond to fungal species known only from ECM root sequences.
Fourteen of the 52 fungal OTUs documented on ECM roots corresponded with formally described species and 24 of the 52 OTUs could be matched with vouchered sporocarp specimens from the Pegaima site. Approximately 67% of the fungi detected on ECM roots in this study (35/52 OTUs) have also been detected previously at other Guyanese rainforest sites dominated by leguminous trees, either directly as sequences on ECM roots or as sporocarps in long term research plots , . Seventeen of the 52 OTUs detected on roots from the Pegaima site have not previously been found in Guyana on ECM roots or as sporocarps (Henkel & Smith, unpublished data). However, most of the common ECM species detected were fungi that have been found previously at other sites; among the 17 most common ECM fungal species at Pegaima, only four had not been detected before (Russula TH9503, Tomentella ECM40-5, Tomentella ECM34-4 and Cortinarius ECM34-5). Sampling effort curves indicated that while overall ECM fungal diversity was lower at the Pegaima site as compared with the Upper Potaro site of Smith et al. , much more sampling was needed at the Pegaima site to recover the total diversity of ECM fungi (Fig. 3).
). The two studies differed in their sampling procedure; this study was based on random sampling of eight ECM roots per tree whereas the study by Smith et al.  was based on sampling of 20 morphotyped ECM roots per tree.
Incomplete recovery of ECM fungal diversity by root sampling was corroborated by the fact that 82 species of putative or confirmed ECM fungi have now been collected as sporocarps from the same P. dipterocarpacea-D. jenmanii stands at Pegaima (Table 2). While the majority of these taxa are conspecific with species described or awaiting description from the Potaro sites , 26 are currently known only from the Pegaima sites, including species of Russula, Lactarius, Clavulina, Cortinarius, Elaphomyces, and Sarcodon. Taxonomic work on these potentially new species is ongoing.
Affinities of the Unique Species Agaricales TH9235
The sporocarp ITS rDNA sequence from Agaricales TH9235 was identical to ITS rDNA sequences from ECM roots of D. jenmanii, confirming the ECM status of this fungus (Fig. 4). While most of the ECM fungal taxa in Guyana can be identified to genus or species by morphology and/or sequence homology in the ITS and 28s rDNA, this is not the case for Agaricales TH9235. Although the fungus superficially resembles species in the genus Tricholoma (Agaricales, Basidiomycota), a genus of terrestrial white-spored mushrooms that form ectomycorrhizas, blastN searches based on ITS rDNA indicated that this species is not closely related to Tricholoma and shares no obvious homology in the ITS1 or ITS2 regions with any Tricholoma taxa in GenBank. BlastN matches of the 5.8s region were inconsistent and provide only low matches to named sequences of Agaricales, suggesting that this mushroom may represent a unique lineage within the order (e.g. a previously unknown genus or family). The equivocal phylogenetic relationships of Agaricales TH9235 were corroborated by inconclusive blastN results from three other gene regions (mtLSU, 28S rDNA and 18S rDNA) that are more conserved than the ITS1 and ITS2 spacer regions (Table 3). For example, blastN results based on 28s rDNA suggested affinities with pink-spored agaricoid fungi in the genera Entoloma and Claudopus, 18S blastN results suggest possible relationships with white-spored taxa in the genera Clitocybe or Hydropus, and mtLSU blastN results suggest possible relationships with either the white-spored genus Amanita or the pink-spored genera Pluteus and Volvariella (Table 3). Some of these genera are known to form ECM associations (e.g. Amanita, some Entoloma) whereas others are considered saprotrophic (e.g. Clitocybe, Pluteus) .
Maximum likelihood phylogeny (A) based on 28S rDNA shows inconclusive placement of Agaricales TH9235 within the mushroom-forming fungal order Agaricales (Basidiomycota). Nodes with bootstrap support ≥70 are indicated by black circles. Taxa considered ectomycorrhizal (ECM) based on Tedersoo et al.  are indicated by bold text, all other species are considered to be either saprotrophic, parasitic, or have an unknown trophic mode. Agaricales TH9235 is nested in a clade that includes pink-spored, saprotrophic and pink-spored ECM Entoloma species as well as the white-spored saprotrophic species Clitocybe hesleri, but this group lacks statistical support. Macroscopic photograph (B) shows fresh orange, tricholomatoid mushrooms of Agaricales TH9235 (Bar = 10 mm). Close-up photograph (C) illustrates a large cluster of ECM Dicymbe roots colonized by the white mycelium of Agaricales TH9235 (Bar = 10 mm).
Efforts to shed light on this enigmatic fungus through phylogenetic analysis based on 28S rDNA were inconclusive. Phylogenies produced by both maximum likelihood (Fig. 4) and maximum parsimony (data not shown) suggest that Agaricales TH9235 may be related to species of Entoloma and Clitocybe, although there is no statistical support for this relationship. The mushrooms that clustered close to TH9235 include pink-spored and white-spored taxa and include both putatively ECM fungi (e.g. Entoloma griseolazulinum from subgenus Entoloma) as well as saprotrophs (Clitocybe hesleri, Entoloma tectonicola in subgenus Inocephalus and Entoloma areneosum in subgenus Pouzarella) .
This is the first in-depth molecular study of ECM fungi associated with P. dipterocarpacea spp. dipterocarpacea. Given the moderate level of sampling and the random selection of ECM roots for molecular analysis, we detected a relatively high ECM species richness on this unusual host plant (40 OTUs) and in general for this tropical savanna-forest mosaic (52 OTUs). Moyersoen’s previous studies ,  that examined the ECM fungal community associated with P. dipterocarpacea spp. nitidum in Venezuela sequenced rDNA from ECM roots and sporocarps to provide evidence of ten ECM fungal lineages (/amanita,/boletus,/cantharellus-craterellus,/clavulina,/coltricia,/cortinarius,/hydnum,/inocybe,/russula-lactarius,/sebacina, and/tomentella-thelephora). We discovered two additional well-established ECM lineages previously unknown on the roots of P. dipterocarpacea (/elaphomyces and/hysterangium) as well as the presence of two putatively new lineages (/polyporales1 and/agaricalesTH9235) (see below). Given that our sampling of P. dipterocarpacea fungi remained far below saturation (Fig. 4), the cumulative data suggest that both subspecies of P. dipterocarpacea probably associate with a wide diversity of ECM fungi over their range in Guyana and Venezuela . At the lineage level, the diversity of ECM fungi associated with P. dipterocarpacea is similar to what has been documented with dipterocarps in Southeast Asia and Africa , , .
Despite the high diversity of ECM fungi documented on the roots of P. dipterocarpacea, we did not find evidence of strong host effects or evidence that the ECM fungal community in this savanna was dramatically different from nearby Dicymbe-dominated rainforest ECM fungal communities. Pakaraimaea dipterocarpacea and D. jenmanii represent two distantly related plant lineages within the angiosperms that have separately evolved the ability to form the ECM symbiosis . Although both plants belong to the rosid radiation within the eudicots, Pakaraimaea belongs to subclass Malvidae whereas Dicymbe belongs to subclass Fabidae . Despite the phylogenetic distance of the host plants, we found that most of the common ECM species were multi-host generalists that were detected on the roots of both hosts (Fig. 2). Also, many of these taxa are present at nearby rainforest sites where P. dipterocarpacea is absent , , a fact reinforced by the overlap of many taxa as sporocarps .
Despite this low level of fungal specialization, it is notable that four common ECM fungi preferred either P. dipterocarpacea or D. jenmanii (Fig. 2). Nonetheless, this level of host preference is low when compared with levels found in some other tropical ecosystems and many temperate ecosystems with multiple sympatric hosts, where species of ECM fungi may exhibit strong preference for one host plant lineage over another , , . The level of host preference in this study, however, was actually more pronounced than in the larger study in nearby rainforest. In that study we examined three leguminous host species, sampled almost four times as many ECM roots, and recovered 118 ECM fungal species but found evidence of host preference in only one fungal species . Two other recent studies of ECM trees in lowland tropical forests in Africa have also reported low levels of ECM host preferences and a high degree of mycobiont sharing among locally sympatric plants in the Fabaceae, Dipterocarpaceae, and Phyllanthaceae , . A similar phenomenon of extensive host-sharing has also been found in some, but certainly not all, temperate forest ecosystems .
The high degree of host sharing may also partially explain why the ECM fungal community in this tropical savanna ecosystem was compositionally similar to that of closed-canopy rainforests of the region. Due to the highly oligotrophic white sand soils, the higher fire frequency, and the presence of a non-leguminous host lineage (e.g. Dipterocarpaceae), we had expected to find a distinct ECM community. However, 67% of the 52 OTUs that we documented on roots at the Pegaima site as well as 56 out of 82 (68.3%) of the ECM sporocarp species had been previously documented at one or more Fabaceae-dominated rainforest sites in the nearby Upper Potaro Basin, suggesting a fairly homogenous pool of regional ECM fungi. This pattern contrasts with many temperate zone forests where ECM fungal communities often exhibit marked spatial autocorrelation and the dominant ECM species can vary significantly in nearby stands of trees , . Furthermore, the importance of edaphic factors in shaping tropical ectomycorrhizal communities has been suggested by studies of paleotropical dipterocarps in Borneo , .
While many ECM fungi are shared by multiple host plant lineages across sites in Guyana, it was notable that in the present study almost all of the dominant ECM fungal lineages either form dense clusters of ECM roots (e.g./russula-lactarius) or produce ectomycorrhizas equipped with extensive extramatrical hyphal cords for medium- to long-distance soil exploration (e.g./boletus and/cortinarius) . In the present study the ECM fungal lineages that were notably less diverse and dominant on ECM roots as compared to the Smith et al.  rainforest study were those which exhibited short-distance exploration strategies with minimal hyphal cord development (e.g./clavulina,/sebacina, and/cantharellus). One explanation for the abundance of ECM fungi with long-distance ECM exploration types and/or large ECM clusters may be that they are adapted to the physico-chemical aspects of the nutrient poor white sand savanna soils or to respond favorably to fire disturbance. Alternatively, it may be that our rapid ECM sampling procedures, which included less root sorting and less intense sampling, were more likely to detect these more robust and noticeable ECM types.
Although most of the ECM types that we documented belonged to fungal lineages known from other parts of the world, one unusual ECM type had the same ITS rDNA sequence as large, orange, tricholomatoid mushrooms found fruiting directly on soil under D. jenmanii and P. dipterocarpacea. This ECM fungus is a member of the Tricholomatoid clade of the Agaricales  but is highly divergent compared to any other known species (Table 3, Fig. 4). Sequences obtained from Agaricales TH9235 suggest that the fungus cannot be convincingly placed within any of the four independently derived ECM lineages within the Tricholomatoid clade (i.e./entoloma,/paralyophyllum,/catathelasma, and/tricholoma) . Agaricales TH9235 may therefore represent a unique evolutionary branch within the order that independently evolved the ability to form ECM in the Neotropics. If this is true, it would be the first documented case of a tropical-endemic ECM fungal group. More robust phylogenetic analyses are necessary to address this hypothesis in the future.
Whether based on the number of species detected per tree or the results of diversity indices, the ECM fungal species diversity indicated by root-based sampling in the Pegaima ecosystem appears lower than that of a nearby rainforest dominated by leguminous trees . This result may be due in part to the comparatively lower sampling intensity, minimal morphological sorting of mycorrhizas, and lower success rate in PCR amplification in the present study. Smith et al.  achieved ca. 90 percent sequencing success when roots were stored directly in CTAB extraction buffer compared to ca. 80 percent success rate in this study where roots were rapidly air-dried with silica gel. Methodological issues aside, the comparison of sampling effort curves from the two different studies suggests that a more complete sampling of this savanna study site would yield similar diversity to nearby rainforest sites (Fig. 3). Additionally, the 86 sporocarp species of ECM fungi collected from the Pegaima site during two short expeditions totals nearly half of the 174 species recovered from the Potaro rainforest site over a 10 year sampling period, suggesting that many more sporocarp species remain to be discovered at Pegaima .
A relatively small number of tropical ECM communities have thus far been studied using molecular techniques but the available data suggest that tropical ecosystems are highly variable in terms of both ECM fungal diversity and the level of ECM host preferences. Ecosystems inhabited by ECM hosts that are large, dominant trees growing close together appear to have relatively high ECM fungal diversity but often low levels of fungal host preferences , , , . In contrast, tropical forests with phylogenetically diverse ECM host plants occurring at low densities in otherwise arbuscular mycorrhizal dominated plant communities have low ECM fungal diversity but often have mycobionts that exhibit distinct host preferences . Thus it appears that several factors, including size of the host plants, host distribution or dominance within the community, and host phylogenetic relationships, may all be important in governing ECM fungal diversity and host associations in tropical habitats. Certainly more studies of tropical ECM plants and their associated ECM fungal communities are warranted to further explore these intriguing patterns.
Permits were granted by the Guyana Environmental Protection Agency. Field assistance was provided by C. Andrew, F. Edmond, L. Edmond, A.C. Grupe, S. Gowan, P. Joseph, and D. Thompson. Dillon Husbands functioned as Guyanese research counterpart and contributed to all phases of the fieldwork. Dr. Francis Brearly and an anonymous reviewer improved this work with constructive editorial critiques. This is paper number 197 of the Smithsonian Institution’s Biological Diversity of the Guiana Shield publication series.
Conceived and designed the experiments: MES TWH JKU HDC. Performed the experiments: MES TWH JKU HDC. Analyzed the data: MES AKF. Contributed reagents/materials/analysis tools: MES RV TWH AKF. Wrote the paper: MES TWH RV.
- 1. Smith S, Read D (2008) Mycorrhizal Symbiosis, Third Edition. Boston: Academic Press.
- 2. Bandou E, Lebailly F, Muller F, Dulormne M, Toribio A, et al. (2006) The ectomycorrhizal fungus Scleroderma bermudense alleviates salt stress in seagrape (Coccoloba uvifera L.) seedlings. Mycorrhiza 16: 559–565.
- 3. Alexander I (2006) Ectomycorrhizas–out of Africa. New Phytol 172: 589–591.
- 4. Henkel T (2003) Monodominance in the ectomycorrhizal Dicymbe corymbosa (Caesalpiniaceae) from Guyana. J. Trop. Ecol. 19: 417–437.
- 5. Fassi B, Fontana A (1962) Micorize ectotrofiche di Brachystegia laurentii e di alcune altre Cesalpiniaceae minori del Congo. Allionia 8: 121–131.
- 6. Tedersoo L, Nara K (2009) General latitudinal gradient of biodiversity is reversed in ectomycorrhizal fungi. New Phytol. 185: 351–354.
- 7. Morris MH, Perez-Perez MA, Smith ME, Bledsoe CS (2009) Influence of host species on ectomycorrhizal communities associated with two co-occurring oaks (Quercus spp.) in a tropical cloud forest. FEMS Microbiol. Ecol. 69: 274–287.
- 8. Peay KG, Kennedy PG, Davies SJ, Tan S, Bruns TD (2010) Potential link between plant and fungal distributions in a dipterocarp rainforest: community and phylogenetic structure of tropical ectomycorrhizal fungi across a plant and soil ecotone. New Phytol. 185: 529–542.
- 9. Smith ME, Henkel TW, Aime MC, Fremier AK, Vilgalys R (2011) Ectomycorrhizal fungal diversity and community structure on three co-occurring leguminous canopy tree species in a Neotropical rainforest. New Phytol. 192: 699–712.
- 10. Tedersoo L, Sadam A, Zambrano M, Valencia R, Bahram M (2009) Low diversity and high host preference of ectomycorrhizal fungi in Western Amazonia, a neotropical biodiversity hotspot. ISME J. 4: 465–71.
- 11. Moyersoen B (2006) Pakaraimaea dipterocarpacea is ectomycorrhizal, indicating an ancient Gondwanaland origin for the ectomycorrhizal habit in Dipterocarpaceae. New Phytol. 172: 753–62.
- 12. Henkel TW, Terborgh J, Vilgalys RJ (2002) Ectomycorrhizal fungi and their leguminous hosts in the Pakaraima Mountains of Guyana. Mycol. Res. 106: 515–531.
- 13. Singer R, Araujo IDJDS (1979) Litter decomposition and ectomycorrhiza in Amazonian Forests 1. A comparison of litter decomposing and ecto mycorrhizal basidiomycetes in latosol terra-firma rain forest and white podzol Campinarana. Acta Amazonica 9: 25–42.
- 14. Tedersoo L, Suvi T, Beaver K, Koljalg U (2007) Ectomycorrhizal fungi of the Seychelles: diversity patterns and host shifts from the native Vateriopsis seychellarum (Dipterocarpaceae) and Intsia bijuga (Caesalpiniaceae) to the introduced Eucalyptus robusta (Myrtaceae), but not Pinus caribea (Pinaceae). New Phytol. 175: 321–33.
- 15. Ashton PS (1988) Dipterocarp biology as a window to the understanding of tropical forest structure. Annu. Rev. Ecol. Syst. 19: 347–370.
- 16. Maury-Lechon G, Curtet L (1998) Biogeography and Evolutionary Systematics of Dipterocarpaceae. In: A Review of Dipterocarps: Taxonomy, ecology and silviculture. Appanah S, Turnbull JM, editors. Center for International Forestry Research, Bogor, Indonesia.
- 17. Brearley FQ (2012) Ectomycorrhizal associations of the Dipterocarpaceae. Biotropica 44: 637–648.
- 18. Ashton PS, Macguire B (1977) Pakaraimaea dipterocarpacea I. systematic, geographic and phyletic considerations. Taxon 26: 343–368.
- 19. Londoño AC, Alvarez E, Forero E, Morton C (1995) A New genus and species of Dipterocarpaceae from the Neotropics. I. Introduction, taxonomy, ecology, and distribution. Brittonia 47: 225–236.
- 20. López-Quintero C, Straatsma G, Franco-Molano AE, Boekhout T (2012) Macrofungal diversity in Colombian Amazon forests varies with regions and regimes of disturbance. Biodivers. Conserv. 21: 2221–2243.
- 21. Maguire B, Ashton PS (1980) Pakaraimaea dipterocarpacea I. systematic, geographic and phyletic considerations. Taxon 29: 225–231.
- 22. Morton C, Dayanandan S, Dissanayake D (1999) Phylogeny and biosystematics of Pseudomonotes (Dipterocarpaceae) based on molecular and morphological data. Plant Syst. Evol. 216: 197–205.
- 23. Dayanandan S, Ashton PS, Williams SM, Primack RB (1999) Phylogeny of the tropical tree family Dipterocarpaceae based on nucleotide sequences of the chloroplast rbcL gene. Am. J. Bot. 86: 1182–1190.
- 24. Moyersoen B (2012) Dispersion, an important radiation mechanism for ectomycorrhizal fungi in Neotropical lowland forests?, Padmini Sudarshana MN-RaJRS, editor: InTech. DOI: 10.5772/33217. Available: www.intechopen.com/books/tropical-forests/dispersion-an-important-radiation-mechanism-for-ectomycorrhizal-fungi-in-neotropical-lowland-forests. Accessed 2013 Jan 8.
- 25. Henkel T, Aime M, Chin M, Miller S, Vilgalys R, et al. (2012) Ectomycorrhizal fungal sporocarp diversity and discovery of new taxa in Dicymbe monodominant forests of the Guiana Shield. Biodivers. Conserv. 26: 2195–2220.
- 26. Fanshawe D (1952) The vegetation of British Guiana: a preliminary review. In: Imperial Forestry Institute. Institute Paper No. 39.
- 27. Hijmans RJ, Cameron SE, Parra JL, Jones PG, Jarvis A (2005) Very high resolution interpolated climate surfaces for global land areas. Int. J. Climat. 25: 1965–1978.
- 28. Gibbs AK, Barron CN (1993) The Geology of the Guiana Shield. New York: Oxford University Press.
- 29. Hammond DS, Ter Steege H (1998) Propensity for fire in Guianan rainforests. Conserv. Biol. 12: 944–947.
- 30. Maguire B, Steryermark JA (1981) Pakaraimaea dipterocarpacea III. Mem. N.Y. Bot. Gard. 32: 306–309.
- 31. Maguire B, Ashton PS (1980) Pakaraimaea dipterocarpacea II. Taxon 29: 225–231.
- 32. Gardes M, Bruns TD (1993) ITS primers with enhanced specificity for basidiomycetes–application to the identification of mycorrhizae and rusts. Mol. Ecol. 2: 113–118.
- 33. Vilgalys R, Hester M (1990) Rapid genetic identification and mapping of enzymatically amplified ribosomal DNA from several Cryptococcus species. J. Bacteriol. 172: 4238–4246.
- 34. White TJ, Bruns TD, Lee S, Taylor J (1990) Amplification and direct sequencing of fungal ribosomal RNA genes for phylogenetics. In: Innis M, Gelfand D, Sninsky J, White T, editors. PCR Protocols: A guide to methods and applications. San Diego: Academic Press. 315–322.
- 35. Glenn T, Schable N (2005) Isolating microsatelline DNA loci. In: Zimmer EA, Roalson EH, eds Molecular evolution: Producing the biochemical data, part B San Diego, CA: Academic Press. 202–222.
- 36. Tedersoo L, May T, Smith ME (2010) Ectomycorrhizal lifestyle in fungi: global diversity, distribution, and evolution of phylogenetic lineages. Mycorrhiza. 20: 217–63.
- 37. McCune B, Mefford M (2011) PC-ORD. Mulitvariate Analysis of Ecological Data. Version 6. Gleneden Beach, Oregon, USA: MjM Software.
- 38. Borneman J, Hartin R (2000) PCR primers that amplify fungal rRNA genes from environmental samples. Appl. Environ. Microbiol. 66: 4356–4360.
- 39. Bruns T, Szaro T, Gardes M, Cullings K, Pan J, et al. (1998) A sequence database for the identification of ectomycorrhizal basidiomycetes by phylogenetic analysis. Mol. Ecol. 7: 257–272.
- 40. Moncalvo JM, Vilgalys R, Redhead SA, Johnson JE, James TY, et al. (2002) One hundred and seventeen clades of euagarics. Mol. Phylogenet. Evol. 23: 357–400.
- 41. Zwickl D (2006) Genetic algorithm approaches for the phylogenetic analysis of large biological sequence datasets under the maximum likelihood criterion. Doctoral Dissertation, University of Texas at Austin. 1–115.
- 42. Co-David D, Langeveld D, Noordeloos ME (2009) Molecular phylogeny and spore evolution of Entolomataceae. Persoonia 23: 147–176.
- 43. Tedersoo L, Bahram M, Jairus T, Bechem E, Chinoya S, et al. (2011) Spatial structure and the effects of host and soil environments on communities of ectomycorrhizal fungi in wooded savannas and rain forests of Continental Africa and Madagascar. Mol. Ecol. 20: 3071–3080.
- 44. Wang B, Qiu Y (2006) Phylogenetic distribution and evolution of mycorrhizas in land plants. Mycorrhiza 16: 299–363.
- 45. Wang HC, Moore MJ, Soltis PS, Bell CD, Brockington SF, et al. (2009) Rosid radiation and the rapid rise of angiosperm-dominated forests. Proc. Natl. Acad. Sci. USA. 106: 3853–3858.
- 46. Ishida T, Nara K, Hogetsu T (2007) Host effects on ectomycorrhizal fungal communities: insight from eight host species in mixed conifer-broadleaf forests. New Phytol. 174: 430–440.
- 47. Smith ME, Douhan GW, Fremier AK, Rizzo DM (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. 182: 295–299.
- 48. Diédhiou AG, Selosse M-A, Galiana A, Diabaté M, Dreyfus B, et al. (2010) Multi-host ectomycorrhizal fungi are predominant in a Guinean tropical rainforest and shared between canopy trees and seedlings. Environ. Microbiol. 12: 2219–2232.
- 49. Kennedy PG, Izzo AD, Bruns TD (2003) There is high potential for the formation of common mycorrhizal networks between understorey and canopy trees in a mixed evergreen forest. J. Ecol. 91: 1071–1080.
- 50. Lilleskov EA, Bruns TD, Horton TR, Taylor D, Grogan P (2004) Detection of forest stand-level spatial structure in ectomycorrhizal fungal communities. FEMS Microbiol. Ecol. 49: 319–332.
- 51. Brearley FQ (2006) Differences in the growth and ectomycorrhizal community of Dryobalanops lanceolata (Dipterocarpaceae) seedlings grown in ultramafic and non-ultramafic soils. Soil Biol. Biochem. 38: 3407–3410.
- 52. Agerer R (2001) Exploration types of ectomycorrhizae. Mycorrhiza. 11: 107–114.
- 53. Matheny P, Curtis J, Hofstetter V, Aime MC, Moncalvo JM, et al. (2006) Major clades of Agaricales: a multilocus phylogenetic overview. Mycologia 98: 982–995.